CN106203646A - Customize stroke commending system and method - Google Patents
Customize stroke commending system and method Download PDFInfo
- Publication number
- CN106203646A CN106203646A CN201610499550.1A CN201610499550A CN106203646A CN 106203646 A CN106203646 A CN 106203646A CN 201610499550 A CN201610499550 A CN 201610499550A CN 106203646 A CN106203646 A CN 106203646A
- Authority
- CN
- China
- Prior art keywords
- data
- user
- stroke
- unit
- content
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/02—Reservations, e.g. for tickets, services or events
- G06Q10/025—Coordination of plural reservations, e.g. plural trip segments, transportation combined with accommodation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/109—Time management, e.g. calendars, reminders, meetings or time accounting
- G06Q10/1093—Calendar-based scheduling for persons or groups
Abstract
The invention discloses customization stroke commending system and method, system includes: data acquisition unit, obtains content-data from WEB terminal, and content-data includes structural data or unstructured data；Data structured unit, directly deposits structured data to data base, and after carrying out data structured process according to the dependency between key word according to the setting tuple of user in non-structural data and be stored in data base；User emotion analytic unit, infers the user emotion of correspondence, result is updated to data base simultaneously according to the user's one section of experience in stroke；Routing unit, according to the content-data in the emotion analysis result in described user emotion analytic unit and described data structured unit, customizes out travel schedule.The present invention fully obtains relevant content information from multiple Data Sources, and quickly analyzes effective focus trend, in conjunction with the psychology and behavior of consumption of user, helps user immediately to process bulk information, and being customized stroke is recommended.
Description
Technical field
The present invention relates to commending contents field, particularly to customizing stroke commending system and method.
Background technology
Along with development and the raising of living standard of economic society, the spiritual cultural demand of the common people increases rapidly, in
Feature the most multi-level, multi-form, diversified, cultural consumption ability is greatly enhanced, and the pursuit to leisure way also gets more and more
Sample.But information that on Web, tourism and leisure is relevant is numerous and jumbled and scattered, content is very different, it is desirable to from the data that these are mixed and disorderly
In the effective cost obtaining useful information more and more higher.And along with the transformation of life style, the common people follow blindly the most simply, more
Many is to pursue individual unique need and showing that self samples, it is individual that this makes popular judgment criteria cannot meet
Property demand.And provide stereotyped with lack the place of intention and trip scheme can not attract 80, this group of consumption after 90s are main
Body, they are it is desirable that the service of the personalization experience sense interesting with novelty is subject to, it is intended to experience sense is strong, and meets oneself product
The leisure way of taste, will allow and oneself be different from other people of circle of friends.
But at present, great majority are engaged in the company of association area and are only to provide the recommendation in single place, these purely by
Editor is according to the fine work of individual's feelings packaging, it is not necessary to be applicable to its location crowd.Study carefully its psychology to understand, compared to stranger
Or the versatility recommendation of tourism intelligent, user is more likely to the recommendation based on self preference, or likes similar good friend's suggestion.And
On Web, although user can advise to the stroke of some with Free Acquisition, but these machine-made trip attack strategys can not be expired
Those crowds having individual character pursuit and custom to utilize the fragmentation time to go on a journey sufficient.Although user is accustomed to passing through social network the most very much
Their tourism and leisure trip experience is shared in network service, but these services but cannot provide customization technically for them
The travel schedule changed.Additionally, each user is intended to the trip requirements of oneself, i.e. travel time, place and scene when trip
Deng, it is possible to sufficiently met.And based on existing technology and implementation method, user first must search for tourism voluntarily from Web
With the leisure relevant information of trip, plan trip scheme further according to oneself trip requirements, thus consume the substantial amounts of time with
Energy.
Summary of the invention
The technical problem to be solved in the present invention is, fully obtains relevant content information from multiple Data Sources, and quickly
Analyze effective focus trend, in conjunction with the psychology and behavior of consumption of user, help user immediately to process bulk information, carry out
Customization stroke is recommended.
Solve above-mentioned technical problem, the invention provides a kind of customization stroke commending system, it is characterised in that including: number
According to collecting unit, data structured unit, user emotion analytic unit and routing unit,
Described data acquisition unit, in order to from WEB terminal obtain content-data, described content-data include structural data or
Person's unstructured data；
Described data structured unit, in order to directly to deposit described structured data to data base, and according to user's
Set after tuple carries out data structured process according to the dependency between key word in described non-structural data and be stored in
In data base；
Described user emotion analytic unit, in order to infer user's feelings of correspondence according to the user's one section of experience in stroke
Thread, is updated to data base by result simultaneously；
Described routing unit, in order to according to the emotion analysis result in described user emotion analytic unit and described number
According to the content-data in structuring unit, customize out travel schedule.
Further, described user set tuple as:
{ time, place, scene }
The described time includes user's stroke start and end time, and described place includes stroke departure place and destination, institute
State scene and include user-defined travel type.
Further, described routing unit, in order to by described data structured unit by time, place and field
The decision tree of scape composition sets according to the trip of user, obtains travel schedule.
Further, described travel schedule is obtained according to such as down stroke custom algorithm:
By CFB unconfined condition algorithm, according to the summation of the evaluation score in all places to go included in traffic path
Obtain the recommendation of route；
Described recommendation computational methods are:
The total length of described route by this route comprised all go between total sum of distance,
Further according to the recommendation of the route of starting point to another place to go, and gone by the method traversal one of above-mentioned recommendation
Virgin concentrates all places to go, then select the traffic path planning of optimum；
Or,
By CBB constraints algorithm, the constraints provided according to user, carry out route planning；
The summation of the evaluation score according to all places to go included in traffic path obtains the recommendation of route；
Further according to the recommendation of the route of starting point to another place to go, and gone by the method traversal one of above-mentioned recommendation
Virgin concentrates all places to go；If route is in constraints, then calculate the best route rule that can meet described constraints
Drawing, described constraints at least includes, the decision tree being made up of time, place and scene.
Further, obtain content-data from WEB terminal and include,
By accessing SNS server, data base searches the content-data in social networks, and rings in WEB terminal
Should；
And/or, by accessing content server, data base searches the content-data published or pay close attention to, and
Respond in WEB terminal；
Further, described data acquisition unit, also in order to carry out following pretreatment:
6-1) carry out unstructured data excavation, and structure data entry according to the stroke special topic of content-data；
6-2) according to natural language processing algorithm, described unstructured data is carried out pretreatment.
Further, described data structured unit, also in order to,
7-1) according to the time in described tuple, place and scene, described unstructured data is changed；
7-2) according to semantic weight, the data after said structure are classified；
7-3) generate stroke to be correlated with corpus prototype.
Further, described user emotion analytic unit, also in order to,
8-1) utilize natural language emotion analytical technology that the emotion of user is analyzed extraction and obtain corresponding key word；
8-2) depict user for the tectonic information needed for the portrait of the preference of stroke according to described key word；
8-3) extract from described tectonic information represent user interest point key word and to user stroke experience in
User emotion understanding.
Further, described routing unit, also in order to,
9-1) the trip scene settings given based on user goes out a stroke list, the row corresponding in order to show different scene
Journey；
9-2) when customizing travel schedule, increase instant data message, in order to carry out different weight couplings.
Based on said system, present invention also offers a kind of stroke that customizes and recommend method, including:
Obtaining content-data from WEB terminal, described content-data includes structured data or non-knot structure data；
Described structured data is directly deposited to data base, and according to the setting tuple of user in described non-structural data
In carry out data structured process according to the dependency between key word after and be stored in data base；
Infer the user emotion of correspondence according to the user's one section of experience in stroke, result is updated to data simultaneously
Storehouse；
According to the content in the emotion analysis result in described user emotion analytic unit and described data structured unit
Data, customize out travel schedule.
Beneficial effects of the present invention:
1) the customization stroke commending system in the present invention, it is provided that data acquisition unit, data structured unit, user
Emotion analytic unit and routing unit, described data acquisition unit, in order to obtain content-data from WEB terminal, described interior
Hold data and include structural data or unstructured data；Described data structured unit, in order to by straight for described structured data
Connect and deposit to data base, and according to the setting tuple of user in described non-structural data according to the dependency between key word
After carrying out data structured process and be stored in data base；Described user emotion analytic unit, in order to according to user at stroke
In one section of experience infer correspondence user emotion, result is updated to data base simultaneously；Described routing unit, in order to
According to the content-data in the emotion analysis result in described user emotion analytic unit and described data structured unit, customization
Go out travel schedule.The present invention is by fully obtaining relevant content information from multiple Data Sources, and quickly analyzes effective
Focus trend, in conjunction with the psychology and behavior of consumption of user, helps user immediately to process bulk information, and being customized stroke pushes away
Recommend.After user's stroke commending system in the application of the invention, greatly save searching and produced user-specific customization
Time cost spent by travel schedule and energy, also greatly facilitate tourism and the leisure trip planning of user simultaneously.
2) due to described data acquisition unit, in order to obtain content-data from WEB terminal, described content-data includes structuring
Data or unstructured data；Data acquisition unit can collect the unstructured data that tourism and leisure trip is relevant simultaneously, and
The data that data acquisition unit collection then can be come by data structured unit are based on one or more time, place and scene
Construction of condition.Each user is intended to oneself trip requirements when trip, i.e. travel time, place and scene etc. can
Enough sufficiently met.
3) due to described user emotion analytic unit, in order to infer correspondence according to the user's one section of experience in stroke
User emotion, is updated to data base by result simultaneously.For user, the impression of oneself certain section of trip experience can be analyzed.
4) due to routing unit, in order to according to the emotion analysis result in described user emotion analytic unit and described
Content-data in data structured unit, customizes out travel schedule, by utilizing in user emotion analytic unit user's
The data handled well in the result of stroke impression analysis and data structured unit generate a customization for this user
Stroke list.Trip requirements planning trip scheme according to user oneself, saves substantial amounts of time and efforts.
Accompanying drawing explanation
Fig. 1 is the structural representation customized in stroke commending system one embodiment of the present invention.
Fig. 2 is the custom algorithm schematic flow sheet of travel schedule in routing unit in Fig. 1.
Fig. 3 is another custom algorithm schematic flow sheet of travel schedule in routing unit in Fig. 1.
Fig. 4 is the setting tuple structure schematic diagram of user in Fig. 1.
Fig. 5 is that the stroke that customizes of the present invention recommends the schematic flow sheet in method one embodiment.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with specific embodiment, and reference
Accompanying drawing, the present invention is described in more detail.
Fig. 1 is the structural representation customized in stroke commending system one embodiment of the present invention.
Customization stroke commending system in the present embodiment, including: data acquisition unit 100, data structured unit
101, user emotion analytic unit 102 and routing unit 103,
Described data acquisition unit 100, in order to obtain content-data from WEB terminal, described content-data includes structuring number
According to or unstructured data；Described data acquisition unit 100 gathers particularly travel and leisure from WEB terminal in order to unification and goes out
The unstructured data that row is relevant.The content-data that WEB terminal obtains refers to, such as social networking service platform, blog and special topic
On website, gather the content-data that tourism is relevant with leisure trip the most automatically.Therefore the core technology of this module is to utilize
To can automatically search for and index the various structurings on webpage and non-structured letter by Web information delver based on theme
These information and are stored in program by breath, and user then need not just to allow journey by any link in webpage clicking
Sequence completes the lookup to new web page and index further according to these information got before repeatedly.This make disposable or
Continuous print gathers substantial amounts of tourism and leisure trip related data and becomes possibility.
In certain embodiments, the acquisition mode of WEB terminal includes but not limited to, WEB web crawlers.
In certain embodiments, the content-data of acquisition includes but not limited to, in Sina's microblogging, in wechat circle of friends, discusses
On altar and RSS special subject network station etc..
In certain embodiments, structural data includes but not limited to be stored in lane database, can use bivariate table structure
Carry out the data that logical expression realizes.
In certain embodiments, unstructured data includes but not limited to office documents, text, picture, XML, HTML, each
Class form, image and audio/visual information etc..
Described data structured unit 101, in order to directly to deposit described structured data to data base, and according to user
Setting tuple in described non-structural data, carry out data structured process according to the dependency between key word after and store
In data base；In addition, described data structured unit 101 can also determine data need according to the particular demands of user
Specification to be structured.It is to say, this module can also utilize the tourism and leisure trip that body (Ontology) defines
The relevant dependency between key word or key word carries out structuring process to data, and the structured specification of these data
To be possible not only to be centered by time, place and scene, it is also possible to be other any rule that can meet user's particular demands
Model.Body is that a kind of description is traveled and the conceptual model of relation between leisure trip key word and these key words.The shape of body
Formula can simply can be complicated, it can be formal and have the conceptual model shared in specific area of clear and definite specification, it is also possible to
It is the data base containing related notion.
In certain embodiments, described user set tuple as:
{ time, place, scene }
The described time includes user's stroke start and end time, and described place includes stroke departure place and destination, institute
State scene and include user-defined travel type.
In certain embodiments, described routing unit, in order to by described data structured unit by time, place
And the decision tree of scene composition sets according to the trip of user, obtain travel schedule.
In certain embodiments, key word is set as by user: overseas, diving and honeymoon；Humanity, temple and history
Etc..
In certain embodiments, the dependency between key word includes but not limited to, the conventional degree of key word, key word
Distance, the position of key word and form, word frequency density, link analysis and page weight etc..
Such as, user inputs " our tourism " to the conventional degree of key word, and it is the highest that degree commonly used in " we " this word,
All can occur on a lot of pages, so identification degree and meaning dependency to " our tourism " this search word are contributed very
Little.And " tourism " this word has practical significance, usage degree ratio is relatively low, the meaning contribution to " our tourism " this search word
Value wants big many.So those pages comprising " tourism " this word are the most relevant to " our tourism " this search word.Word frequency
Density refers to, in the case of not having key word to pile up, the frequency that search key word occurs at WEB webpage is the highest, and search is described
Word is the most relevant to the page, such as search " Beijing ", WEB webpage occur frequency higher be Beijing's Imperial Palace, Beijing Great Wall etc..
Position and the form of key word refer to, when being heading label, H1, black matrix such as the form of key word, the most relevant to key word.Crucial
The distance of word refers to, the key word after participle, if complete match, illustrates relevant to search word, such as " hotel accommodations ", " hotel
Subscribe ".Link analysis and page weight refer to, in addition to the factor of the page own.Link and weight between the page also affect pass
The dependency of keyword, most important of which is exactly the word segment of hyperlink on webpage, and the page has the most with search word as anchor
The importing link of word, illustrates that dependency is the strongest.
In some embodiments it may be preferred that data structured module 101 can be based on time, place and scene, will be by data
The unstructured data that acquisition module collection comes carries out structuring.
Described user emotion analytic unit 102, in order to infer the use of correspondence according to the user's one section of experience in stroke
Family emotion, is updated to data base by result simultaneously；User emotion analytic unit 102 can be for gathering the user come to oneself certain section
The emotion of trip experience is analyzed, and infers and draws the most positive or passive result.
In certain embodiments, if emotion is joyful, then include but not limited to those words or word: like, joyful, see
Upper, fascinated, hobby }；Infer and draw actively
If emotion is pleasantly surprised, then include but not limited to those words or word: be smiling happily, happy long, smilingly, Xing Chong
Punching, sound of crying or vomiting, eh, breathe out, he；Infer and draw actively
If emotion is angry, then include but not limited to those words or word: { in a great rage, go up in the air, fire emits three
Zhang, gas；Infer and draw passiveness
If emotion is sad, then include but not limited to those words or word: eh, toot, heart-rending, feel bad, think
Cry }；Infer and draw passiveness
If emotion is frightened, then include but not limited to those words or word: fear, a word used for translation, alarmed, frighten, fear,
Frightened, timid, shrink, nervous, nervous, panic；Infer and draw passiveness
If emotion is for detesting, then include but not limited to those words or word: { disagreeable, annoying, nauseating, boring }, deduction
Draw passiveness.
Described routing unit 103, in order to according to the emotion analysis result in described user emotion analytic unit and institute
State the content-data in data structured unit, customize out travel schedule.Described routing unit 103 can be according to user emotion
Analyze the analysis result that module 102 provides, derive one based on after by data structured module 101 structuring for this user
Data, and the travel schedule customized according to time, place and scene.In order to realize such function, routing module
103 first pass through employing structural data generates a decision tree based on time, place and scene, then reality based on user
Border trip requirements (includes stroke purpose, preference and the anticipated of trip in trip place stay duration etc.), utilizes stroke customization to calculate
Method derives a travel schedule customized, and the most again the program is supplied to user.
Preferred as in the present embodiment, obtains content-data from WEB terminal and includes,
By accessing SNS server, data base searches the content-data in social networks, and rings in WEB terminal
Should；
And/or, by accessing content server, data base searches the content-data published or pay close attention to, and
Respond in WEB terminal.
In certain embodiments, user is by delivering disclosed in apps server requested database or user's concern
Content.
In certain embodiments, described in the content publishing or pay close attention to include but not limited to, the comment of microblogging@, micro-
The user comment delivered in the letter comment of circle of friends@and forum, user are for the scoring of tourist famous-city, tourist famous-city
Temperature ranking list etc..
In certain embodiments, the contents extraction paid close attention to is carried out according to the alexa website ranking of travel-related web sites.
Preferred as in the present embodiment, described data acquisition unit 101, also in order to carry out following pretreatment:
Unstructured data excavation, and structure data entry is carried out according to the stroke special topic of content-data；
According to natural language processing algorithm, described unstructured data is carried out pretreatment.
Described stroke special topic refers to, humane, shopping, seabeach, cuisines, self-driving, outdoor experience, natural, photography, performance joy
Happy, build label, history, parent-offspring, honeymoon, old man, romance, lovers etc..
Preferred as in the present embodiment, described data structured unit 101, also in order to,
According to the time in described tuple, place and scene, described unstructured data is changed；
According to semantic weight, the data after said structure are classified；
Generate stroke to be correlated with corpus prototype.
Preferred as in the present embodiment, described user emotion analytic unit 102, also in order to,
Utilize natural language emotion analytical technology that the emotion of user is analyzed extraction and obtain corresponding key word；
User is depicted for the tectonic information needed for the portrait of the preference of stroke according to described key word；
Extract from described tectonic information and represent the key word of user interest point and to user's use in stroke is experienced
Family emotion understanding.
Preferred as in the present embodiment, described routing unit 103, also in order to,
The trip scene settings given based on user goes out a stroke list, the stroke corresponding in order to show different scene；
When customizing travel schedule, increase instant data message, in order to carry out different weight couplings.
In certain embodiments, weight coupling is based on user's scoring and temperature.
In certain embodiments, increase instant data message to include big but be not limited to, weather, budget and mood of going on a tour
Deng.
In the present embodiment, by fully obtaining relevant content information from multiple Data Sources, and quickly analyze effectively
Focus trend, in conjunction with the psychology and behavior of consumption of user, help user immediately to process bulk information, being customized stroke
Recommend.After user's stroke commending system in the application of the invention, greatly save searching and produced user-specific customization
Change the time cost spent by travel schedule and energy, also greatly facilitate tourism and the leisure trip planning of user simultaneously.Number
Can collect, according to collecting unit 100, the unstructured data that tourism and leisure trip is relevant, data structured unit then can be by simultaneously
The data construction of condition based on one or more time, place and scene that data acquisition unit collection comes.Each user
Be intended to the trip requirements of oneself when trip, i.e. travel time, place and scene etc., can sufficiently be met.Described
User emotion analytic unit 102, can be analyzed the impression of oneself certain section of trip experience for user.Routing unit
103, by utilizing the stroke of 102 couples of users in user emotion analytic unit to experience in result and the data structured unit of analysis
The data handled well generate a customization stroke list for this user.Trip requirements planning trip according to user oneself
Scheme, saves substantial amounts of time and efforts.Native system has greatly been saved searching and has produced user-specific customization travel schedule
Spent time cost and energy, also greatly facilitate tourism and the leisure trip planning of user simultaneously.
Fig. 2 is the custom algorithm schematic flow sheet of travel schedule in routing unit in Fig. 1.
Preferred custom algorithm flow process in the present embodiment includes:
Step S200 passes through CFB unconfined condition algorithm, divides according to the evaluation in all places to go included in traffic path
The summation of number obtains the recommendation of route；
Described in step S201, recommendation computational methods are:
The total length of route described in step S202 by this route comprised all go between total sum of distance,
Step S203 is further according to the recommendation of the route of starting point to another place to go, and by the method time of above-mentioned recommendation
Go through all places to go in a place to go subset, then select the traffic path planning of optimum.
Specifically, unconfined condition (CFB) route developing algorithm is a kind of based on famous Dick Si Tela
(Dijkstra) improved algorithm of algorithm.Dijkstra algorithm is that the shortest path from a summit to remaining each summit is calculated
Method, solution is shortest route problem in directed graph.Accordingly, summit is the place to go information having effective evaluation mark, and connects
The limit on each summit, it is simply that the physical distance between each place to go.Therefore, CFB algorithm can first count in data acquisition unit 101 and obtain
Physical distance between all of going and the evaluation score in each place to go, then on the basis of this, calculate every potential road
The recommendation of line.In CFB algorithm, the recommendation of route is exactly the total of the evaluation score in all places to go that this route is comprised
With, then the total length divided by this route, namely this route comprised all go between the summation of distance.Then, CFB calculates
Method can find and can go over one according to the recommendation of the route of starting point to another place to go and comprise a number of place to go subset
In all places to go but route that total recommendation is maximum.
Fig. 3 is another custom algorithm schematic flow sheet of travel schedule in routing unit in Fig. 1.
In the present embodiment, another preferably custom algorithm flow process comprises the steps:
Step S301 passes through CBB constraints algorithm, the constraints provided according to user, carries out route planning；
Step S302 obtains the recommendation of route according to the summation of the evaluation score in all places to go included in traffic path
Value；
Step S303 is further according to the recommendation of the route of starting point to another place to go, and by the method time of above-mentioned recommendation
Go through all places to go in a place to go subset；If route is in constraints, then calculates and can meet described constraints
Good route planning, described constraints at least includes, the decision tree being made up of time, place and scene.In machine learning, institute
Stating decision tree is a forecast model, and what he represented is object properties (such as time, place and scene) and object value (distribution
Weight) between a kind of mapping relations.In tree, each node represents certain that certain object, each diverging paths then represent
Possible property value, each leaf node then corresponding from root node to the object represented by the path that this leaf node is experienced
Value.
Specifically, it is the route developing algorithm relative to CFB algorithm based on particular constraints condition (CBB) algorithm.CBB algorithm
The particular constraints condition that user can be proposed, as time and scene (the most such as: the time, 3 days, Qin Ziyou；5 days time, lovers
Romantic recreation trip；1 day time, the poor trip of donkey friend.), the most such as scene and (the most such as: Qin Ziyou, Disneyland, place；Feelings
Companion's romance recreation trip, coconut trees and rhythmic sea；The poor trip of donkey friend, Dabie Mountain；Red journey, the site of a meeting, Zun Yi).Take into account in route planning.This
Also it is that it is uniquely different from the place of unconfined condition (CFB) algorithm.In the core of algorithm, the operation principles of algorithm is with upper
The CFB algorithm stated is consistent.When user in client application by preference acquiring unit 102 by specific constraints,
Such as route expection cost time and budget, pass to CBB algoritic module (132) as parameter, and this algorithm can be at CFB algorithmic derivation
Screen further on the basis of the route gone out.It is to say, planning unit 104 can ensure that cooked up route not can exceed that
Expection cost time and budget, and the potential route planning that can not meet condition can be denied, and can meet this until calculating
The best route planning of condition.
Fig. 4 is the setting tuple structure schematic diagram of user in Fig. 1.
Preferred as in the present embodiment, user sets tuple and includes { time 400, place 401, scene 402}.Described
Time 400 includes that user's stroke start and end time, described place 401 include stroke departure place and destination, described scene
402 include user-defined travel type.
| Time 400 | Place 401 | Scene 402 |
| Vacation on May Day | Beijing and Tianjin Qinhuangdao | Parent-offspring swims |
| 11 vacations | Canberra, Melbourne, Sydney | Honeymoon |
| Spring Festival holiday | Europe ten state travels | Family forms a team |
| The vacation Dragon Boat Festival | Shanghai, Hangzhou, Anhui | Company group builds |
| Vacation in the mid-autumn | Sichuan, Yunnan | The poor trip of donkey friend |
Fig. 5 is that the stroke that customizes of the present invention recommends the schematic flow sheet in method one embodiment.
A kind of stroke that customizes in the present embodiment recommends method, comprises the following steps that
Step S500 obtains content-data from WEB terminal, and described content-data includes structured data or non-knot structure data；
In step S500, it is preferable that use the information data method for digging of special topicization to obtain content-data.In step S500, excellent
Selection of land, uses based on natural language processing algorithm, carries out semantic analysis after extracting text data again.Described natural language
The process model of Processing Algorithm includes but not limited to, Markov model and hidden Markov model, N meta-model, maximum entropy
Model, support vector machine etc..
Described structured data is directly deposited to data base by step S501, and according to the setting tuple of user described non-
After structured data carries out data structured process according to the dependency between key word and be stored in data base；In step
S501 carries out the standardization of data, such as based on { time, place, scene }；The described time includes that user's stroke starts
And the end time, described place includes that stroke departure place and destination, described scene include user-defined travel type.One
In a little embodiments, step S501 also needs to according to the time in described tuple, place and scene described unstructured data
Change；According to semantic weight, the data after said structure are classified；Generate stroke to be correlated with corpus prototype.
Step S502 infers the user emotion of correspondence according to the user's one section of experience in stroke, result is updated simultaneously
To data base；In step S502, infer that the user emotion of correspondence can be based on the association between key word or key word
Property.
Step S503 is according to the emotion analysis result in described user emotion analytic unit and described data structured unit
In content-data, customize out travel schedule.In step S503, utilize the natural language emotion analytical technology emotion to user
It is analyzed extraction and obtains corresponding key word；According to described key word depict user for the portrait of the preference of stroke needed for
Tectonic information；Extract from described tectonic information and represent the key word of user interest point and to user in stroke is experienced
User emotion recognizes.The step customizing out travel schedule includes, the trip scene settings given based on user goes out a stroke list,
The stroke corresponding in order to show different scene；When customizing travel schedule, increase instant data message, in order to carry out different power
Heavily mate.
The stroke that customizes in the present embodiment recommends method fully to obtain relevant content information from multiple Data Sources, and
Quickly analyze effective focus trend, in conjunction with the psychology and behavior of consumption of user, help user immediately to process bulk information,
Being customized stroke is recommended.
Those of ordinary skill in the field it is understood that more than, described be only the present invention specific embodiment, and
Be not used in the restriction present invention, all within the spirit and principles in the present invention, any modification, equivalent substitution and improvement etc. done, all
Within protection scope of the present invention should being included in.
Claims (10)
1. customize stroke commending system, it is characterised in that including: data acquisition unit, data structured unit, user emotion
Analytic unit and routing unit,
Described data acquisition unit, in order to obtain content-data from WEB terminal, described content-data includes structural data or non-
Structural data；
Described data structured unit, in order to described structured data is directly deposited to data base, and the setting according to user
After tuple carries out data structured process according to the dependency between key word in described non-structural data and be stored in data
In storehouse；
Described user emotion analytic unit, in order to infer the user emotion of correspondence according to the user's one section of experience in stroke,
Result is updated to data base simultaneously；
Described routing unit, in order to tie according to the emotion analysis result in described user emotion analytic unit and described data
Content-data in structure unit, customizes out travel schedule.
Customization stroke commending system the most according to claim 1, it is characterised in that described user set tuple as:
{ time, place, scene }
The described time includes user's stroke start and end time, and described place includes stroke departure place and destination, described field
Scape includes user-defined travel type.
Customization stroke commending system the most according to claim 2, it is characterised in that described routing unit, in order to
The decision tree being made up of time, place and scene in described data structured unit is set according to the trip of user, obtains
Travel schedule.
Customization stroke commending system the most according to claim 3, it is characterised in that obtain described travel schedule according to such as
Down stroke custom algorithm:
By CFB unconfined condition algorithm, obtain according to the summation of the evaluation score in all places to go included in traffic path
The recommendation of route；
Described recommendation computational methods are:
The total length of described route by this route comprised all go between total sum of distance,
Further according to the recommendation of the route of starting point to another place to go, and go virgin by the method traversal one of above-mentioned recommendation
Concentrate all places to go, then select the traffic path planning of optimum；
Or,
By CBB constraints algorithm, the constraints provided according to user, carry out route planning；
The summation of the evaluation score according to all places to go included in traffic path obtains the recommendation of route；
Further according to the recommendation of the route of starting point to another place to go, and go virgin by the method traversal one of above-mentioned recommendation
Concentrate all places to go；If route is in constraints, then calculate the best route planning that can meet described constraints, institute
State constraints at least to include, the decision tree being made up of time, place and scene.
Customization stroke commending system the most according to claim 1, it is characterised in that obtain content-data bag from WEB terminal
Include,
By accessing SNS server, data base searches the content-data in social networks, and responds in WEB terminal；
And/or, by accessing content server, data base searches the content-data published or pay close attention to, and at WEB
End responds.
Customization stroke commending system the most according to claim 1, it is characterised in that described data acquisition unit, also uses
To carry out following pretreatment:
6-1) carry out unstructured data excavation, and structure data entry according to the stroke special topic of content-data；
6-2) according to natural language processing algorithm, described unstructured data is carried out pretreatment.
Customization stroke commending system the most according to claim 2, it is characterised in that described data structured unit, also
In order to,
7-1) according to the time in described tuple, place and scene, described unstructured data is changed；
7-2) according to semantic weight, the data after said structure are classified；
7-3) generate stroke to be correlated with corpus prototype.
Customization stroke commending system the most according to claim 1, it is characterised in that described user emotion analytic unit,
Also in order to,
8-1) utilize natural language emotion analytical technology that the emotion of user is analyzed extraction and obtain corresponding key word；
8-2) depict user for the tectonic information needed for the portrait of the preference of stroke according to described key word；
8-3) extract from described tectonic information represent user interest point key word and to user stroke experience in use
Family emotion understanding.
Customization stroke commending system the most according to claim 2, it is characterised in that described routing unit, also uses
With,
9-1) the trip scene settings given based on user goes out a stroke list, the stroke corresponding in order to show different scene；
9-2) when customizing travel schedule, increase instant data message, in order to carry out different weight couplings.
10. one kind customizes stroke recommendation method, it is characterised in that including:
Obtaining content-data from WEB terminal, described content-data includes structured data or non-knot structure data；
Described structured data is directly deposited to data base, and press in described non-structural data according to the setting tuple of user
After carrying out data structured process according to the dependency between key word and be stored in data base；
Infer the user emotion of correspondence according to the user's one section of experience in stroke, result is updated to data base simultaneously；
According to the content-data in the emotion analysis result in described user emotion analytic unit and described data structured unit,
Customize out travel schedule.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610499550.1A CN106203646A (en) | 2016-06-29 | 2016-06-29 | Customize stroke commending system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610499550.1A CN106203646A (en) | 2016-06-29 | 2016-06-29 | Customize stroke commending system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106203646A true CN106203646A (en) | 2016-12-07 |
Family
ID=57462800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610499550.1A Pending CN106203646A (en) | 2016-06-29 | 2016-06-29 | Customize stroke commending system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106203646A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107463698A (en) * | 2017-08-15 | 2017-12-12 | 北京百度网讯科技有限公司 | Method and apparatus based on artificial intelligence pushed information |
| CN108710969A (en) * | 2018-04-23 | 2018-10-26 | 华南师范大学 | A kind of intelligent paths planning method of tourism |
| CN110288107A (en) * | 2019-06-04 | 2019-09-27 | 北京汽车股份有限公司 | The method and system of more new trip, car-mounted terminal, mobile terminal and cloud platform |
| CN110516922A (en) * | 2019-08-06 | 2019-11-29 | 北京创鑫旅程网络技术有限公司 | Distribute the method and device of data processing object |
| CN110929178A (en) * | 2018-09-20 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Information recommendation method and device, terminal device and server |
| CN110991651A (en) * | 2019-11-30 | 2020-04-10 | 航天科技控股集团股份有限公司 | Energy consumption prediction analysis system and method for user driving habits based on TBOX |
| CN112148985A (en) * | 2020-10-09 | 2020-12-29 | 苏州淘喜网络科技有限公司 | Intelligent recommendation system and method for urban scenic spot |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150145727A1 (en) * | 2010-10-08 | 2015-05-28 | HJ Laboratories, LLC | Tracking a mobile computer indoors using wi-fi and motion sensor information |
| CN104978700A (en) * | 2015-06-29 | 2015-10-14 | 昆明理工大学 | Real-time recommendation method for self-help travel of old people in tourism website |
| CN105547306A (en) * | 2015-08-11 | 2016-05-04 | 深圳大学 | Route pushing method and system thereof |
| CN105610884A (en) * | 2014-11-21 | 2016-05-25 | 阿里巴巴集团控股有限公司 | Method and device for providing travel information |
-
2016
- 2016-06-29 CN CN201610499550.1A patent/CN106203646A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150145727A1 (en) * | 2010-10-08 | 2015-05-28 | HJ Laboratories, LLC | Tracking a mobile computer indoors using wi-fi and motion sensor information |
| CN105610884A (en) * | 2014-11-21 | 2016-05-25 | 阿里巴巴集团控股有限公司 | Method and device for providing travel information |
| CN104978700A (en) * | 2015-06-29 | 2015-10-14 | 昆明理工大学 | Real-time recommendation method for self-help travel of old people in tourism website |
| CN105547306A (en) * | 2015-08-11 | 2016-05-04 | 深圳大学 | Route pushing method and system thereof |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107463698A (en) * | 2017-08-15 | 2017-12-12 | 北京百度网讯科技有限公司 | Method and apparatus based on artificial intelligence pushed information |
| CN108710969A (en) * | 2018-04-23 | 2018-10-26 | 华南师范大学 | A kind of intelligent paths planning method of tourism |
| CN110929178A (en) * | 2018-09-20 | 2020-03-27 | 阿里巴巴集团控股有限公司 | Information recommendation method and device, terminal device and server |
| CN110288107A (en) * | 2019-06-04 | 2019-09-27 | 北京汽车股份有限公司 | The method and system of more new trip, car-mounted terminal, mobile terminal and cloud platform |
| CN110516922A (en) * | 2019-08-06 | 2019-11-29 | 北京创鑫旅程网络技术有限公司 | Distribute the method and device of data processing object |
| CN110516922B (en) * | 2019-08-06 | 2023-07-04 | 北京创鑫旅程网络技术有限公司 | Method and device for distributing data processing objects |
| CN110991651A (en) * | 2019-11-30 | 2020-04-10 | 航天科技控股集团股份有限公司 | Energy consumption prediction analysis system and method for user driving habits based on TBOX |
| CN110991651B (en) * | 2019-11-30 | 2023-04-28 | 航天科技控股集团股份有限公司 | Energy consumption predictive analysis system and method for user driving habit based on TBOX |
| CN112148985A (en) * | 2020-10-09 | 2020-12-29 | 苏州淘喜网络科技有限公司 | Intelligent recommendation system and method for urban scenic spot |
| CN113515693A (en) * | 2020-10-09 | 2021-10-19 | 苏州淘喜网络科技有限公司 | City scenic spot wisdom recommendation system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106203646A (en) | Customize stroke commending system and method | |
| Deng et al. | Different cultures, different photos: A comparison of Shanghai's pictorial destination image between East and West | |
| Park et al. | A study on topic models using LDA and Word2Vec in travel route recommendation: focus on convergence travel and tours reviews | |
| US20110238608A1 (en) | Method and apparatus for providing personalized information resource recommendation based on group behaviors | |
| CN105740366A (en) | Inference method and device of MicroBlog user interests | |
| Xu et al. | A dynamic topic model and matrix factorization-based travel recommendation method exploiting ubiquitous data | |
| CN104598588A (en) | Automatic generating algorithm of microblog user label based on biclustering | |
| Lu et al. | Travel attractions recommendation with knowledge graphs | |
| CN106202252A (en) | Method, system are recommended in a kind of trip analyzed based on user emotion | |
| Zahálka et al. | Interactive multimodal learning for venue recommendation | |
| CN109284443A (en) | A kind of tourism recommended method and system based on crawler technology | |
| Liu et al. | Recommending attractive thematic regions by semantic community detection with multi-sourced VGI data | |
| Barreneche | The order of places: Code, ontology and visibility in locative media | |
| Rozenkiewicz et al. | Geotourism and the 21st century–NTOs’ website information availability on geotourism resources in selected central European countries: international perspective | |
| Mukhina et al. | Orienteering problem with functional profits for multi-source dynamic path construction | |
| CN105678383A (en) | Mobile knowledge service system on the basis of ontology model | |
| Ma et al. | How do uses of and gratifications from social media platforms drive responsible birdwatching behavior? | |
| Havlícek et al. | Agritourism farms-evaluation of their websites quality and Web 2.0 | |
| Gao et al. | Towards travel recommendation interpretability: Disentangling tourist decision-making process via knowledge graph | |
| Haris et al. | Mining graphs from travel blogs: a review in the context of tour planning | |
| CN104794200B (en) | A kind of event distribution subscription method of the support fuzzy matching based on body | |
| Nelson | Curating Object-Oriented Collections Using the TEI | |
| Ruan et al. | Exploring emotion differences in tourist attractions based on online travel notes: a case study in Nanjing, China | |
| Wei et al. | Visual representation of a linear tourist destination based on social network photos: a comparative analysis of cross-cultural perspectives | |
| CN106570130B (en) | Text region judgment method and its system based on RDF knowledge base |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161207 |