CN107092621A - Information search method and device - Google Patents
Information search method and device Download PDFInfo
- Publication number
- CN107092621A CN107092621A CN201611045792.XA CN201611045792A CN107092621A CN 107092621 A CN107092621 A CN 107092621A CN 201611045792 A CN201611045792 A CN 201611045792A CN 107092621 A CN107092621 A CN 107092621A
- Authority
- CN
- China
- Prior art keywords
- search
- search result
- label
- hunting zone
- result
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present invention provides a kind of information search method and device, is related to Internet technical field.Wherein, methods described includes：According to keyword, the first search result is searched out；Based on first search result, search need is determined；Search meets the second search result of the search need.Scanned for based on search need, the result searched is closer to user's real demand, and the search degree of accuracy is high.
Description
Technical field
The present invention relates to Internet technical field, more particularly to a kind of information search method and device.
Background technology
As internet is fast-developing, the spring after a large amount of O2O (on Online To Offline, line to line under) platform such as rain
Appearance as bamboo shoot.O2O is combined the commercial chance under line with internet, allows internet to become the platform of off-line transaction, significantly
Change the life style of people.For example, the appearance for taking out industry O2O platforms changes the food and drink behavior of people.
O2O platforms can provide the user with a client application (APP), and user can search for trade company by APP, subsequently into
Corresponding trade company interface such as carries out commodity selection, placed an order at the operation.Sometimes, when user is scanned for using APP, it may appear that search is not
To the situation for expecting trade company, the expectation of user is not being met.
The content of the invention
Based on keyword carry out text matches using obtain the process of search result as：First, participle is carried out to keyword to obtain
To a collection of term (participle), participle collection tokens={ term1, term2 ..., termN } is designated as.Then, based on tokens,
From the inverted index of name of firm and menu name recall obtaining trade company set shoplist, the condition recalled is
Tokens being capable of some or all of hit name of firm or menu name.Finally, by the trade company set shoplist recalled
Feed back to user.Above-mentioned use keyword participle carries out the way of search of text matches, searches in obtained result and occurs with using
The probability that family real demand differs greatly is higher, it is impossible to comprehensively meet user's request；In addition, using simple text matches
Mode, which is easily appeared in, to be there is search for sky because that can not match part or all of participle in a certain hunting zone or searches for knot
Really few situation, the expectation of user is not being met.For example, user expects to find entitled B trade company in position A, still
Position A and its trade company that entitled B is nearby not present, now search result will be empty, and the expectation of user be not being met.
Then, there is provided a kind of information search method in one embodiment of the invention.This method includes：According to pass
Keyword, searches out the first search result；Based on first search result, search need is determined；Search meets the search need
The second search result asked.
Optionally, in above-mentioned method, first search result can be one or more.When the described first search knot
It is above-mentioned to be based on first search result when fruit is multiple, search need is determined, can adopt and realize with the following method：To multiple institutes
State the first search result and carry out Commonness Analysis, obtain common feature；According to the common feature, the search need is determined.
Optionally, it is above-mentioned that Commonness Analysis is carried out to multiple first search results, common feature is obtained, including：Obtain
Multiple first search results, which are distinguished in corresponding label dictionary, the label dictionary, includes label and the corresponding power of label
Weight；The weight of label in multiple label dictionaries is weighted and collected；The label that summarized results is come to top n is used as institute
State common feature.Wherein, the N is the first default value.
Optionally, above-mentioned method may also include：Multiple first search results are ranked up；Accordingly, to many
Individual first search result carries out Commonness Analysis, obtains common feature, can be specially：M first search before to coming
As a result Commonness Analysis is carried out, the common feature is obtained.Wherein, the M is the second default value.From M first before coming
Search result carries out Commonness Analysis, and one is that can reduce the data volume of analysis, improves analysis efficiency；Another be sort it is more forward
Search result and keyword matching degree it is higher, therefore the general character that Commonness Analysis is obtained is carried out to the high search result of matching degree
Feature will be more accurate.
Optionally, above-mentioned search meets the second search result of the search need, including：Obtain in hunting zone
The corresponding label dictionary of data；The corresponding label dictionary of the data is matched with the search need；If matching degree expires
Foot matching is required, then regard the data as second search result.
Optionally, above-mentioned method also includes：If second search result is multiple, according to the matching degree to many
Individual second search result is ranked up.In a kind of achievable mode, according to height from high in the end suitable of matching degree
Multiple second search results of ordered pair are ranked up.The purpose of sequence is in order to high with the search need matching degree second is searched
Hitch fruit, which comes, above to be showed, and preferable search experience is brought to user.
Optionally, above-mentioned method also includes：Obtain the characteristic information of multiple second search results；And, it is described
Multiple second search results are ranked up according to matching degree, including：It is right with reference to the matching degree and the characteristic information
Multiple second search results are ranked up.The characteristic information of the second search result is added as sort by so that sequence
According to more comprehensive, the search experience of user is favorably improved.
Optionally, the first above-mentioned search result and the second search result can be the numbers in two different hunting zones
According to for example, first search result is the data in the first hunting zone；Second search result is the second hunting zone
Interior data.Accordingly, above-mentioned according to keyword, searching out the first search result can be specially：In the first hunting zone,
First search result is gone out according to the keyword search.Above-mentioned search meets the second search result of the search need
Can be specially：In the second hunting zone, search meets the second search result of the search need.
Optionally, above-mentioned method also includes：In second hunting zone, search out and match with the keyword
The 3rd search result；If the quantity of the 3rd search result searched out is less than threshold value, expand the second search model
Enclose and obtain first hunting zone.Method provided in an embodiment of the present invention can be triggered by above-mentioned event, i.e., search for model second
It is searching in enclosing with Keywords matching the 3rd search result seldom or be space-time, expanding hunting zone and analyzing search needs
Ask, return the second hunting zone and scanned for based on search need, enrich search result, and then a variety of of user can be met
Demand.
Optionally, above-mentioned method may also include：For second search result, rearmounted display properties is set；Wherein, institute
State rearmounted display properties and indicate and the 3rd search result is placed in behind the display location of second search result.Receive
The client of two search results and the 3rd search result can be according to the rearmounted display properties of the second search result, preferentially in client
Search interface on show the 3rd search result, second search result is then presented behind the 3rd search result again.
There is provided a kind of information retrieval device in another embodiment of the present invention.The device includes：First search mould
Block, determining module and the second search module.Wherein, the first search module is used to go out the first search result according to keyword search；
Determining module is used to be based on first search result, determines search need；Second search module is used to search for be searched described in satisfaction
Second search result of rope demand.
Technical scheme provided in an embodiment of the present invention, search need, then base are determined according to the first search result searched
Scanned in search need, the second search result searched easily meets the search need of user close to the real demand of user
Ask.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are this hairs
Some bright embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can be with root
Other accompanying drawings are obtained according to these accompanying drawings.
The schematic flow sheet for the information search method that Fig. 1 provides for one embodiment of the invention；
The schematic flow sheet for the information search method that Fig. 2 provides for another embodiment of the present invention；
The schematic flow sheet for the information search method that Fig. 3 provides for further embodiment of this invention；
The structured flowchart for the information retrieval device that Fig. 4 provides for one embodiment of the invention；
The structured flowchart for the information retrieval device that Fig. 5 provides for another embodiment of the present invention.
Embodiment
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described.
In some flows of description in description and claims of this specification and above-mentioned accompanying drawing, contain according to
Particular order occur multiple operations, but it should be clearly understood that these operation can not herein occur according to it is suitable
Sequence is performed or performed parallel, and the sequence number such as 101,102 etc. of operation is only used for distinguishing each different operation, sequence number
Any execution sequence is not represented for itself.In addition, these flows can include more or less operations, and these operations can
To perform or perform parallel in order.It should be noted that the description such as " first ", " second " herein, is to be used to distinguish not
Same message, equipment, module etc., does not represent sequencing, it is different types also not limit " first " and " second ".
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, the every other implementation that those skilled in the art are obtained under the premise of creative work is not made
Example, belongs to the scope of protection of the invention.
The schematic flow sheet for the information search method that Fig. 1 provides for one embodiment of the invention.As shown in figure 1, this method bag
Include：
101st, according to keyword, the first search result is searched out.
102nd, based on first search result, search need is determined.
103rd, search meets the second search result of the search need.
In above-mentioned 101, the first search result matched with keyword can be searched for by the way of existing text matches.Tool
The search principle of body can be found in prior art, and here is omitted.
In above-described embodiment and following embodiments, above-mentioned the first search result searched out can be one or more.
When first search result is one, search need can be determined by analyzing the content information in the first search result.
For example, in class application scenarios are taken out, the keyword of user's search is " KFC "；Search out the letter in a KFC shop
Breath.In a kind of achievable scheme, the KFC shop corresponding label dictionary { western fast food, fried chicken, hamburger, coffee are obtained
Coffee }, directly using each label in the label dictionary as search need, to be scanned for again based on search need, find out tool
There is the second search result of these labels.
When first search result is multiple, Commonness Analysis can be carried out to multiple first search results, to obtain this
The common feature of a little first search results；Then the search need is determined according to common feature.Certainly, in actual applications
Also directly the common feature can be regard as the search need.Wherein, Commonness Analysis can be realized using following several ways：
Mode one：The corresponding label dictionary of each first search result is obtained, then multiple label dictionaries to getting
In label collected, will appear from label often as search need.For example, according to keyword search to first search
Hitch fruit has：KFC shop, Pizza Hut shop, Subway shop and McDonald shop.Wherein, the corresponding mark in KFC shop
Allusion quotation of signing is { western fast food, fried chicken, hamburger, coffee }；The corresponding label dictionary in Pizza Hut shop is { western fast food, Pizza, cake
Select, afternoon tea；The corresponding label dictionary in Subway shop is { western fast food, hamburger }；The corresponding label dictionary in McDonald shop
For { western fast food, fried chicken, hamburger, coffee }；The occurrence number for collecting each label in above-mentioned label dictionary is obtained：Western fast food is
4 times, fried chicken be 2 times, hamburger be 3 times, coffee be 2 times, Pizza be 1 time, cake be 1 time, afternoon tea be 1 time；By above-mentioned appearance
Number of times is more than or equal to the label of 2 times as common feature, obtains common feature for { western fast food, fried chicken, hamburger, coffee }.
Mode two：Obtain include in the corresponding label dictionary of each first search result, the label dictionary label and
The corresponding weight of label；Then the weight of label in the multiple label dictionaries got is weighted and collected；Finally will weighting
Summarized results comes N (N is the first default value, and the value can be manually set) individual label above as common feature.For example, root
According to keyword search to the first search result have：KFC shop, Pizza Hut shop, Subway shop and McDonald shop.
Wherein, the corresponding label dictionary in KFC shop is { western fast food：2nd, fried chicken：2nd, hamburger：2nd, coffee：1}；Pizza Hut shop pair
The label dictionary answered is { western fast food：2nd, Pizza：2nd, cake：1st, afternoon tea:1}；The corresponding label dictionary in Subway shop is
{ western fast food：2nd, hamburger：4}；The corresponding label dictionary in McDonald shop is { western fast food:2nd, fried chicken:2nd, hamburger：2nd, coffee：
1}.The weight of each label in all label dictionaries is weighted to collect and obtained：Western fast food=2+2+2+2=8；Fried chicken=
2+2=4；Hamburger=2+2+4=8；Coffee=1+1=2；Pizza=2；Cake=1；Afternoon tea=1；By above-mentioned each label according to
Summarized results is ranked up：Western fast food, hamburger, fried chicken, coffee, Pizza, cake, afternoon tea；Summarized results is come preceding 4
Label obtains common feature for { western fast food, hamburger, fried chicken, coffee } as common feature.Certainly, in the specific implementation for
It the efficiency of aggregation process is improved, can first find out the label repeated from multiple label dictionaries, then to repeating
The weight of label, which is weighted, to collect.
Mode three, obtains the content of each first search result, general character is extracted from the content of all first search results
Information, regard common information as the common feature.For example, in class application scenarios are taken out, search result is believed for multiple shops
Breath, obtains all dish informations in each shop, extracts the vegetable having in all shops as common feature.
In above-mentioned 103, when search meets the second search result of search need, it can recall all in hunting zone first
Data, then calculate the data in hunting zone and the matching degree of search need, matching degree are met to the number for matching and requiring respectively
According to being used as the second search result.Specifically, above-mentioned 103 can adopt and realize with the following method：Obtain the data correspondence in hunting zone
Label dictionary；The corresponding label dictionary of the data is matched with the search need；Will if matching degree meets matching
Ask, then regard the data as second search result.
The technical scheme that the present embodiment is provided, search need is determined according to the first search result searched, then based on searching
Rope demand is scanned for, and the second search result searched easily meets the search need of user close to the real demand of user.
Further, in above-described embodiment and following embodiments, when first search result searched is multiple
When, the above method may also include：Multiple first search results are ranked up.Accordingly, it is above-mentioned based on the first search
As a result, search need is determined, can be specially：Based on M first search results before coming, search need is determined.Wherein, institute
M is stated for the second default value, second default value can artificially be set.General character is carried out from M the first search results before coming
Analysis, one is that can reduce the data volume of analysis, improves analysis efficiency；Another is sort more forward search result and key
The matching degree of word is higher, therefore the common feature that the search result progress Commonness Analysis high to matching degree is obtained will be more accurate.
Such a case occurs in actual applications：When being scanned for using keyword in existing hunting zone,
Search is less than the first search result.For example, in class application scenarios are taken out, hunting zone is the geographical position according to where user
Come what is determined；Assuming that the geographical position that user is currently located is Xizhimen, the user using application when being scanned for, its correspondence
Hunting zone be exactly all data (such as trade company) residing for Xizhimen in commercial circle.Now, if user * * * trade companies to be searched for, and
In hunting zone residing for it without * * * trade companies, then scanned for using keyword, search result is sky.Now,
Search need can not be determined based on the first search result.For such case, the invention provides another embodiment.It is following to implement
The first search result and the second search result are the data in two different hunting zones in example.Fig. 2 is implemented for the present invention is another
The schematic flow sheet for the information search method that example is provided.As shown in Fig. 2 this method includes：
201st, in the first hunting zone, multiple first search results are gone out according to keyword search.
202nd, multiple first search results are ranked up.
203rd, Commonness Analysis are carried out to coming preceding M first search results, obtains the common feature.
204th, according to the common feature, the search need is determined.
205th, the corresponding label dictionary of data in second hunting zone is obtained.
206th, the corresponding label dictionary of data in second hunting zone is matched with the search need.
If the 207, matching degree meets matching and required, the data are regard as second search result.
If the 208, second search result is multiple, multiple second search results are arranged according to matching degree
Sequence.
The technical scheme that the present embodiment is provided, determines to search based on the first search result searched in the first hunting zone
Rope demand, to search out the second search result for meeting search need in the second hunting zone, it is to avoid exist in the prior art
There is search result because keyword can not be matched in second hunting zone for empty situation to occur；In addition, being needed based on search
Ask to scan for, the result searched is closer to user's real demand, and the search degree of accuracy is high.
Above-mentioned steps 202 in the specific implementation, can be entered according to the ordering rule of setting to multiple first search results
Row sequence.Wherein, ordering rule can be：With the matching degree of keyword it is higher come before, or according to pre-set level
Quantity is ranked up etc..Wherein, application scenarios are different, and pre-set level can be different.For example, taking out class application scenarios
Under, the first search result is merchant information；Accordingly, pre-set level can be that the order volume of trade company, the history of trade company are searched
Rope number of times etc..In above-described embodiment and following embodiments, can according to trade company day order volume or moon order volume etc., will order
Before many trade companies of single amount come.Above-mentioned N numerical value can be manually set, for example, N takes 10,20 etc..Obtain sequence preceding 10 or 20
Individual first search result is analyzed, and one is that can reduce the data processing amount of analysis, improves analysis efficiency；Another is to improve
The accuracy of Commonness Analysis.Because in the case where taking out class application scenarios, being that user recommends the high trade company of temperature to user, more expiring
The demand of sufficient user, user's search experience Du Genggao.
M first search results carry out Commonness Analysis before 203 pairs of above-mentioned steps come, and obtain the common feature,
It can specifically adopt and realize with the following method：
S1, multiple first search results of acquisition, which are distinguished in corresponding label dictionary, the label dictionary, includes mark
Label and the corresponding weight of label.
S2, the weight of label in multiple label dictionaries is weighted and collected.
S3, the label that summarized results comes to top n are used as the common feature.
Wherein, the N is the first default value.
For example, in class application scenarios are taken out, multiple first search results searched out are merchant information.Obtain each
The corresponding label dictionary tags={ tag1 of trade company:W1, tag2:W2 ..., tagi:wi,…,tagN:WN }, wherein wi is i-th
Weight of the individual tag (label) in current trade company tags, N is the tag quantity of current trade company；Then according to temperature (such as order volume
Number) all tag of trade company in target_shops are weighted and collected, it is final to obtain preceding 10 (N values) individual weight is most
Big tag, is denoted as mean_tags={ m_tag1:M_w1, m_tag2:M_w2 ..., m_tagi:m_wi,…,m_tagN:m_
WN }, it is used as user search query demand dictionary.
A kind of above-mentioned 204 achievable mode is directly to regard the common feature as the search need；For example, general character
It is characterized as { hamburger, French fries }, { hamburger, French fries } is directly used as search need.Another achievable mode is by described
Common feature is analyzed, and regard analysis result as search need.Specifically, default Requirements Analysis Model can be used, by institute
Common feature is stated as the input of the model, corresponding search need is analyzed.Wherein, the Requirements Analysis Model can artificially be set
It is fixed.
In above-mentioned 206, the corresponding label dictionary of data in second hunting zone and the search need are carried out
Matching, its matching process can be realized using cosine similarity method.For example, the corresponding label of data in second hunting zone
Dictionary D1=(wi1, wi2..., win), search need D2=(wj1, wj2..., wjn), then the data in the second hunting zone are corresponding
The cosine similarity of label dictionary and the search need is calculated as：
Cosine similarity minimum value is 0, and maximum is 1.Its geometric meaning of cosine similarity is：Two vectors are calculated in higher-dimension
The angle spent in space, angle is smaller, and cosine similarity is bigger, and (when angle is 0,1) similarity is；Angle is bigger, and cosine is similar
Degree is smaller, and (when 90 degree of angle is orthogonal, 0) similarity is.
Further, in above-described embodiment and following embodiments, the above method may also include：Obtain multiple described second
The characteristic information of search result.Accordingly, it is above-mentioned that multiple second search results are ranked up according to matching degree, bag
Include：With reference to the matching degree and the characteristic information, multiple second search results are ranked up.Wherein, application scenarios
Difference, above-mentioned characteristic information would also vary from.For example, in application scenarios are taken out, the second search result is merchant information,
The corresponding characteristic information of merchant information may include：It is distance (path length of the merchant location apart from user current location), price, excellent
Favour rate etc..In the specific implementation, can using matching degree, distance, price and concessionary rate as sequence consider the factor, to it is above-mentioned because
Son is weighted read group total score, is then ranked up according to the height of score.
Further, in above-described embodiment and following embodiments, the above method may also include：
In second hunting zone, the 3rd search result matched with the keyword is searched out；
If the quantity of the 3rd search result searched out is less than threshold value, expands second hunting zone and obtain institute
State the first hunting zone.
On process nature described by the step can as above-described embodiment startup event, i.e., in the second hunting zone
When searching seldom with Keywords matching the 3rd search result, expand hunting zone and simultaneously analyze search need, return the
Two hunting zones are scanned for based on search need.Certainly, in actual applications, on the execution of above-described embodiment can not needed yet
The event triggering described by step is stated, any search all can first search for the 3rd search result in the second hunting zone, then arrive
Searched in first hunting zone to obtain search need, return to search for the second hunting zone in and meet the second of search need and search
Hitch fruit.Search result can be so enriched, a variety of demands of user are met.
Further, in above-described embodiment and following embodiments, search obtains the second search result and the 3rd search result
Afterwards, second search result and the 3rd search result can be sent to client, with the corresponding search interface of client
Showed.When showing, the second search result and the 3rd search result can be ranked up, for example, the second search can be tied
Fruit is come after the 3rd search result.Accordingly, above-mentioned method may also include the steps of：
For second search result, rearmounted display properties is set.
Wherein, the rearmounted display properties, which is indicated, is placed in the described 3rd behind the display location of second search result and searches
Hitch fruit.In other words, it will be presented in and scanned for using search need with the literal search result matched of keyword
Before the search result arrived.Client can be searched after the 3rd search result and the second search result is received according to described second
The rearmounted display properties of hitch fruit, second search result is included to be shown after the 3rd search result.
The first hunting zone mentioned in each embodiment is two different search models with the second hunting zone herein
Enclose.Wherein, first hunting zone can include the second hunting zone, for example, in the case where taking out class application scenarios, hunting zone is
Geographical position according to where user is determined.Assuming that the geographical position that user is currently located is Xizhimen, the second search model
Enclose all trade companies in commercial circle residing for Xizhimen.First hunting zone can be all trade companies in all commercial circles in whole city.When
So, first hunting zone can not include the second hunting zone yet.For example, the second hunting zone is residing for Xizhimen in commercial circle
All trade companies；First hunting zone be commercial circle residing for described Xizhimen close to commercial circle in all trade companies.
Below exemplified by taking out class application scenarios, the client end AP P that user is provided using O2O platforms searches for it and wants inquiry
Trade company.The schematic flow sheet for the information search method that Fig. 3 provides for further embodiment of this invention.As shown in figure 3, this method bag
Include：
301st, in corresponding second hunting zone in region that user is presently in, the business that search matches with keyword
Family, the 3rd search result is designated as by the trade company searched.
If the 302, the 3rd search result is less than threshold value for the quantity of empty or described 3rd search result, expand described
Second hunting zone obtains the first hunting zone.
Need exist for explanation be：The present embodiment illustrate only by extended mode to obtain the first hunting zone
Mode, substantially can also be by way of changing hunting zone.
303rd, search and the trade company of Keywords matching in the first hunting zone, the first search is designated as by the trade company searched
As a result.
If the 304, the first search result is multiple, multiple first search results are arranged according to trade company's temperature
Sequence.
Wherein, the temperature of trade company can be embodied in more than order volume, or is searched often.For example, according to trade company
Blanket order amount be ranked up, or be ranked up according to searched number of times.
305th, to the first search result of temperature highest top n added in the set of target trade company, and by the target trade company
Set is denoted as target_shops.
306th, each corresponding label dictionary tags={ tag1 of trade company in target_shops are obtained:W1, tag2:
W2 ..., tagi:wi,…,tagN:wN}.
Wherein, wi is weight of i-th of tag (label) in all tag of current trade company, and N is the tag numbers of current trade company
Amount.
307th, each tag in the corresponding label dictionary of all trade companies in target_shops is weighted and collected.
308th, the tag of M, is denoted as demand label mean_tags={ m_tag1 before summarized results is come:M_w1, m_
tag2:M_w2 ..., m_tagi:m_wi,…,m_tagN:m_wN}.
Wherein, m_tagi is the label of i before summarized results comes.M_wi collects knot for i-th of demand label is corresponding
Really.
309th, in corresponding second hunting zone in region that user is presently in, all trade companies is recalled and obtain trade company's set
recall_shops。
310th, the label dictionary recall_tags={ r_tag1 of each trade company in recall_shops are obtained:R_w1, r_
tag2:R_w2 ..., r_tagi:r_wi,…,r_tagN:r_wN}.
311st, the recall_tags and demand label mean_tags of above-mentioned each trade company matching degree match_ are calculated
ratio。
The trade company that the 312nd, match_ratio is met to matching requirement is used as the second search result.
If the 313, second search result is multiple, the characteristic information of second search result is obtained.
314th, with reference to match_ratio and characteristic information, multiple second search results are ranked up.
315th, rearmounted display properties is set for second search result.
Wherein, the rearmounted display properties, which is indicated, is placed in the described 3rd behind the display location of second search result and searches
Hitch fruit.
If the 316, in the presence of the 3rd search result, the 3rd search result and second search result are sent to visitor
Family end.
Client is received after the 3rd search result and the second search result, and it is shown.The process of displaying
In, client can will be placed in the 3rd according to the rearmounted display properties of the second search result behind the display location of the second search result
Search result.To allow user to see the 3rd search result with Keywords matching at first, subsequently see again and meet user's search need
The second search result asked.
On above-mentioned 301~316, reference can be made to the description of foregoing embodiments, here is omitted.
It should be noted that：The executive agent that above-described embodiment provides each step of method may each be same equipment,
Or, this method is also used as executive agent by distinct device.Such as, the executive agent of step 101 to step 103 can be equipment
A；Again such as, step 101 and 102 executive agent can be device A, and the executive agent of step 103 can be equipment B；Etc..
The structural representation for the information retrieval device that Fig. 4 provides for one embodiment of the invention.As shown in figure 4, the present embodiment
The device of offer includes：First search module 410, the search module 430 of determining module 420 and second.Wherein, the first search module
410 are used for according to keyword, search out the first search result.Determining module 420 is used to be based on first search result, it is determined that
Search need.Second search module 430 is used to search for the second search result for meeting the search need.
The technical scheme that the present embodiment is provided, search need is determined according to the first search result searched, then based on searching
Rope demand is scanned for, and the second search result searched easily meets the search need of user close to the real demand of user.
First search result described in above-described embodiment can be one or more.When the first search result is one,
Above-mentioned determining module 420 is additionally operable to：Content analysis is carried out to the described first search section, with search need.When the first search knot
When fruit is multiple, above-mentioned determining module can be realized using the structure shown in Fig. 5.Specifically, the determining module 420 includes dividing
Analyse unit 421 and determining unit 422.Wherein, analytic unit 421 is used to carry out general character point to multiple first search results
Analysis, obtains common feature.Determining unit 422 is used for according to the common feature, determines the search need.
Further, above-mentioned analytic unit 421 is additionally operable to：Obtain multiple first search results and distinguish corresponding mark
Include label and the corresponding weight of label in signature allusion quotation, the label dictionary；By the power of label in multiple label dictionaries
It is weighted and collects again；The label that summarized results is come to top n is used as the common feature；Wherein, the N is first default
Numerical value.
As shown in figure 5, the information retrieval device that above-described embodiment is provided may also include：First order module 440.Described
One order module 440 is used to be ranked up multiple first search results.Accordingly, above-mentioned analytic unit is also used：To row
Commonness Analysis is carried out in preceding M first search results, the common feature is obtained.Wherein, the M is the second present count
Value.
Further, the second above-mentioned search module 430 is additionally operable to：Obtain the corresponding mark of data in the hunting zone
Signature allusion quotation；The corresponding label dictionary of the data is matched with the search need；If matching degree meets matching and required,
It regard the data as second search result.
Further, as shown in figure 5, above-mentioned information retrieval device may also include：Second order module 450.Second row
Sequence module 450 is used for：If second search result is multiple, multiple second search results are carried out according to matching degree
Sequence.
Further, as shown in figure 5, above-mentioned information search module may also include：Acquisition module 460.The acquisition mould
Block 460 is used for the characteristic information for obtaining multiple second search results.Accordingly, the second above-mentioned order module 450 is also used
In：With reference to the matching degree and the characteristic information, multiple second search results are ranked up.
Further, first search result is the data in the first hunting zone；Second search result is the
Data in two hunting zones；And first search module 410 is additionally operable in the first hunting zone, according to the key
Word searches out first search result；Second search module 430 is additionally operable in the second hunting zone, and search meets institute
State the second search result of search need.Determine that search is needed based on the first search result searched in the first hunting zone
Ask, to search out the second search result for meeting search need in the second hunting zone, it is to avoid in the prior art second
There is search result because keyword can not be matched in hunting zone for empty situation to occur；In addition, based on search need come
Scan for, the result searched is closer to user's real demand, and the search degree of accuracy is high.
Further, as shown in figure 5, above-mentioned information search module may also include：3rd search module 470 and processing mould
Block 480.Wherein, the 3rd search module 470 is used in second hunting zone, searches out what is matched with the keyword
3rd search result.If the quantity that processing module 480 is used for the 3rd search result searched out is less than threshold value, expand institute
State the second hunting zone and obtain first hunting zone.
Further, as shown in figure 5, above-mentioned information search module may also include：Setup module 490.The setting mould
Block 490 is used to set rearmounted display properties for second search result.Wherein, the rearmounted display properties indicates described
The 3rd search result is placed in behind the display location of two search results.
Need exist for explanation be：The information retrieval device that above-described embodiment is provided can realize that above-mentioned each method embodiment is carried
The technical scheme of confession.Each module or unit implements principle reference can be made to the above method is real in the embodiment of information retrieval device
Corresponding content in example is applied, here is omitted.
Device embodiment described above is only schematical, wherein the unit illustrated as separating component can
To be or may not be physically separate, the part shown as unit can be or may not be physics list
Member, you can with positioned at a place, or can also be distributed on multiple NEs.It can be selected according to the actual needs
In some or all of module realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying creativeness
Work in the case of, you can to understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
Realized by the mode of software plus required general hardware platform, naturally it is also possible to pass through hardware.Understood based on such, on
The part that technical scheme substantially in other words contributes to prior art is stated to embody in the form of software product, should
Computer software product can be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disc, CD, including some fingers
Order is to cause a computer equipment (can be personal computer, server, or network equipment etc.) to perform each implementation
Method described in some parts of example or embodiment.
Finally it should be noted that：The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although
The present invention is described in detail with reference to the foregoing embodiments, it will be understood by those within the art that：It still may be used
To be modified to the technical scheme described in foregoing embodiments, or equivalent substitution is carried out to which part technical characteristic；
And these modification or replace, do not make appropriate technical solution essence depart from various embodiments of the present invention technical scheme spirit and
Scope.
Claims (20)
1. a kind of information search method, it is characterised in that including：
According to keyword, the first search result is searched out；
Based on first search result, search need is determined；
Search meets the second search result of the search need.
2. according to the method described in claim 1, it is characterised in that first search result is multiple；And,
It is described to be based on first search result, search need is determined, including：
Commonness Analysis is carried out to multiple first search results, common feature is obtained；
According to the common feature, the search need is determined.
3. method according to claim 2, it is characterised in that described that general character point is carried out to multiple first search results
Analysis, obtains common feature, including：
Obtain in the corresponding label dictionary of multiple first search results difference, the label dictionary and include label and label
Corresponding weight；
The weight of label in multiple label dictionaries is weighted and collected；
The label that summarized results is come to top n is used as the common feature；
Wherein, the N is the first default value.
4. according to the method in claim 2 or 3, it is characterised in that also include：
Multiple first search results are ranked up；And,
It is described that Commonness Analysis is carried out to multiple first search results, common feature is obtained, including：
M first search results carry out Commonness Analysis before to coming, and obtain the common feature.
Wherein, the M is the second default value.
5. according to the method in any one of claims 1 to 3, it is characterised in that the search meets the search need
The second search result, including：
Obtain the corresponding label dictionary of data in hunting zone；
The corresponding label dictionary of the data is matched with the search need；
If matching degree meets matching and required, the data are regard as second search result.
6. method according to claim 5, it is characterised in that also include：
If second search result is multiple, multiple second search results are ranked up according to the matching degree.
7. method according to claim 6, it is characterised in that also include：
Obtain the characteristic information of multiple second search results；
And, it is described that multiple second search results are ranked up according to matching degree, including：
With reference to the matching degree and the characteristic information, multiple second search results are ranked up.
8. according to the method in any one of claims 1 to 3, it is characterised in that first search result is searched for first
Data in the range of rope；Second search result is the data in the second hunting zone；And
It is described that first search result is searched out according to keyword, including：
In the first hunting zone, first search result is gone out according to the keyword search；
The search meets the second search result of the search need, including：
In the second hunting zone, search meets the second search result of the search need.
9. the method according to any one of claim 8, it is characterised in that also include：
In second hunting zone, the 3rd search result matched with the keyword is searched out；
If the quantity of the 3rd search result searched out is less than threshold value, expands second hunting zone and obtain described the
One hunting zone.
10. method according to claim 9, it is characterised in that also include：
For second search result, rearmounted display properties is set；
Wherein, the rearmounted display properties, which is indicated, is placed in the 3rd search knot behind the display location of second search result
Really.
11. a kind of information retrieval device, it is characterised in that including：
First search module, for according to keyword, searching out the first search result；
Determining module, for based on first search result, determining search need；
Second search module, the second search result of the search need is met for searching for.
12. device according to claim 11, it is characterised in that first search result is multiple；And
The determining module, including：
Analytic unit, for carrying out Commonness Analysis to multiple first search results, obtains common feature；
Determining unit, for according to the common feature, determining the search need.
13. device according to claim 12, it is characterised in that the analytic unit is additionally operable to：
Obtain in the corresponding label dictionary of multiple first search results difference, the label dictionary and include label and label
Corresponding weight；
The weight of label in multiple label dictionaries is weighted and collected；
The label that summarized results is come to top n is used as the common feature；
Wherein, the N is the first default value.
14. the device according to claim 12 or 13, it is characterised in that also include：
First order module, for being ranked up to multiple first search results；And
The analytic unit is additionally operable to：M first search results carry out Commonness Analysis before to coming, and obtain the general character special
Levy.
Wherein, the M is the second default value.
15. the device according to any one of claim 11 to 13, it is characterised in that second search module is also used
In：
Obtain the corresponding label dictionary of data in hunting zone；
The corresponding label dictionary of the data is matched with the search need；
If matching degree meets matching and required, the data are regard as second search result.
16. device according to claim 15, it is characterised in that also include：
Second order module, if being multiple for second search result, according to the matching degree to multiple described second
Search result is ranked up.
17. device according to claim 16, it is characterised in that also include：
Acquisition module, the characteristic information for obtaining multiple second search results；
And, second order module is additionally operable to：
With reference to the matching degree and the characteristic information, multiple second search results are ranked up.
18. the method according to any one of claim 11 to 13, it is characterised in that first search result is first
Data in hunting zone；Second search result is the data in the second hunting zone；And
First search module, is additionally operable in the first hunting zone, and first search is gone out according to the keyword search
As a result；
Second search module, is additionally operable in the second hunting zone, and search meets the second search knot of the search need
Really.
19. device according to claim 18, it is characterised in that also include：
3rd search module, in second hunting zone, searching out the 3rd search matched with the keyword
As a result；
Processing module, if the quantity of the 3rd search result for searching out is less than threshold value, expands second search
Scope obtains first hunting zone.
20. device according to claim 19, it is characterised in that also include：
Setup module, for setting rearmounted display properties for second search result；
Wherein, the rearmounted display properties, which is indicated, is placed in the 3rd search knot behind the display location of second search result
Really.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611045792.XA CN107092621A (en) | 2016-11-24 | 2016-11-24 | Information search method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611045792.XA CN107092621A (en) | 2016-11-24 | 2016-11-24 | Information search method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107092621A true CN107092621A (en) | 2017-08-25 |
Family
ID=59648673
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611045792.XA Pending CN107092621A (en) | 2016-11-24 | 2016-11-24 | Information search method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107092621A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109636536A (en) * | 2018-12-19 | 2019-04-16 | 拉扎斯网络科技（上海）有限公司 | Service product providing method, device, electronic equipment and storage medium |
| CN110020082A (en) * | 2017-12-11 | 2019-07-16 | 北京搜狗科技发展有限公司 | A kind of searching method and device |
| CN111461833A (en) * | 2020-03-31 | 2020-07-28 | 拉扎斯网络科技（上海）有限公司 | Information loading method, system, server, terminal and storage medium |
| CN111881352A (en) * | 2020-07-28 | 2020-11-03 | 腾讯科技（北京）有限公司 | Content pushing method and device, computer equipment and storage medium |
| CN112559903A (en) * | 2020-12-15 | 2021-03-26 | 广州市贺氏办公设备有限公司 | Community member searching method, system, device and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110295688A1 (en) * | 2010-05-28 | 2011-12-01 | Microsoft Corporation | Defining user intent |
| CN102999520A (en) * | 2011-09-15 | 2013-03-27 | 北京百度网讯科技有限公司 | Method and device for identifying search request |
| CN103020066A (en) * | 2011-09-21 | 2013-04-03 | 北京百度网讯科技有限公司 | Method and device for recognizing search demand |
-
2016
- 2016-11-24 CN CN201611045792.XA patent/CN107092621A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110295688A1 (en) * | 2010-05-28 | 2011-12-01 | Microsoft Corporation | Defining user intent |
| CN102999520A (en) * | 2011-09-15 | 2013-03-27 | 北京百度网讯科技有限公司 | Method and device for identifying search request |
| CN103020066A (en) * | 2011-09-21 | 2013-04-03 | 北京百度网讯科技有限公司 | Method and device for recognizing search demand |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110020082A (en) * | 2017-12-11 | 2019-07-16 | 北京搜狗科技发展有限公司 | A kind of searching method and device |
| CN109636536A (en) * | 2018-12-19 | 2019-04-16 | 拉扎斯网络科技（上海）有限公司 | Service product providing method, device, electronic equipment and storage medium |
| CN111461833A (en) * | 2020-03-31 | 2020-07-28 | 拉扎斯网络科技（上海）有限公司 | Information loading method, system, server, terminal and storage medium |
| CN111881352A (en) * | 2020-07-28 | 2020-11-03 | 腾讯科技（北京）有限公司 | Content pushing method and device, computer equipment and storage medium |
| CN112559903A (en) * | 2020-12-15 | 2021-03-26 | 广州市贺氏办公设备有限公司 | Community member searching method, system, device and medium |
| CN112559903B (en) * | 2020-12-15 | 2024-04-05 | 广州市贺氏办公设备有限公司 | Community member searching method, system, device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424043B (en) | Product recommendation method and device and electronic equipment | |
| WO2019056661A1 (en) | Search term pushing method and device, and terminal | |
| CN107092621A (en) | Information search method and device | |
| US8078601B1 (en) | Determining unambiguous geographic references | |
| US9767144B2 (en) | Search system with query refinement | |
| CN110020128B (en) | Search result ordering method and device | |
| JP2020523714A (en) | Recommended information acquisition method and device, electronic device | |
| CN110532351B (en) | Recommendation word display method, device and equipment and computer readable storage medium | |
| CN103136228A (en) | Image search method and image search device | |
| CN110413875A (en) | A kind of method and relevant apparatus of text information push | |
| CN103377200A (en) | Method and device for collecting user preference information | |
| CN106886518A (en) | A kind of method of microblog account classification | |
| CN106776860A (en) | One kind search abstraction generating method and device | |
| CN102968419B (en) | Disambiguation method for interactive Internet entity name | |
| CN103577432A (en) | Method and system for searching commodity information | |
| WO2018176913A1 (en) | Search method and apparatus, and non-temporary computer-readable storage medium | |
| CN108431829A (en) | System and method for searching for product in catalogue | |
| CN110287307A (en) | A kind of search result ordering method, device and server | |
| CN111400507A (en) | Entity matching method and device | |
| CN113254810B (en) | Search result output method and device, computer equipment and readable storage medium | |
| CN103902549B (en) | The method and apparatus for searching for data sorting, the method and apparatus of data search | |
| CN110609958A (en) | Data pushing method and device, electronic equipment and storage medium | |
| CN112488781A (en) | Search recommendation method and device, electronic equipment and readable storage medium | |
| CN112084307A (en) | Data processing method and device, server and computer readable storage medium | |
| CN114820123A (en) | Group purchase commodity recommendation method, device, equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170825 |
|
| RJ01 | Rejection of invention patent application after publication |