CN109241198A - Rival's method for digging, device, electronic equipment and storage medium - Google Patents

Rival's method for digging, device, electronic equipment and storage medium Download PDF

Info

Publication number
CN109241198A
CN109241198A CN201810797322.1A CN201810797322A CN109241198A CN 109241198 A CN109241198 A CN 109241198A CN 201810797322 A CN201810797322 A CN 201810797322A CN 109241198 A CN109241198 A CN 109241198A
Authority
CN
China
Prior art keywords
trade company
competition
candidate
target
competitive relation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810797322.1A
Other languages
Chinese (zh)
Other versions
CN109241198B (en
Inventor
黄剑飞
罗恒亮
梁世朴
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Sankuai Online Technology Co Ltd
Original Assignee
Beijing Sankuai Online Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing Sankuai Online Technology Co Ltd filed Critical Beijing Sankuai Online Technology Co Ltd
Priority to CN201810797322.1A priority Critical patent/CN109241198B/en
Publication of CN109241198A publication Critical patent/CN109241198A/en
Application granted granted Critical
Publication of CN109241198B publication Critical patent/CN109241198B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
    • G06F2216/03—Data mining

Landscapes

  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

This application discloses a kind of rival's method for digging, belong to field of computer technology, solve the problems, such as the rival excavated in the prior art inaccuracy.Rival's method for digging disclosed in the present application includes: to determine candidate competition trade company and corresponding first competition relative strength index of the target trade company based on the competitive relation Web Mining according to the competitive relation network constructed in advance；According to the first competition relative strength index, candidate competition trade company to be compared is determined from the candidate competition trade company based on the competitive relation Web Mining；By the way that the candidate competition trade company to be compared to be compared with the competition seed trade company of the target trade company, the competition trade company of the target trade company is determined.Rival's method for digging disclosed in the present application improves the accuracy rate excavated and arrive rival.

Description

Rival's method for digging, device, electronic equipment and storage medium
Technical field
This application involves field of computer technology, more particularly to a kind of rival's method for digging, device, electronic equipment And storage medium.
Background technique
In time discovery rival, facilitate trade company in time, comprehensively adjust marketing strategy, better services client.It is existing Have in technology, rival's method for digging of trade company usually first collect trade company geographical location, trade company positioning (such as star, Type, price etc.), the data such as category information that product is hung under trade company, then, through sorting technique (such as utilization based on category The classification model of automatic classification technology) classified automatically to the data of collection and automatic re-arrangement, finally, for collecting Data in content of text feature calculation content of text between similarity, and then determine the sequence of association trade company, and rely on Expertise come to sequence effect assess, to determine rival.It has been found that being dug using method in the prior art Obtained rival is dug, rival's Result is assessed based on expertise, appraisal procedure is not objective enough, exists The defect of Result inaccuracy；Meanwhile without objective accuracy balancing method, it is unfavorable for continuing to be promoted excavating accuracy.
Summary of the invention
The application provides a kind of rival's method for digging, at least solves the competition pair existing in the prior art excavated The problem of hand inaccuracy.
In a first aspect, the embodiment of the present application provides a kind of rival's method for digging includes:
According to the competitive relation network constructed in advance, candidate of the target trade company based on the competitive relation Web Mining is determined Trade company is competed, and, first competition of the candidate competition trade company based on corresponding competitive relation network and the target trade company is strong Weak index；
According to the first competition relative strength index, from the candidate competition trade company based on the competitive relation Web Mining Middle determination candidate competition trade company to be compared；
By the way that the candidate competition trade company to be compared to be compared with the competition seed trade company of the target trade company, determine The competition trade company of the target trade company.
Second aspect, the embodiment of the present application provide a kind of rival's excavating gear, comprising:
Candidate's competition merchant information determining module, for determining target trade company according to the competitive relation network constructed in advance Candidate competition trade company based on the competitive relation Web Mining, and, the candidate competition trade company is based on corresponding competitive relation First competition relative strength index of network and the target trade company；
Candidate competition trade company's determining module to be compared, for what is determined according to the candidate competition merchant information determining module First competition relative strength index determines candidate to be compared from the candidate competition trade company based on the competitive relation Web Mining Compete trade company；
It competes merchant list and excavates module, for by the way that the candidate to be compared is competed trade company and the target trade company Competition seed trade company is compared, and determines the competition trade company of the target trade company.
The third aspect, the embodiment of the present application also disclose a kind of electronic equipment, including memory, processor and are stored in institute The computer program that can be run on memory and on a processor is stated, the processor realizes this when executing the computer program Apply for rival's method for digging described in embodiment.
Fourth aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey Sequence, when which is executed by processor disclosed in the embodiment of the present application the step of rival's method for digging.
Rival's method for digging disclosed in the embodiment of the present application, the competitive relation network constructed in advance by basis, really Set the goal the candidate competition trade company based on the competitive relation Web Mining of trade company, and, the candidate competition trade company is based on First competition relative strength index of corresponding competitive relation network and the target trade company；Relative strength index is competed according to described first, from Based on determining candidate competition trade company to be compared in the candidate competition trade company of the competitive relation Web Mining；By will be described Candidate competition trade company to be compared is compared with the competition seed trade company of the target trade company, determines the competition of the target trade company Trade company solves the problems, such as the rival excavated in the prior art inaccuracy.Rival disclosed in the embodiment of the present application Method for digging is competed trade company by the competitive relation Web Mining candidate constructed based on data, and competed further combined with seed Trade company is ranked up the candidate competition trade company excavated, improves the accuracy rate excavated and arrive rival.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be in embodiment or description of the prior art Required attached drawing is briefly described, it should be apparent that, the accompanying drawings in the following description is only some realities of the application Example is applied, it for those of ordinary skill in the art, without any creative labor, can also be attached according to these Figure obtains other attached drawings.
Fig. 1 is rival's method for digging flow chart of the embodiment of the present application one；
Fig. 2 is the flow chart of rival's method for digging of the embodiment of the present application two；
Fig. 3 is one of rival's excavating gear structural schematic diagram of the embodiment of the present application three；
Fig. 4 is the second structural representation of rival's excavating gear of the embodiment of the present application three.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application carries out clear, complete Site preparation description, it is clear that described embodiment is some embodiments of the present application, instead of all the embodiments.Based on this Shen Please in embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts Example, shall fall in the protection scope of this application.
First competition relative strength index, the second competition relative strength index and third competition described in the embodiment of the present application are strong and weak Index etc. competes relative strength index for the competitive relation power or size between constant trade company.For example, the first competition relative strength index The value the big, indicates that the competitive relation between trade company is stronger or bigger；First competition relative strength index value is smaller, indicates trade company Between competitive relation it is weaker or smaller.
Embodiment one
A kind of rival's method for digging disclosed in the present embodiment, as shown in Figure 1, this method comprises: step 110 is to step 130。
Step 110, according to the competitive relation network constructed in advance, determine that target trade company is dug based on the competitive relation network The candidate competition trade company of pick, and, the candidate competition trade company based on corresponding competitive relation network and the target trade company the One competition relative strength index.
When it is implemented, determining the candidate competition trade company of target trade company according to the competitive relation network constructed in advance first Information, wherein the candidate competition merchant information includes: the candidate competition trade company based on the competitive relation Web Mining, with And first competition relative strength index of the candidate competition trade company based on corresponding competitive relation network and the target trade company.
Competitive relation network is used to record competitive relation and the first competition relative strength index between trade company, the one of the application In a little specific embodiments, competitive relation network can be constructed by the digraph between building trade company.The application when it is implemented, Can static data based on trade company, trade company line on dynamic data, trade company line under dynamic data, trade company user's line on move State data, trade company user's line under any one or more data in dynamic data combination, excavate that there are competitive relations Trade company, and determine that there are the competition relative strength indexs of first between the trade company of competitive relation.Wherein, there are the trade companies of competitive relation can To be determined according to specific business need, for example, can be with are as follows: multiple trade companies that user successively operates in a behavior, geographical position Set similar multiple trade companies, manage the similar multiple trade companies of category etc..It is described first competition relative strength index for measure trade company it Between competition it is strong and weak, can be determined according to the behavioral data of trade company or the user behavior data of trade company.
Then, using trade company as node, directed edge is constructed according to the competitive relation between determining trade company, is established oriented Figure.Wherein, directed edge is directed toward destination node by source node, and user indicates the corresponding trade company of destination node quotient corresponding with source node There are competitive relations at family.Further, by the first competition between the trade company between destination node trade company corresponding with source node Attribute information of the relative strength index as the directed edge.So far, the building of a competitive relation network is completed.
The application when it is implemented, can from multiple dimensions determine there are the trade companies of competitive relation, and, between trade company First competition relative strength index, and the competitive relation determined according to each dimension and the first competition relative strength index construct accordingly respectively Competitive relation network can also only construct a competitive relation network.
In the specific application process, target trade company pair can be searched for respectively in each competitive relation network constructed in advance The node answered, and the corresponding trade company's conduct of destination node pointed by the determining directed edge with the target trade company as source node Candidate competition trade company of the target trade company in current competitive relational network, using the attribute information of the directed edge as corresponding First competition relative strength index of candidate competition trade company and the target trade company.
Step 120, according to the first competition relative strength index, from the candidate based on the competitive relation Web Mining It competes and determines candidate competition trade company to be compared in trade company.
When it is implemented, can determine that one group of candidate of target trade company competes trade company by a network, by different Network can determine the multiple groups candidate of target trade company compete trade company and every group of candidate compete in trade company each candidate competition trade company with First competition relative strength index of the target trade company.The first competition relative strength index refers to true according to a competitive relation network First competition relative strength index of fixed the target trade company and a certain candidate competition trade company.When it is implemented, if building in advance One competitive relation network, to be based on the competitive relation network to the target trade company according to the first competition relative strength index One group of determining candidate competes trade company and is ranked up, and then, selects the first competition relative strength index in candidate competition trade company highest The candidate competition trade company of preset quantity is as candidate competition trade company to be compared.If constructing multiple competitive relation networks in advance, The one group of candidate that the target trade company can be determined based on each competitive relation network according to the first competition relative strength index Competition trade company is ranked up respectively, then, every group of candidate is selected to compete the first competition highest present count of relative strength index in trade company The candidate competition trade company of amount is as candidate competition trade company to be compared corresponding with corresponding competitive relation network.Wherein, described default Quantity according to business it needs to be determined that, such as can for the target trade company competition seed trade company quantity.
Step 130, by carrying out the competition seed trade company of the candidate competition trade company to be compared and the target trade company Compare, determines the competition trade company of the target trade company.
It, can be by calculating the candidate competition trade company pair to be compared if constructing a competitive relation network in advance It is described to determine whether the candidate competition trade company to be compared can be used as the coverage rate of the competition seed trade company of the target trade company The competition trade company of target trade company.
If constructing multiple competitive relation networks in advance, the application in the specific implementation, needs to preassign the mesh The competition seed trade company of trade company is marked, and determines that the third competition with the target trade company of each competition seed trade company is strong and weak Index.Then, by being based on different competitions to the candidate competition trade company to be compared determined according to each competitive relation network First competition relative strength index of relational network is weighted fusion, determines that the second competition of each candidate competition trade company to be compared is strong Weak index.Later, based on the second competition relative strength index, that is, the first competition relative strength index of multiple competitive relation networks is integrated Obtained competition strong or weak relation calculates the difference of the competition seed trade company of the candidate competition trade company to be compared and the target trade company Different index, to determine whether be ranked up excavation to the candidate competition trade company to be compared based on current second competition relative strength index Accurately.For example, calculate the candidate competition trade company to be compared the second competition relative strength index and the competition seed trade company the The root-mean-square error of three competition relative strength indexs, as difference index.Alternatively, the list based on the candidate competition trade company to be compared With the difference index of the list of the competition seed trade company, relative strength index is competed to described to be compared based on current second to determine Whether candidate's competition trade company is ranked up excavation accurate.For example, calculating described in the list of the candidate competition trade company to be compared The accuracy rate or recall rate for competing seed trade company, as difference index.
When the difference index meets default difference index condition, based on current second competition relative strength index to it is described to Compare candidate competition trade company and is ranked up excavation.When the difference index is unsatisfactory for default difference index condition, by adjusting Each competitive relation network corresponding first competes the weight of relative strength index, and iteration is executed to pass through and be closed to according to each competition It is that the candidate competition trade company to be compared that network determines is weighted based on the first competition relative strength index of different competitive relation networks Fusion determines the step of the step of the second of each candidate competition trade company to be compared competes relative strength index and difference index judge, Until the difference index meets default difference index condition.Wherein, the default difference index condition is according to specific business need Ask determining.
Rival's method for digging disclosed in the embodiment of the present application, the competitive relation network constructed in advance by basis, really Set the goal candidate competition trade company of the trade company based on the competitive relation Web Mining, and, the candidate competition trade company is based on phase Answer the first competition relative strength index of competitive relation network and the target trade company；According to the first competition relative strength index, from base Candidate competition trade company to be compared is determined in the candidate competition trade company of the competitive relation Web Mining；By will it is described to Compare candidate competition trade company to be compared with the competition seed trade company of the target trade company, determines the competition quotient of the target trade company Family solves the problems, such as the rival excavated in the prior art inaccuracy.Rival disclosed in the embodiment of the present application digs Pick method is competed trade company by the competitive relation Web Mining candidate constructed based on data, and competes quotient further combined with seed Family is ranked up determining candidate competition trade company, improves the accuracy rate excavated and arrive rival.
Embodiment two
A kind of rival's method for digging disclosed in the present embodiment, as shown in Fig. 2, this method comprises: step 210 is to step 250。
Step 210, competitive relation network is constructed according to the merchant data of acquisition.
In some embodiments of the present application, the competitive relation network is to be constructed according to the merchant data of acquisition.Example Such as, some dimension can be based on according to merchant data and constructs a competitive relation network, it can also be according to the merchant data of acquisition From the building of at least two dimensions and the one-to-one competitive relation network of at least two dimension.
In some embodiments of the present application, the competitive relation network that the basis constructs in advance determines target trade company base In the candidate competition trade company of the competitive relation Web Mining, and, the candidate competition trade company is based on corresponding competitive relation net Before first competition relative strength index of network and the target trade company, further includes: construct competitive relation according to the merchant data of acquisition Network；Wherein, the merchant data includes dynamic data, and the dynamic data includes following at least one: being moved on the line of trade company State data, trade company line under dynamic data, trade company user's line on dynamic data, trade company user's line under dynamic data.At this In some embodiments of application, the merchant data further includes static data, and the static data further comprises: trade company consolidates There is attribute information.Wherein, static data refers to relatively stable data, such as the build-in attribute information of trade company.It is with trade company Hotel's citing, the build-in attribute information include: the geographical location of trade company, star, opening time, fitting date etc..Dynamically Data refer to the data often changed, construct competitive relation network in conjunction with dynamic data, the timeliness of data are effectively ensured, into one Step can promote the accuracy of the competitive relation network of building.When it is implemented, dynamic on the line for the trade company that dynamic data includes Data may include recent sales volume, nearest price etc.；Dynamic data may include: quotient under the line for the trade company that dynamic data includes Preferential activity in family etc.；Dynamic data may include: the browsing data of user, lower forms data, the use of user on user's line of trade company The comment data etc. that family generates；Dynamic data may include the reality of user to shop data etc. under user's line of trade company.Pass through knot It closes merchant data and user data constructs competitive relation network, data more comprehensively, are conducive to the standard for promoting the rival excavated True property.By combining the dynamic behaviour data of the online and offline of user to construct competitive relation network, real user is accomplished Data-driven helps to promote the practicability and accuracy rate that rival excavates.
After getting merchant data, competitive relation network is further constructed according to the merchant data of acquisition.
Competitive relation network is used to record competitive relation and the first competition relative strength index between trade company, the one of the application In a little specific embodiments, competitive relation network can be constructed by the digraph between building trade company.The competitive relation network It include: node and directed edge, wherein the corresponding trade company of each node, the node include source node and destination node, There are the corresponding node of the trade company of competitive relation, the attribute informations of the directed edge to have including described for the directed edge connection First competition relative strength index of the destination node trade company corresponding with source node to side.It is wherein, described that there are the quotient of competitive relation Family is determined by any one or more following method: determining that trade company involved in the behavior of user is that there are competitive relations Trade company；Determine that trade company involved in the identical behavior of user is the trade company there are competitive relation；Determine the user of designated user Trade company operated by equipment is the trade company there are competitive relation；The trade company for determining that attribute information meets default related condition is to deposit In the trade company of competitive relation；The association trade company for determining default trade company's behavior is the trade company there are competitive relation.
For example, the page of some user in trade company A, B, C browses, the finally single act under trade company A generation, then it is assumed that There are competitive relations between trade company A, B and C；For another example user generated lower uniline in trade company A and trade company B respectively in one week For, then it is assumed that there are competitive relations by trade company A and trade company B；In another example user, which was executed by equipment 1 to shop, consumes operation, then may be used Operate between the trade company being related to that there are competitive relations on line to think the equipment 1；Further for example, if trade company C and trade company D Geographical location is closer or lower extension product category is identical, then it is assumed that there are competitive relations by trade company C and trade company D.The one of the application In a little embodiments, trade company A has browsed trade company E, it may be considered that there are competitive relations between trade company A and trade company E.
In some embodiments of the present application, described there are the trade companies of competitive relation also to determine by the following method: will be with There are the trade companies of competitive relation as there are the trade companies of competitive relation for the same trade company.For example, if trade company A and trade company B exists There are competitive relations by competitive relation, trade company B and trade company F, it may be considered that there is also competitive relations by trade company A and trade company F.
When it is implemented, screening strategy can be arranged according to specific business need, to determine that there are the trade companies of competitive relation. Further, competitive relation can be determined from different dimensions.In some embodiments of the present application, at least two dimensions choosing From in user's evaluation dimension, trade company's inflow and outflow dimension, be overlapped to shop dimension, commercial circle dimension, price dimension, in star dimension The combination of independent dimension or dimension.When constructing competitive relation network based on different dimensions, using corresponding data analysing method pair The merchant data is analyzed, to determine trade company's competitive relation that building competitive relation network needs, and with current building The corresponding trade company of competitive relation network between first competition relative strength index.The first competition relative strength index is for measuring quotient Competition between family is strong and weak, can be obtained by the competition network to building using general ranking algorithm such as pagerank method It arrives.It can also be by carrying out analysis determination in dimension corresponding with the competitive relation network to merchant data, for example, according to quotient The attribute information at family, trade company are predetermined the associated number of behavior, trade company by any one in the number of the default behavior choice Item or multinomial determination.
The specific building process of several competitive relation networks is introduced in citing below.
User's evaluation number when based on user's evaluation dimension building competitive relation network, first in extraction merchant data According to then, according to the determination of the similarity of user's evaluation data, there are the trade companies of competitive relation.For example, some user or some refer to Determine user group, all includes in message at different trade company A and trade company's B page " breakfast is eaten very well ", it is determined that trade company A and trade company There are competitive relations by B, and using trade company A as source node NodeA, trade company B constructs directed edge AB as destination node NodeB.Into one Step, full dose digraph can be constructed by the comment data of the whole network user, it is then, oriented to full dose by ranking method Figure scans for, that is, can determine the competition relative strength index of trade company A and trade company B.It can also be existed according to the similar comment of same user The number occurred in trade company A and trade company B determines the first competition relative strength index of trade company A and trade company B based on user's evaluation dimension.Example Such as, the probability of occurrence by the similar comment of the same user in the comment of trade company A, to the similar comment of the same user Probability of occurrence in the comment of trade company B does difference, then, will ask reciprocal after difference plus 1, using obtained inverse as trade company A With the first competition relative strength index based on user's evaluation dimension of trade company B.Finally, being tieed up based on user's evaluation by family A and trade company B Attribute information of the first competition relative strength index of degree as directed edge AB, according to source node NodeA, destination node NodeB and has Competitive relation network is constructed based on user's evaluation dimension to side AB building.When it is implemented, being constructed based on user's evaluation dimension competing Striving can also include that there are other trade companies pair of competitive relation with trade company A or trade company B based on user's evaluation dimension in relational network The node answered and other there are the nodes of competitive relation.
When based on trade company's inflow and outflow dimension building competitive relation network, the clear of the user in merchant data is extracted first Then the data such as look at, place an order, moving in determine that there are the trade companies of competitive relation according to the data of extraction.In some realities of the application Apply in example, the browsing of user selects and moves in present on line and imperfect, need to integrate under line and line on data carry out jointly Competitive relation network struction.For example, moving in certain hotel under some user's line, platform is obtained by accessing the system of moving in the hotel The device-dependent message of the user, and the hotel's selection course for restoring the user is recalled according to the device-dependent message, it will select It selects hotel involved in process and is determined as trade company there are competitive relation.And in the competition constructed based on trade company's inflow and outflow dimension Increase in relational network with determination there are the corresponding node of the trade company of competitive relation and directed edges.Further, can pass through Then the browsing data building full dose digraph of the whole network user scans for full dose digraph by ranking method, i.e., It can determine the competition relative strength index of trade company A and trade company B.Same use can also be appeared according to the trade company there are competitive relation Probability curve during the placing an order of family determines first there are the trade company of competitive relation based on trade company's inflow and outflow dimension Compete relative strength index.For example, the mean difference for the probability distribution for taking trade company A and trade company B to appear in during the placing an order of multiple users Inverse as trade company A and trade company B based on trade company's inflow and outflow dimension first compete relative strength index.Finally, by described first Compete attribute letter of the relative strength index as the directed edge for connecting trade company A and the corresponding node of trade company B in current competitive relational network Breath.
Geographical location, administrative area when based on commercial circle dimension building competitive relation network, first in extraction merchant data Then the static datas such as Domain Properties determine that there are competing according to information such as the geographical location of trade company distance and administrative region attributes Strive the trade company of relationship.For example, the trade company in same administrative region will be in as there are the trade companies of competitive relation.Further, The first competition relative strength index of trade company can be determined according to the distance between geographical location of trade company, distance is closer, competes stronger. Then, increase node and directed edge in the competitive relation network based on commercial circle dimension according to above-mentioned similar approach, it is competing to construct Strive relational network.
In some embodiments of the present application, it is also based on price dimension, dimension is overlapped to shop, the building such as star dimension Corresponding competitive relation network.Competitive relation network can also be constructed based on price and commercial circle dimension, commercial circle and star dimension etc.. With the promotion of merchant data type enriched constantly with business demand, competitive relation net can be constructed based on more dimensions Network.
Step 220, according to the competitive relation network constructed in advance, determine that target trade company is dug based on the competitive relation network The candidate competition trade company of pick, and, the candidate competition trade company based on corresponding competitive relation network and the target trade company the One competition relative strength index.
In some embodiments of the present application, closed if being based on some dimension previously according to merchant data and constructing a competition It is network, then the competitive relation network that the basis constructs in advance determines the candidate competition merchant information of target trade company, wherein The candidate competition merchant information includes: the candidate competition trade company based on the competitive relation Web Mining, and, the candidate Compete first competition relative strength index of the trade company based on corresponding competitive relation network and the target trade company.
In some embodiments of the present application, at least two competitive relation networks that the basis constructs in advance determine mesh Candidate competition trade company of the trade company based on each competitive relation Web Mining is marked, and, each candidate competition trade company's base In the first competition relative strength index of corresponding competitive relation network and the target trade company, comprising: competing for each of constructing in advance Strive relational network and execute following operation respectively: determine the corresponding source node of linking objective trade company in the competitive relation network has It is target directed edge to side；The corresponding trade company of destination node that the target directed edge is directed toward, as target trade company base In the candidate competition trade company of the competitive relation network；It is oriented as the target by the attribute information of the target directed edge The corresponding candidate competition trade company of destination node pointed by side is based on the competitive relation network and the first of the target trade company Compete relative strength index.
To include the corresponding node citing of trade company A, B, C in competitive relation network N et1, it is assumed that trade company's A corresponding node NodeA is source node, and trade company B and trade company's C corresponding node pass through directed edge AB and AC respectively and connect with source node NodeA, and refer to To trade company B and trade company's C corresponding node, then trade company B and trade company's C corresponding node are destination node, are denoted as NodeB and NodeC.First The corresponding node NodeA of target trade company A is searched in competitive relation network N et1, according to the network information；Further traversal with The directed edge of NodeA connection determines by the directed edge of source node of NodeA to be target directed edge；Target directed edge is being determined After AC and AB, destination node NodeB and NodeC that target directed edge AC and AB are directed toward can be determined, then NodeB and NodeC Corresponding trade company B and trade company C is candidate competition trade company of the target trade company A based on competitive relation network N et1.Then, respectively by mesh Mark the of the attribute information of directed edge AC and AB as trade company B and trade company C and target trade company A based on competitive relation network N et1 One competition relative strength index, i.e. the first competition relative strength index based on competitive relation network N et1.
Other competitive relation networks are traversed after the same method, can determine mesh corresponding with each competitive relation network One group of candidate for marking trade company A competes trade company and every group of candidate competes in trade company, each candidate competition trade company and target trade company A First competition relative strength index.
In some embodiments of the present application, the attribute information by the target directed edge has as the target To the corresponding candidate competition trade company of destination node pointed by side based on the competitive relation network and the target trade company the After one competition relative strength index, further includes: execute following operation respectively for each competitive relation network constructed in advance: will Candidate competition trade company of the target trade company based on the competitive relation network, as hypothesis target trade company；Determine the hypothesis Candidate competition trade company of the target trade company based on the competitive relation network, is based on the competitive relation net as the target trade company The predicting candidate of network competes trade company；Trade company is competed based on the candidate of the competitive relation network according to the hypothesis target trade company The first competition of first competition relative strength index, the hypothesis target trade company based on the competitive relation network and the target trade company Relative strength index determines first competition of the predicting candidate competition trade company based on the competitive relation network and the target trade company Relative strength index；The predicting candidate is competed into trade company, is supplemented to candidate of the target trade company based on the competitive relation network Compete trade company.
After the completion of competitive relation network struction, further potential competition relationship can also be carried out by competitive relation network It excavates, to avoid potential competition trade company is omitted.Such as a certain competitive relation nodes A and the corresponding trade company of node B deposit In competitive relation, there are competitive relations for node C and the corresponding trade company of node B, it may be considered that node A and the corresponding quotient of node C There are competitive relations at family.When it is implemented, including with the candidate competition trade company excavated by competitive relation network N et1 to trade company A Trade company B citing, excavates the candidate competition trade company of trade company B by competitive relation network N et1 first, such as obtains trade company B competition and close The candidate competition trade company for being network N et1 includes trade company H, then is based on the competitive relation for trade company H as the target trade company A The predicting candidate of network N et1 competes trade company.Then, it is determined that trade company H is based on the competitive relation network N et1 and candidate competition quotient The single network of family B competes relative strength index, and is based on the competitive relation network N et1 and candidate competition trade company B according to trade company H Single network competition relative strength index, target trade company A based on the competitive relation network N et1 and candidate compete the single of trade company B Network competition relative strength index determines that single network competition of the trade company H based on the competitive relation network N et1 and target trade company A is strong Weak index.Finally, the predicting candidate is competed trade company H, the target trade company A is supplemented to based on the competitive relation network In candidate's competition trade company, then the target trade company A based on the candidate competition trade company of the competitive relation network be extended for trade company B and Trade company H.
Step 230, according to the first competition relative strength index, from the candidate based on the competitive relation Web Mining It competes and determines candidate competition trade company to be compared in trade company.
When it is implemented, can determine that one group of candidate competes trade company by each competitive relation network.If building in advance Multiple competitive relation networks can determine that the multiple groups candidate of target trade company competes trade company by different competitive relation network, And every group of candidate competes the first competition relative strength index of each candidate competition trade company and the target trade company in trade company.Described first Competition relative strength index refers to that the target trade company determined according to a competitive relation network competes the competing of trade company with a certain candidate Strive relative strength index.When it is implemented, each competition can be based on to the target trade company according to the first competition relative strength index One group of candidate that relational network determines competes trade company and is ranked up respectively, then, selects every group of candidate to compete in trade company first competing The candidate competition trade company of the highest preset quantity of relative strength index is striven as candidate to be compared corresponding with corresponding competitive relation network Compete trade company.Wherein, the preset quantity according to business it needs to be determined that, such as can be the target trade company competition seed quotient The quantity at family.According to the first competition relative strength index, from the candidate competition quotient of each competitive relation Network records Selected section candidate competes trade company and is compared with competition seed trade company in family, can promote the accuracy rate that competition trade company excavates.
It is assumed that the competitive relation network of building includes competitive relation network N et1 based on the building of commercial circle dimension and based on valence The competitive relation network N et2 of lattice dimension building, according to the candidate competition quotient of the competitive relation network N et1 target trade company A determined Family includes trade company B (Net1_Socre_B), C (Net1_Socre_C), D (Net1_Socre_D) and E (Net1_Socre_E), root It include trade company B (Net2_Socre_B), C according to the candidate competition trade company of the competitive relation network N et2 target trade company A determined (Net2_Socre_C), D (Net2_Socre_D) and F (Net2_Socre_F).Wherein, Net1_Socre_B, Net1_Socre_ C, Net1_Socre_D and Net1_Socre_E is respectively that trade company B, C, D, E are recorded in competitive relation network N et1 and target The first competition relative strength index of trade company A, Net2_Socre_B, Net2_Socre_C, Net2Socre_D and Net2_Socre_F points Not Wei trade company B, C, D, F recorded in competitive relation network N et2 compete relative strength index with the first of target trade company A.It assumes again that The seed competition trade company of the target trade company includes trade company B, trade company C and trade company D, then can be from according to competitive relation network N et1 The candidate competition quotient of the first competition of selection relative strength index highest 3 in candidate competition trade company B, C, D, E of determining target trade company A Family is as one group of candidate competition trade company to be compared, for example, the candidate competition trade company to be compared determined is trade company B, C and E；Again from root According to selection the first competition relative strength index in candidate competition trade company B, C, D, F of the competitive relation network N et2 target trade company A determined Highest 3 candidate competition trade companies are as one group of candidate competition trade company to be compared, for example, the candidate competition trade company to be compared determined For trade company B, C and D.
Step 240, by carrying out the competition seed trade company of the candidate competition trade company to be compared and the target trade company Compare, determines the competition trade company of the target trade company.
When constructing a competitive relation network in advance, by the way that the candidate to be compared is competed trade company and the target The competition seed trade company of trade company is compared, and determines the competition trade company of the target trade company, comprising: determines the candidate to be compared Trade company is competed to the coverage rate of the competition seed trade company；If the coverage rate meets default coverage rate threshold value, it is determined that described Competition trade company of the candidate competition trade company to be compared as the target trade company；If the coverage rate is unsatisfactory for the default coverage rate Threshold value then exports the failure of excavation competition trade company.
Still competing trade company below with the candidate to be compared of the target trade company A determined according to competitive relation network N et1 includes quotient Family B (Net1_Socre_B), C (Net1_Socre_C) and E (Net1_Socre_E), the competition seed trade company packet of target trade company A Trade company B, C and F are included, for default coverage rate threshold value is 80%, the candidate competition trade company to be compared is to the competition seed quotient The coverage rate at family is 66.7%, and 66.7% less than 80%, it is believed that the coverage rate is unsatisfactory for the default coverage rate threshold value, then defeated The failure of competition trade company is excavated out.If default coverage rate threshold value is 60%, the candidate competition trade company to be compared is to the competition The coverage rate of seed trade company is 66.7%, and 66.7% is greater than 60%, then it is assumed that the coverage rate meets the default coverage rate threshold All candidate competition trade companies to be compared can be determined as the competition trade companies of the target trade company by value, can also will be described The first competition highest preset quantity of relative strength index candidate competition trade company to be compared is determined as institute in candidate competition trade company to be compared State the competition trade company of target trade company.
Still competing trade company below with the candidate to be compared of the target trade company A determined according to competitive relation network N et1 includes quotient Family B (Net1_Socre_B), C (Net1_Socre_C) and E (Net1_Socre_E) are determined according to competitive relation network N et2 The candidate competition trade company to be compared of target trade company A includes trade company B (Net2_Socre_B), C (Net2_Socre_C) and D (Net2_ Socre_D it) illustrates, the competition seed trade company by the way that the candidate to be compared to be competed to trade company and the target trade company is described in detail It is compared, determines the technical solution of the competition trade company of the target trade company.
It is described by the way that the candidate to be compared is competed trade company and the target trade company in some embodiments of the present application Competition seed trade company be compared, determine the competition trade company of the target trade company, comprising: by competing to the candidate to be compared The the first competition relative strength index for striving trade company is weighted fusion with corresponding weight value, determines the candidate competition trade company to be compared and institute State the second competition relative strength index of target trade company；Determine the difference of the candidate competition trade company to be compared and the competition seed trade company Different index；If the difference index meets the default difference index condition, relative strength index is competed from institute according to described second State the competition trade company that the target trade company is determined in candidate competition trade company to be compared；If the difference index is unsatisfactory for described default Difference index condition then meets the default difference index condition as target using the difference index, adjusts first competition The corresponding weight of relative strength index, jump to the first competition relative strength index by the candidate competition trade company to be compared with Corresponding weight value is weighted fusion, determines that the candidate competition trade company to be compared and the second competition power of the target trade company refer to Target step.Difference index described in the embodiment of the present application is for the difference size between constant trade company.
In the present embodiment, according to the candidate competition to be compared of competitive relation network N et1 and Net2 the target trade company A determined Trade company includes trade company B, C, D and E.Firstly, it is necessary to compete relative strength index to the first of the candidate competition trade company to be compared with phase It answers weight to be weighted fusion, determines that the candidate competition trade company to be compared and the second competition power of the target trade company refer to Mark.By taking trade company B as an example, it is according to the first competition relative strength index of the competitive relation network N et1 trade company B determined and target trade company A Net1_Socre_B, competing relative strength index according to the first of the competitive relation network N et2 trade company B determined and target trade company A is Net2_Socre_B, it is assumed that the initial weight of competitive relation network N et1 and Net2 are 0.5, then trade company B and target trade company A Second competition relative strength index is calculated by the following formula:
Net_Socre (i)=a*Net1_Socre (i)+b*Net2_Socre (i), wherein i=B, Net_Socre (i) For the second competition relative strength index of trade company B and target trade company A, a=b=0.5, Net1_Socre (i)=Net1_Socre_B, Net2_Socre (i)=Net2_Socre_B.
Referring to above-mentioned formula, the second competition relative strength index of candidate competition trade company B, C, D and E to be compared can be determined.
Then, it is determined that the difference index of the candidate competition trade company to be compared and the competition seed trade company.
In some embodiments of the present application, each competition seed trade company corresponding obtaining with the target quotient in advance The third at family competes relative strength index, and the difference of the determination candidate competition trade company to be compared and the competition seed trade company refers to Mark includes at least one of the following: according to the candidate competition trade company to be compared and the competition seed trade company, determine described in than The difference index of more candidate competition trade company and the competition seed trade company；It is competing according to the second of the candidate competition trade company to be compared The third competition relative strength index for striving relative strength index and the competition seed trade company and the target trade company, determines the time to be compared The difference index of choosing competition trade company and the competition seed trade company.
By taking the competition seed trade company of target trade company A includes trade company B, C and F as an example, candidate competition trade company to be compared include: B, C, D and E, wherein including competing seed trade company B and C, then can determine that recalling accuracy rate is 66.7%, i.e. difference index is 66.7%.
In some embodiments of the present application, trade company and the competition seed trade company are competed according to the candidate to be compared, Determine that the difference index of the candidate competition trade company to be compared and the competition seed trade company can be with are as follows: by the competition seed quotient Recall rate or to be compared candidate competition trade company of the family in the candidate competition trade company to be compared are relative to the competition seed quotient The accuracy rate at family, the difference index as the candidate competition trade company to be compared and the competition seed merchant information.
In some embodiments of the present application, according to the second competition relative strength index of the candidate competition trade company to be compared and The third of the competition seed trade company and the target trade company competes relative strength index, determine the candidate competition trade company to be compared with The difference index of the competition seed trade company can be with are as follows: by the second competition relative strength index of candidate competition trade company to be compared and described The root-mean-square error of the third competition relative strength index of seed trade company and the target trade company is competed, it is competing as the candidate to be compared Strive the difference index of trade company and the competition seed merchant information.
Still by taking the competition seed trade company of target trade company A includes trade company B, C and F as an example, it is assumed that competition seed trade company B, C and F Third competition relative strength index with target trade company A is respectively 0.8,0.1 and 0.1, and candidate competition trade company to be compared includes: B, C, D The second competition relative strength index with E is respectively 0.55,0.3,0.1,0.1, then the third competition for competing seed trade company B, C and F is strong The root-mean-square error of weak index and the second competition relative strength index of candidate competition trade company to be compared are as follows:
After the difference index that the competition seed trade company and candidate competition trade company to be compared has been determined, according to default difference Different indicator conditions judge the difference index, to determine according to current second competition relative strength index in the time to be compared Whether the competition trade company that target trade company is excavated in choosing competition trade company is suitable.If the difference index meets the default difference index Condition is less than preset root-mean-square error threshold value greater than preset accuracy rate threshold value or root-mean-square error for example, recalling accuracy rate, Then according to determining candidate competition trade companies all to be compared or the second competition highest part of relative strength index time to be compared Competition trade company of the choosing competition trade company as the target trade company.If the difference index is unsatisfactory for the default difference index item Part, for example, it is square more than or equal to preset less than or equal to preset accuracy rate threshold value or root-mean-square error to recall accuracy rate Root error threshold then meets the default difference index condition as target using the difference index, and first is competing described in iteration adjustment The corresponding weight of relative strength index is striven, and by the first competition relative strength index to the candidate competition trade company to be compared accordingly to weigh Value is weighted fusion, determines the second competition relative strength index of the candidate competition trade company to be compared and the target trade company, directly The competition trade company that target trade company is excavated in trade company is competed in the candidate to be compared with current second competition relative strength index to determining It is suitable.
So far, it has been determined that the competition trade company of target trade company.
When it is implemented, iteration adjustment is substantially the process of data fitting, each competitive relation network, which can correspond to, to be directed to First competition relative strength index of any two trade company, the first competition relative strength index are the input of data fitting, specific data Approximating method can carry out linear fit and optimization using the method for least square, can also using such as logistic regression method its His optimization method to carry out tuning to the ranking results of candidate competition trade company to be compared.
In some embodiments of the present application, the competition seed trade company is determined by any one following method: being passed through The default behavioral data of the target trade company is excavated, determines the competition seed trade company of the target trade company；According to described Target trade company determines the competition seed trade company of the target trade company to the markup information of trade company.By taking trade company of hotel as an example, target wine Shop in hotel end software or can log in hotel's account, then voluntarily select other hotels, and the competition wine by that will select Shop is added to the hotel " my concern " list Zhong Jianggai and is labeled as competition seed trade company；Alternatively, by target hotel in certain time The interior log for browsing other hotels is for statistical analysis, and sorts according to hotel of the browsing frequency to browsing, by high frequency browsing Similar hotel is as competition seed trade company.
When it is implemented, the third competition relative strength index of the competition seed trade company and target trade company is referred to determine time The determination method of first competition relative strength index of choosing competition trade company and target trade company, details are not described herein again.In some of the application In embodiment, the third competition relative strength index of the competition seed trade company and the target trade company can also be according to by the competition Seed is added to the time of " my concern " list, or browsing frequency determines.It is described competing in other embodiments of the application Striving seed trade company and the third competition relative strength index of the target trade company can also be arranged by trade company.
In some embodiments of the present application, the method also includes: according to the target trade company to the competition seed The operation of trade company updates the competition seed trade company.It is described by will be described to be compared in other embodiments of the application Candidate's competition trade company is compared with the competition seed trade company of the target trade company, excavates the competition trade company column of the target trade company After table, further includes: according to the operation of the competition trade company excavated, update the competition seed trade company.In the other of application In embodiment, the competition seed trade company can also be updated using above two method simultaneously.Also, according to the target Operation of the trade company to the competition seed trade company, the operation for updating the competition seed trade company are not limited to determining target trade company After competition trade company.
Step 250, according to the target trade company to the behaviour of the competition seed trade company and/or the competition trade company excavated Make, updates the competition seed trade company.
After the competition trade company that target trade company has been determined, the competition merchant list can be fed back into target trade company.Tool When body is implemented, the trade company in the competition merchant list can be appointed as seed competition trade company by target trade company, for example, passing through boundary Some or certain trade companies in the competition merchant list is added in " my concern " by face interaction.Target trade company can also incite somebody to action Some trade company removes from " my concern ", and then updates the competition seed trade company.
When excavating the competition trade company of target trade company next time, then by determining candidate competition trade company to be compared and after updating Competition seed trade company make comparisons, with the competition trade company of the determination target trade company.
Rival's method for digging disclosed in the embodiment of the present application, by being tieed up according to the merchant data of acquisition from least two Degree building and the one-to-one competitive relation network of at least two dimension, according at least two competitive relations constructed in advance Network determines candidate competition trade company of the target trade company based on each competitive relation Web Mining, and, each candidate First competition relative strength index of the trade company based on corresponding competitive relation network and the target trade company is competed, later, according to described the One competition relative strength index determines time to be compared from the candidate competition trade company based on each competitive relation Web Mining Choosing competition trade company, further by carrying out the competition seed trade company of the candidate competition trade company to be compared and the target trade company Compare, determine the competition trade company of the target trade company, finally, according to the target trade company to the competition seed trade company and/or The operation of the competition trade company excavated updates the competition seed trade company, solves the competition excavated in the prior art The problem of opponent's inaccuracy.Rival's method for digging disclosed in the embodiment of the present application is updated competing by the operation according to trade company Seed trade company is striven, the dynamic order of personalized competition trade company may be implemented, export the competition merchant list of self-learning type, it can Trade company's demand is got a real idea of, accomplishes intelligence, more accurately excavate rival.
Embodiment three
A kind of rival's excavating gear disclosed in the present embodiment, as shown in figure 3, described device includes:
Candidate's competition merchant information determining module 310, for determining target quotient according to the competitive relation network constructed in advance Candidate competition trade company of the family based on the competitive relation Web Mining, and, the candidate competition trade company is based on corresponding competition and closes It is the first competition relative strength index of network and the target trade company；
Candidate competition trade company's determining module 320 to be compared, for according to the candidate competition merchant information determining module 310 Determining first competition relative strength index, determined from the candidate competition trade company based on the competitive relation Web Mining to than More candidate competition trade company；
It competes merchant list and excavates module 330, for passing through the candidate competition trade company to be compared and the target quotient The competition seed trade company at family is compared, and determines the competition trade company of the target trade company.
Optionally, as shown in figure 4, described device further includes competition seed trade company update module 340, the competition seed quotient Family update module 340 is further used for:
Operation according to the target trade company to the competition seed trade company, updates the competition seed trade company；And/or
According to the operation of the competition trade company excavated, the competition seed trade company is updated.
Optionally, the competitive relation network is to be constructed according to the merchant data of acquisition, in some implementations of the application In example, as shown in figure 4, described device further include:
Competitive relation network struction module 350, for constructing competitive relation network according to the merchant data of acquisition；
Wherein, the merchant data includes dynamic data, and the dynamic data includes following at least one: on the line of trade company Dynamic data, trade company line under dynamic data, trade company user's line on dynamic data, trade company user's line under dynamic data.It is logical The dynamic behaviour data building competitive relation network for crossing the online and offline in conjunction with user, has accomplished that real user data drives It is dynamic, help to promote the practicability and accuracy rate that rival excavates.In some embodiments of the present application, the merchant data It further include static data, the static data further comprises: the build-in attribute information of trade company.
Optionally, the competitive relation network includes: node and directed edge, wherein each corresponding quotient of the node Family, the node include source node and destination node, and there are the corresponding sections of the trade company of competitive relation for the directed edge connection Point, the attribute information of the directed edge include that the first competition of the destination node trade company corresponding with source node of the directed edge is strong Weak index, the candidate competition merchant information determining module 310 are further used for:
The directed edge for determining the corresponding source node of linking objective trade company in the competitive relation network is target directed edge；
The corresponding trade company of destination node that the target directed edge is directed toward is based on the competition as the target trade company The candidate competition trade company of relational network；
By the attribute information of the target directed edge, as the corresponding time of destination node pointed by the target directed edge First competition relative strength index of the choosing competition trade company based on the competitive relation network and the target trade company.
Optionally, the candidate competition merchant information determining module 310 is also used to:
Candidate competition trade company by the target trade company based on the competitive relation network, as hypothesis target trade company；
Candidate competition trade company of the hypothesis target trade company based on the competitive relation network is determined, as the target quotient Family competes trade company based on the predicting candidate of the competitive relation network；
First competition of the candidate competition trade company according to the hypothesis target trade company based on the competitive relation network is strong and weak The the first competition relative strength index of index, the hypothesis target trade company based on the competitive relation network and the target trade company, really First competition relative strength index of the fixed predicting candidate competition trade company based on the competitive relation network and the target trade company；
The predicting candidate is competed into trade company, it is competing to be supplemented to candidate of the target trade company based on the competitive relation network Strive trade company.
Optionally, the competition merchant list excavates module 330 and is further used for:
Determine the candidate competition trade company to be compared to the coverage rate of the competition seed trade company；
If the coverage rate meets default coverage rate threshold value, it is determined that the candidate competition trade company to be compared is as the mesh Mark the competition trade company of trade company；
If the coverage rate is unsatisfactory for the default coverage rate threshold value, the failure of excavation competition trade company is exported.
Optionally, the competition merchant list excavates module 330 and is further used for:
Fusion is weighted with corresponding weight value by the first competition relative strength index to the candidate competition trade company to be compared, Determine the second competition relative strength index of the candidate competition trade company to be compared and the target trade company；
Determine the difference index of the candidate competition trade company to be compared and the competition seed trade company；
If the difference index meets the default difference index condition, relative strength index is competed from institute according to described second State the competition trade company that the target trade company is determined in candidate competition trade company to be compared；
If the difference index is unsatisfactory for the default difference index condition, met with the difference index described default Difference index condition is target, adjusts the corresponding weight of the first competition relative strength index, jump to it is described by it is described to The the first competition relative strength index for comparing candidate competition trade company is weighted fusion with corresponding weight value, determines that the candidate to be compared is competing Strive the second competition relative strength index of trade company and the target trade company.
Optionally, each competition seed trade company is corresponding obtaining in advance strong and weak with target trade company third competition Index, the difference index of the determination candidate competition trade company to be compared and the competition seed trade company, including it is following at least One:
According to the candidate competition trade company to be compared and the competition seed trade company, the candidate competition quotient to be compared is determined The difference index at family and the competition seed trade company；
According to the second competition relative strength index of the candidate competition trade company to be compared and the competition seed trade company with it is described The third of target trade company competes relative strength index, determines the difference of the candidate competition trade company to be compared and the competition seed trade company Index.
Optionally, the competition seed trade company is determined by any one following method:
It is excavated by the default behavioral data to the target trade company, determines the competition seed quotient of the target trade company Family；
According to the target trade company to the markup information of trade company, the competition seed trade company of the target trade company is determined.
Rival's excavating gear disclosed in the embodiment of the present application, for realizing in the embodiment of the present application one and embodiment two Each step of rival's method for digging, the specific embodiment of each module of device is referring to corresponding steps, herein not It repeats again.
Rival's excavating gear disclosed in the embodiment of the present application, the competitive relation network constructed in advance by basis, really Set the goal the candidate competition trade company based on the competitive relation Web Mining of trade company, and, the candidate competition trade company is based on First competition relative strength index of corresponding competitive relation network and the target trade company；Relative strength index is competed according to described first, from Based on determining candidate competition trade company to be compared in the candidate competition trade company of the competitive relation Web Mining；By will be described Candidate competition trade company to be compared is compared with the competition seed trade company of the target trade company, determines the competition of the target trade company Trade company solves the problems, such as the rival excavated in the prior art inaccuracy.Rival disclosed in the embodiment of the present application Excavating gear is competed trade company by the competitive relation Web Mining candidate constructed based on data, and competed further combined with seed Trade company is ranked up the candidate competition trade company excavated, improves the accuracy rate excavated and arrive rival.
Further, rival's excavating gear disclosed in the embodiment of the present application is updated competing by the operation according to trade company Seed trade company is striven, the dynamic order of personalized competition trade company may be implemented, export the competition merchant list of self-learning type, it can Trade company's demand is got a real idea of, accomplishes intelligence, more accurately excavate rival.
Correspondingly, disclosed herein as well is a kind of electronic equipment, including memory, processor and it is stored in the memory Computer program that is upper and can running on a processor, the processor are realized when executing the computer program as the application is real Apply rival's method for digging described in example one and embodiment two.The electronic equipment can be PC machine, mobile terminal, a number Word assistant, tablet computer etc..
Disclosed herein as well is a kind of computer readable storage mediums, are stored thereon with computer program, which is located Manage the step of realizing rival's method for digging as described in the embodiment of the present application one and embodiment two when device executes.
All the embodiments in this specification are described in a progressive manner, the highlights of each of the examples are with The difference of other embodiments, the same or similar parts between the embodiments can be referred to each other.For Installation practice For, since it is basically similar to the method embodiment, so being described relatively simple, referring to the portion of embodiment of the method in place of correlation It defends oneself bright.
A kind of rival's method for digging provided by the present application and device are described in detail above, it is used herein The principle and implementation of this application are described for specific case, and the above embodiments are only used to help understand The present processes and its core concept；At the same time, for those skilled in the art is having according to the thought of the application There will be changes in body embodiment and application range, in conclusion the content of the present specification should not be construed as to the application Limitation.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware realization.Based on such reason Solution, substantially the part that contributes to existing technology can embody above-mentioned technical proposal in the form of software products in other words Come, which may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including Some instructions are used so that a computer equipment (can be personal computer, server or the network equipment etc.) executes respectively Method described in certain parts of a embodiment or embodiment.

Claims (20)

1. a kind of rival's method for digging characterized by comprising
According to the competitive relation network constructed in advance, candidate competition of the target trade company based on the competitive relation Web Mining is determined Trade company, and, the candidate competition trade company is referred to based on corresponding competitive relation network and the first competition power of the target trade company Mark；
According to the first competition relative strength index, from the candidate competition trade company based on the competitive relation Web Mining really Fixed candidate competition trade company to be compared；
By the way that the candidate competition trade company to be compared to be compared with the competition seed trade company of the target trade company, described in determination The competition trade company of target trade company.
2. the method according to claim 1, wherein further include:
Operation according to the target trade company to the competition seed trade company, updates the competition seed trade company；And/or
According to the operation of the competition trade company excavated, the competition seed trade company is updated.
3. method according to claim 1 or 2, which is characterized in that the competitive relation network is the trade company according to acquisition Data building, wherein the merchant data includes dynamic data, and the dynamic data includes following at least one: trade company Dynamic data on line, trade company line under dynamic data, trade company user's line on dynamic data, trade company user's line under dynamic number According to.
4. according to the method described in claim 3, it is characterized in that, the competitive relation network includes: node and directed edge, In, each corresponding trade company of the node, the node includes source node and destination node, and there are competing for the directed edge connection The corresponding node of trade company of relationship is striven, the attribute information of the directed edge includes that the destination node of the directed edge and source are saved First competition relative strength index of the corresponding trade company of point, the competitive relation network that the basis constructs in advance determine target trade company base In the candidate competition trade company of the competitive relation Web Mining, and, the candidate competition trade company is based on corresponding competitive relation net The step of first competition relative strength index of network and the target trade company, comprising:
The directed edge for determining the corresponding source node of linking objective trade company in the competitive relation network is target directed edge；
The corresponding trade company of destination node that the target directed edge is directed toward is based on the competitive relation as the target trade company The candidate competition trade company of network；
By the attribute information of the target directed edge, as destination node pointed by the target directed edge, corresponding candidate is competing Strive first competition relative strength index of the trade company based on the competitive relation network and the target trade company.
5. according to the method described in claim 4, it is characterized in that, the attribute information by the target directed edge, as The corresponding candidate competition trade company of destination node pointed by the target directed edge is based on the competitive relation network and the mesh After the step of marking the first competition relative strength index of trade company, further includes:
Candidate competition trade company by the target trade company based on the competitive relation network, as hypothesis target trade company；
Candidate competition trade company of the hypothesis target trade company based on the competitive relation network is determined, as target trade company base Trade company is competed in the predicting candidate of the competitive relation network；
First competition relative strength index of the candidate competition trade company according to the hypothesis target trade company based on the competitive relation network, It is described to assume first competition relative strength index of the target trade company based on the competitive relation network and the target trade company, determine described in Predicting candidate competes first competition relative strength index of the trade company based on the competitive relation network and the target trade company；
The predicting candidate is competed into trade company, is supplemented to candidate competition quotient of the target trade company based on the competitive relation network Family.
6. the method according to claim 1, wherein described by the way that the candidate to be compared is competed trade company and institute State the step of competition seed trade company of target trade company is compared, determines the competition trade company of the target trade company, comprising:
Determine the candidate competition trade company to be compared to the coverage rate of the competition seed trade company；
If the coverage rate meets default coverage rate threshold value, it is determined that the candidate competition trade company to be compared is as the target quotient The competition trade company at family；
If the coverage rate is unsatisfactory for the default coverage rate threshold value, the failure of excavation competition trade company is exported.
7. the method according to claim 1, wherein described by the way that the candidate to be compared is competed trade company and institute State the step of competition seed trade company of target trade company is compared, determines the competition trade company of the target trade company, comprising:
Fusion is weighted with corresponding weight value by the first competition relative strength index to the candidate competition trade company to be compared, is determined Second competition relative strength index of the candidate competition trade company to be compared and the target trade company；
Determine the difference index of the candidate competition trade company to be compared and the competition seed trade company；
If the difference index meets the default difference index condition, according to it is described second competition relative strength index from it is described to Compare the competition trade company that the target trade company is determined in candidate competition trade company；
If the difference index is unsatisfactory for the default difference index condition, the default difference is met with the difference index Indicator conditions are target, adjust the corresponding weight of the first competition relative strength index, jump to described by described to be compared First competition relative strength index of candidate's competition trade company is weighted fusion with corresponding weight value, determines the candidate competition quotient to be compared The step of second competition relative strength index of family and the target trade company.
8. the method according to the description of claim 7 is characterized in that each competition seed trade company it is corresponding it is obtaining in advance with The third of the target trade company competes relative strength index, the determination candidate competition trade company to be compared and the competition seed quotient It the step of difference index at family, includes at least one of the following:
According to the candidate competition trade company to be compared and the competition seed trade company, determine the candidate competition trade company to be compared and The difference index of the competition seed trade company；
According to the second competition relative strength index of the candidate competition trade company to be compared and the competition seed trade company and the target The third of trade company competes relative strength index, determines that the candidate competition trade company to be compared and the difference of the competition seed trade company refer to Mark.
9. the method according to claim 1, wherein the competition seed trade company passes through any one following method It determines:
It is excavated by the default behavioral data to the target trade company, determines the competition seed trade company of the target trade company；
According to the target trade company to the markup information of trade company, the competition seed trade company of the target trade company is determined.
10. a kind of rival's excavating gear characterized by comprising
Candidate's competition merchant information determining module, for determining that target trade company is based on according to the competitive relation network constructed in advance The candidate competition trade company of the competitive relation Web Mining, and, the candidate competition trade company is based on corresponding competitive relation network Relative strength index is competed with the first of the target trade company；
Candidate competition trade company's determining module to be compared, first for being determined according to the candidate competition merchant information determining module Relative strength index is competed, determines that candidate to be compared competes from the candidate competition trade company based on the competitive relation Web Mining Trade company；
It competes merchant list and excavates module, for the competition by the way that the candidate to be compared to be competed to trade company and the target trade company Seed trade company is compared, and determines the competition trade company of the target trade company.
11. device according to claim 10, which is characterized in that further include: competition seed trade company update module, it is described competing Seed trade company update module is striven to be further used for:
Operation according to the target trade company to the competition seed trade company, updates the competition seed trade company；And/or
According to the operation of the competition trade company excavated, the competition seed trade company is updated.
12. device described in 0 or 11 according to claim 1, which is characterized in that the competitive relation network is the quotient according to acquisition User data building；
Wherein, the merchant data includes dynamic data, and the dynamic data includes following at least one: dynamic on the line of trade company Data, trade company line under dynamic data, trade company user's line on dynamic data, trade company user's line under dynamic data.
13. device according to claim 12, which is characterized in that the competitive relation network includes: node and directed edge, Wherein, the corresponding trade company of each node, the node include source node and destination node, and the directed edge connection exists The corresponding node of the trade company of competitive relation, the attribute information of the directed edge include destination node and the source of the directed edge First competition relative strength index of the corresponding trade company of node, the candidate competition merchant information determining module are further used for:
The directed edge for determining the corresponding source node of linking objective trade company in the competitive relation network is target directed edge；
The corresponding trade company of destination node that the target directed edge is directed toward is based on the competitive relation as the target trade company The candidate competition trade company of network；
By the attribute information of the target directed edge, as destination node pointed by the target directed edge, corresponding candidate is competing Strive first competition relative strength index of the trade company based on the competitive relation network and the target trade company.
14. device according to claim 13, which is characterized in that the candidate competition merchant information determining module is also used In:
Candidate competition trade company by the target trade company based on the competitive relation network, as hypothesis target trade company；
Candidate competition trade company of the hypothesis target trade company based on the competitive relation network is determined, as target trade company base Trade company is competed in the predicting candidate of the competitive relation network；
First competition relative strength index of the candidate competition trade company according to the hypothesis target trade company based on the competitive relation network, It is described to assume first competition relative strength index of the target trade company based on the competitive relation network and the target trade company, determine described in Predicting candidate competes first competition relative strength index of the trade company based on the competitive relation network and the target trade company；
The predicting candidate is competed into trade company, is supplemented to candidate competition quotient of the target trade company based on the competitive relation network Family.
15. device according to claim 10, which is characterized in that the competition merchant list excavates module and further uses In:
Determine the candidate competition trade company to be compared to the coverage rate of the competition seed trade company；
If the coverage rate meets default coverage rate threshold value, it is determined that the candidate competition trade company to be compared is as the target quotient The competition trade company at family；
If the coverage rate is unsatisfactory for the default coverage rate threshold value, the failure of excavation competition trade company is exported.
16. device according to claim 10, which is characterized in that the competition merchant list excavates module and further uses In:
Fusion is weighted with corresponding weight value by the first competition relative strength index to the candidate competition trade company to be compared, is determined Second competition relative strength index of the candidate competition trade company to be compared and the target trade company；
Determine the difference index of the candidate competition trade company to be compared and the competition seed trade company；
If the difference index meets the default difference index condition, according to it is described second competition relative strength index from it is described to Compare the competition trade company that the target trade company is determined in candidate competition trade company；
If the difference index is unsatisfactory for the default difference index condition, the default difference is met with the difference index Indicator conditions are target, adjust the corresponding weight of the first competition relative strength index, jump to described by described to be compared First competition relative strength index of candidate's competition trade company is weighted fusion with corresponding weight value, determines the candidate competition quotient to be compared The second competition relative strength index at family and the target trade company.
17. device according to claim 16, which is characterized in that each competition seed trade company correspondence obtained in advance Relative strength index, the determination candidate competition trade company to be compared and the competition seed are competed with the third of the target trade company The difference index of trade company, includes at least one of the following:
According to the candidate competition trade company to be compared and the competition seed trade company, determine the candidate competition trade company to be compared and The difference index of the competition seed trade company；
According to the second competition relative strength index of the candidate competition trade company to be compared and the competition seed trade company and the target The third of trade company competes relative strength index, determines that the candidate competition trade company to be compared and the difference of the competition seed trade company refer to Mark.
18. device according to claim 10, which is characterized in that the competition seed trade company passes through any one following side Method determines:
It is excavated by the default behavioral data to the target trade company, determines the competition seed trade company of the target trade company；
According to the target trade company to the markup information of trade company, the competition seed trade company of the target trade company is determined.
19. a kind of electronic equipment, including memory, processor and it is stored on the memory and can runs on a processor Computer program, which is characterized in that the processor realizes claim 1 to 9 any one when executing the computer program Rival's method for digging.
20. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor The step of rival's method for digging described in claim 1 to 9 any one is realized when execution.
CN201810797322.1A 2018-07-19 2018-07-19 Competitor mining method and device, electronic equipment and storage medium Active CN109241198B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810797322.1A CN109241198B (en) 2018-07-19 2018-07-19 Competitor mining method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810797322.1A CN109241198B (en) 2018-07-19 2018-07-19 Competitor mining method and device, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN109241198A true CN109241198A (en) 2019-01-18
CN109241198B CN109241198B (en) 2021-04-23

Family

ID=65072246

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810797322.1A Active CN109241198B (en) 2018-07-19 2018-07-19 Competitor mining method and device, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN109241198B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050261952A1 (en) * 2004-05-24 2005-11-24 Ford Motor Company Computer-implemented method and system for modeling and estimating vehicle sales
CN103455593A (en) * 2013-09-01 2013-12-18 北京航空航天大学 Service competitiveness realization system and method based on social contact network
CN105139106A (en) * 2015-07-31 2015-12-09 山东大学 Enterprise network architecture based on bill of product materials and constructing method therefor
CN106919582A (en) * 2015-12-24 2017-07-04 阿里巴巴集团控股有限公司 The association of network articles and related information statistical method and device
CN107807818A (en) * 2017-10-17 2018-03-16 云南大学 A kind of multilayered model and its construction method towards SaaS multi-tenants

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050261952A1 (en) * 2004-05-24 2005-11-24 Ford Motor Company Computer-implemented method and system for modeling and estimating vehicle sales
CN103455593A (en) * 2013-09-01 2013-12-18 北京航空航天大学 Service competitiveness realization system and method based on social contact network
CN105139106A (en) * 2015-07-31 2015-12-09 山东大学 Enterprise network architecture based on bill of product materials and constructing method therefor
CN106919582A (en) * 2015-12-24 2017-07-04 阿里巴巴集团控股有限公司 The association of network articles and related information statistical method and device
CN107807818A (en) * 2017-10-17 2018-03-16 云南大学 A kind of multilayered model and its construction method towards SaaS multi-tenants

Also Published As

Publication number Publication date
CN109241198B (en) 2021-04-23

Similar Documents

Publication Publication Date Title
CN105701216B (en) A kind of information-pushing method and device
CN103339623B (en) It is related to the method and apparatus of Internet search
CN109345302A (en) Machine learning model training method, device, storage medium and computer equipment
CN103177090B (en) A kind of topic detection method and device based on big data
CN109684554A (en) The determination method and news push method of the potential user of news
CN104992347B (en) A kind of method and device of video matching advertisement
US11238233B2 (en) Artificial intelligence engine for generating semantic directions for websites for automated entity targeting to mapped identities
KR102340463B1 (en) Sample weight setting method and device, electronic device
CN105045901A (en) Search keyword push method and device
CN102053971A (en) Recommending method and equipment for sequencing-oriented collaborative filtering
CA2797401A1 (en) Automated social networking graph mining and visualization
US11526773B1 (en) Predicting accuracy of submitted data
US10565262B2 (en) Methods for refining search results in an application
CN108460082A (en) A kind of recommendation method and device, electronic equipment
CN103390044B (en) Method and device for identifying linkage type POI (Point Of Interest) data
JP2018077821A (en) Method, program, server device, and processor for generating predictive model of category of venue visited by user
CN107978189A (en) Intelligent exercise pushing method and system and terminal equipment
CN107436914A (en) Recommend method and device
CN108062366A (en) Public culture information recommendation system
CN113157752A (en) Scientific and technological resource recommendation method and system based on user portrait and situation
Deng Towards objective benchmarking of electronic government: an inter‐country analysis
CN113902185A (en) Method and device for determining regional land property, electronic equipment and storage medium
AU2015203108B1 (en) Automated predictive tag management system
Miklosik et al. A framework for constructing optimisation models to increase the visibility of organizations' information in search engines
CN106407316A (en) Topic model-based software question and answer recommendation method and device

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant