CN113515697A - Group dynamic tour route recommendation method and system based on multiple intentions of user - Google Patents
Group dynamic tour route recommendation method and system based on multiple intentions of user Download PDFInfo
- Publication number
- CN113515697A CN113515697A CN202110589140.7A CN202110589140A CN113515697A CN 113515697 A CN113515697 A CN 113515697A CN 202110589140 A CN202110589140 A CN 202110589140A CN 113515697 A CN113515697 A CN 113515697A
- Authority
- CN
- China
- Prior art keywords
- user
- representation
- group
- preference
- constructing
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000004458 analytical method Methods 0.000 claims abstract description 19
- 238000012512 characterization method Methods 0.000 claims abstract description 10
- 238000012216 screening Methods 0.000 claims abstract description 9
- 230000004927 fusion Effects 0.000 claims description 14
- 238000013135 deep learning Methods 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 6
- 230000006399 behavior Effects 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 230000002776 aggregation Effects 0.000 claims description 4
- 238000004220 aggregation Methods 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 230000004308 accommodation Effects 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
-
- 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
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/14—Travel agencies
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Economics (AREA)
- Molecular Biology (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- Human Resources & Organizations (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Fuzzy Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Navigation (AREA)
Abstract
The invention provides a group dynamic tour route recommendation method and system based on multi-intentions of users, wherein the method comprises the following steps: constructing a user representation based on multi-view characterization learning, the user representation comprising a user base representation and a content representation, establishing a user preference representation based on the user base representation and the content representation; constructing a group dynamic preference model according to the difference of individual influences, and meeting the preference requirements of group users to the maximum extent; and screening and recommending the travel routes obtained by deep clustering analysis based on the multi-intention preference and the group preference of the user. By the scheme, accurate recommendation of the tour routes of the group users can be realized, user preference and group preference are sufficiently mined, and tour experience in accompanying the tour is improved.
Description
Technical Field
The invention relates to the field of tour route recommendation, in particular to a group dynamic tour route recommendation method and system based on multi-intentions of users.
Background
With the gradual improvement of the living standard of the public, tourism has become extremely popular as an important activity for people to enjoy daily leisure. Abundant tourism data provides massive multi-source information, and also brings the information overload problem, causes the visitor to need to spend a large amount of time and energy when formulating the tourism route simultaneously. Travel route recommendation is an important way for alleviating the problem, and has become a hot direction in the field of current travel data mining, so that meeting personalized preference requirements of different tourists and improving travel experience of the tourists are targets of intelligent travel route recommendation.
The method mainly considers the independent travel scene facing individual users, generally adopts the traditional static search matching mode, firstly obtains the access records of the users, then searches and matches the access records in the existing travel route database according to the similarity or obtains the optimal route by using optimization algorithms such as greedy algorithm and the like, and expands the basic route by using cluster sorting algorithm, thereby generating the recommendation result according with the preference of the users. Due to the gradual increase of multi-source tourism information and the generation of the data sparsity problem, the traditional method has the disadvantages of unreliable high-dimensional feature modeling, unsatisfactory characterization effect, low intelligentization level, poor recommendation performance and serious reduction of the tourism experience of a user. Moreover, most of the existing methods focus on single preference of the user to scenic spots, do not fully consider the multi-intention tourism demands of the user, and are deficient in research on group tourism route recommendation.
Therefore, aiming at the problems that the current travel route only considers the personal travel scene and the individual travel recommendation algorithm, the group dynamic travel route recommendation method and system based on the multi-intention of the user are provided.
Disclosure of Invention
In view of this, embodiments of the present invention provide a group dynamic travel route recommendation method and system based on multiple intentions of a user, so as to solve the problem that the existing travel route recommendation only considers the travel of an individual user.
In a first aspect of the embodiments of the present invention, a group dynamic travel route recommendation method based on multiple intentions of a user is provided, including:
constructing a user representation based on multi-view characterization learning, the user representation comprising a user base representation and a content representation, establishing a user preference representation based on the user base representation and the content representation;
constructing a group dynamic preference model according to the difference of individual influences, and meeting the preference requirements of group users to the maximum extent;
and screening and recommending the travel routes obtained by deep clustering analysis based on the multi-intention preference and the group preference of the user.
In a second aspect of the embodiments of the present invention, there is provided a group dynamic travel route recommendation system based on multi-intentions of users, including:
a user representation construction module for constructing a user representation based on multi-view representation learning, the user representation comprising a user base representation and a content representation, the user preference representation being established based on the user base representation and the content representation;
the group preference representation module is used for constructing a group dynamic preference model according to the difference of individual influences, and maximally meeting the preference requirements of group users;
and the tourism route recommendation module is used for screening and recommending the tourism routes obtained by deep clustering analysis based on the multi-intention preference and the group preference of the user.
In a third aspect of the embodiments of the present invention, there is provided an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the steps of the method according to the first aspect of the embodiments of the present invention.
In a fourth aspect of the embodiments of the present invention, a computer-readable storage medium is provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the method provided in the first aspect of the embodiments of the present invention.
In the embodiment of the invention, the user preference representation is realized through multi-view representation learning, and dynamic group preference modeling is constructed by combining a deep learning technology to dynamically learn the influence of different individuals in groups based on the influence difference of the same individual in different groups; considering a plurality of intentions of a user when selecting a tour route, filtering and screening of the tour route are realized through deep clustering analysis, so that a candidate set is generated, and a more accurate and effective tour route recommendation result is generated through scoring and predicting in the candidate set. Therefore, the problem that the traditional travel route recommendation method is only single in an object-oriented mode can be solved, multiple intention requirements of a user in travel are fully considered, accurate personalized group travel route recommendation is achieved, and travel experience of tourists in the accompanying travel is improved. Reliable and effective user preference representation is realized by utilizing multi-view representation fusion and deep learning technology, different influences of group members are fully excavated, and the problems of object-oriented simplification, insufficient information utilization, poor representation learning capability and the like in the current travel route recommendation can be effectively solved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flowchart illustrating a group dynamic travel route recommendation method based on user multi-intent according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of building a group dynamic preference model according to an embodiment of the present invention;
FIG. 3 is a flow diagram illustrating a travel route recommendation process according to one embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a group dynamic travel route recommendation system based on multi-intent of a user according to an embodiment of the present invention.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons skilled in the art without any inventive work shall fall within the protection scope of the present invention, and the principle and features of the present invention shall be described below with reference to the accompanying drawings.
The terms "comprises" and "comprising," when used in this specification and claims, and in the accompanying drawings and figures, are intended to cover non-exclusive inclusions, such that a process, method or system, or apparatus that comprises a list of steps or elements is not limited to the listed steps or elements.
Referring to fig. 1, fig. 1 is a schematic flow chart of a group dynamic travel route recommendation method based on multiple intentions of a user according to an embodiment of the present invention, including:
s101, constructing a user portrait based on multi-view characterization learning, wherein the user portrait comprises a user basic portrait and a content portrait, and establishing user preference representation based on the user basic portrait and the content portrait;
the multi-view representation learning is to perform feature description on a certain user from multiple angles and perform feature learning, and the multi-view representation of the user may include user basic features and behavior features. The data high-dimensional characteristics can be obtained through multi-view representation, and the data representation effect is improved. The user portrait is a data representation mode based on user characteristics, and different characteristic sets can be adopted for representing different view angles.
Wherein, the basic portrait of the user is constructed according to the average embedding of the gender, age and geographic information of the user; and constructing the user content portrait according to the embedding of the user high-frequency tags and the categories extracted from the user behaviors. Embedding refers to representing the characteristic information in an embedded form.
Based on a user portrait construction model, a user portrait can be constructed through multi-view representation learning, the user portrait contains rich user information, and the user portrait construction model mainly comprises the characteristics of two views, namely a user Basic portrait (Basic profile) and a Content profile (Content profile). Wherein, the user basic portrait construction is average embedding of gender, age and geographic information; and constructing a user content portrait according to the high-frequency tags and category embedding of the user extracted from the user behaviors. Thus, the characteristic information of the user basic image and the content image can be obtained.
User preferences may be represented by multi-view representation fusion. View representation fusion utilizes complementary knowledge contained in multiple views to comprehensively represent user preferences.
Specifically, based on the user basic portrait and the content portrait, the user preference is expressed through multi-view representation fusion, wherein the multi-view representation fusion formula is as follows:
wherein h is the fused user preference representation, x1,x2Respectively a user base representation and a user content representation,
the mode of multi-view representation fusion can be Sum (Sum), maximum (Max) or cascade (coordination) fusion.
S102, constructing a group dynamic preference model according to differences of individual influences, and meeting preference requirements of group users to the maximum extent;
in the group, influence of different users in different groups can be distinguished according to travel records of different individuals, travel experience differences of specific scenic spots and the like, if a member K in one group has richer travel experience for scenic spots of scenic spots and historical sites relative to other members, influence of the member K in the group is higher than that of other members to a certain extent in travel route planning of scenic spots and historical sites, so that contribution of the member K in group decision is improved, and overall travel experience and satisfaction of the group can be improved as much as possible.
The group dynamic preference model is used for constructing the group preference model according to the influence of interests (preferences) of different users in a group, and the model needs to meet the preference requirements of the group users to the maximum extent, namely, the optimization goal of meeting the maximum extent by the preference of the group users is taken as the optimization goal, and the satisfaction (or preference requirements) of all the users in the group is guaranteed to be higher.
In one embodiment, as shown in fig. 2, the group dynamic preference model is constructed by the following steps:
distinguishing influences of different users in different groups according to travel records of the different users and the difference of travel experiences of scenic spots;
dynamically learning influence weights of different users in different groups based on a deep learning technology;
assigning a corresponding influence weight to each group member, and representing the group members with the influence weights in an aggregation manner, wherein the specific formula is as follows:
in the formula, λinfFor the impact weights, G, obtained by the deep neural networkmIs the m-th group, u, of usersjDenotes the jth user, gmIs the m-th group preference representation after aggregation.
Compared with the traditional weight assignment method based on a statistical method, the method adopts a deep learning technology to dynamically learn the influence weights of different users in different groups, and avoids the problems of poor flexibility, low accuracy and the like of the traditional method. In this way, each group member is given a certain influence weight, and then the group members with the additional influence weights are collectively represented
S103, screening and recommending the travel routes obtained through deep clustering analysis based on the multi-intention preference and the group preference of the user.
According to the user preference representation in the step S101 and the group preference representation in the step S102, interactive learning is respectively carried out on the user preference representation and the scenic spot representations by utilizing a deep learning technology, linear interaction and nonlinear interaction relations are mined to the maximum extent, implicit features are better learned, and the problems of cold start and data sparsity can be solved to a certain extent.
The deep clustering analysis is combined with deep learning and clustering analysis methods to learn and iteratively cluster the characteristics, and finally the travel route set can be obtained.
And generating a travel route set by utilizing deep clustering analysis in the acquired travel routes, namely mapping the attribute information of the travel routes to a low-dimensional-strong semantic space by utilizing a deep clustering analysis technology, and further dividing the low-dimensional-strong semantic space to generate the travel route set. Then, considering a plurality of intention information of the user, constructing an initial expression of the intention characteristics of the user, namely considering the intention requirements of the user in six aspects of catering (eating), accommodation (living), traffic (walking), tourism (traveling), shopping (buying) and entertainment (entertainment), and performing characterization learning on the multiple intentions of the group user by utilizing deep learning; and generating a filtered travel route set according to the multi-intention characterization of the user and by utilizing sequencing learning. And finally, calculating the score of each scenic spot in the filtered travel routes through a scoring function, and taking Top-N travel routes according to the score as the recommendation result of the group travel route.
Specifically, in one embodiment, as shown in FIG. 3, the travel route recommendation process based on user preferences and group preferences includes:
s301, obtaining tour route information;
the existing travel route information is obtained from a system database or network.
S302, generating a tour route set based on deep clustering analysis;
s303, filtering the tour route based on the multi-intentions of the user;
constructing representation of user intention characteristics based on the user multi-intention information, and filtering the tour route according to the user multi-intention;
s304, calculating the score of each sight spot in the route based on the group preference;
based on the group preferences, value scores for each sight spot in the tour route are calculated.
Wherein, the value score of each sight spot is calculated according to the following formula:
fg(gm,li)＝gm·li；
in the formula (f)gAs a scoring function, representing a value score of the attraction, gmRepresenting aggregated group preferences,/iDenote sights and denote multiplications.
S305, taking Top-N routes to generate a recommendation result.
And taking a certain number of tour routes ranked at the top as group recommended routes according to the size of the scene point values in the tour routes. After the score of each scenic spot in the tour route is calculated, a certain number of tour routes can be selected for recommendation according to the total score or average score of the tour routes.
In the embodiment, a plurality of feature information in two views of a user basic portrait and a content portrait are fully considered, reliable and effective user preference representation is realized by utilizing multi-view representation fusion and a deep learning technology, and the problems of unreliable feature modeling, weak representation capability, incomplete information utilization and the like of the traditional method are solved; meanwhile, different influences of the deep neural network on the group members are mined, the contribution degree of the group members to the group decision is fully considered, and the accuracy of dynamic group preference modeling is improved. The group dynamic travel route recommendation is realized based on multiple intentions of the user, the problem of object-oriented simplification of the traditional travel route recommendation method can be solved, multiple intention requirements of the user in travel are fully considered, and the accurate personalized travel route recommendation is facilitated to be realized.
The intelligent travel route recommendation method based on the deep learning technology can map high-dimensional sparse features into low-dimensional dense features through the deep learning technology, and effectively solves the problems of unsatisfactory characterization modeling, insufficient information utilization and dynamic preference change of the traditional method.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
FIG. 4 is a schematic structural diagram of a group dynamic travel route recommendation system based on multi-intentions of a user according to an embodiment of the present invention, where the system includes:
a user representation construction module 410 for constructing a user representation based on multi-view representation learning, the user representation comprising a user base representation and a content representation, a user preference representation being established based on the user base representation and the content representation;
wherein, the basic portrait of the user is constructed according to the average embedding of the gender, age and geographic information of the user; and constructing the user content portrait according to the embedding of the user high-frequency tags and the categories extracted from the user behaviors.
Specifically, the establishing of the user preference representation based on the user basic portrait and the content portrait is specifically as follows:
representing user preferences through multi-view representation fusion based on a user base portrait and a content portrait, wherein the multi-view representation fusion formula is as follows:
wherein h is the fused user preference representation, x1,x2Respectively a user base representation and a user content representation,
is a way of multi-view characterization fusion.
The group preference representation module 420 is used for constructing a group dynamic preference model according to the difference of individual influences, so that the preference requirements of group users are met to the maximum extent;
optionally, the group preference representing module 420 includes:
the influence expression unit is used for distinguishing the influence of different users in different groups according to travel records of different users and the difference of travel experiences of scenic spots;
the dynamic learning unit is used for dynamically learning the influence weights of different users in different groups based on a deep learning technology;
the aggregation expression unit is used for giving corresponding influence weight to each group member and aggregating and expressing the group members with the influence weight, and the concrete formula is as follows:
in the formula, λinfFor the impact weights, G, obtained by the deep neural networkmIs a group of users, ujDenotes the jth user, gmIs an aggregated group preference representation.
And the travel route recommending module 430 is used for screening and recommending travel routes obtained by deep clustering analysis based on the user multi-intention preference and the group preference.
Specifically, the travel route recommending module 430 comprises:
the deep clustering analysis unit is used for acquiring the tour route information and generating a tour route set through deep clustering analysis;
the route filtering unit is used for constructing representation of user intention characteristics based on the user multi-intention information and filtering the tour route according to the user multi-intention;
the score calculating unit is used for calculating the value scores of all the scenic spots in the tour route based on the group preference;
and the route recommending unit is used for taking the tourism routes with a certain number of top ranks as the group recommended routes according to the size of the scene point values in the tourism routes.
The calculating of the value scores of the scenic spots in the tour route based on the group preference specifically comprises the following steps:
the value score for each sight is calculated according to the following formula:
fg(gm,li)＝gm·li；
in the formula (f)gAs a scoring function, representing a value score of the attraction, gmRepresenting aggregated group preferences,/iIndicating a sight.
It will be appreciated that in one embodiment, the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor implementing steps S101-S103 when executing the computer program to implement group travel route recommendations.
It will be understood by those skilled in the art that all or part of the steps in the method for implementing the above embodiments may be implemented by instructing the relevant hardware through a program, where the program may be stored in a computer-readable storage medium, and when executed, the program includes steps S101 to S103, and the storage medium includes, for example, ROM/RAM.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A group dynamic travel route recommendation method based on multi-intentions of users is characterized by comprising the following steps:
constructing a user representation based on multi-view characterization learning, the user representation comprising a user base representation and a content representation, establishing a user preference representation based on the user base representation and the content representation;
constructing a group dynamic preference model according to the difference of individual influences, and meeting the preference requirements of group users to the maximum extent;
and screening and recommending the travel routes obtained by deep clustering analysis based on the multi-intention preference and the group preference of the user.
2. The method of claim 1, wherein constructing a user representation based on multi-view representation learning comprises:
constructing a user basic portrait according to average embedding of user gender, age and geographic information;
and constructing the user content portrait according to the embedding of the user high-frequency tags and the categories extracted from the user behaviors.
3. The method of claim 1, wherein establishing a user preference representation based on the user base representation and the content representation is specifically:
representing user preferences through multi-view representation fusion based on a user base portrait and a content portrait, wherein the multi-view representation fusion formula is as follows:
wherein h is the fused user preference representation, x1,x2Respectively a user base representation and a user content representation,
is a way of multi-view characterization fusion.
4. The method of claim 1, wherein the building of the group dynamic preference model according to the differences in individual influence maximizes satisfaction of the preference needs of the group users comprises:
distinguishing influences of different users in different groups according to travel records of the different users and the difference of travel experiences of scenic spots;
dynamically learning influence weights of different users in different groups based on a deep learning technology;
assigning a corresponding influence weight to each group member, and representing the group members with the influence weights in an aggregation manner, wherein the specific formula is as follows:
in the formula, λinfFor the impact weights, G, obtained by the deep neural networkmIs a group of users, ujDenotes the jth user, gmIs an aggregated group preference representation.
5. The method according to claim 1, wherein the screening and recommending of the travel routes obtained by the deep clustering analysis based on the user multi-intent preference and the group preference specifically comprises:
obtaining tour route information, and generating a tour route set through deep clustering analysis;
constructing representation of user intention characteristics based on the user multi-intention information, and filtering the tour route according to the user multi-intention;
calculating value scores of all scenic spots in the tour route based on the group preference;
and taking a certain number of tour routes ranked at the top as group recommended routes according to the size of the scene point values in the tour routes.
6. The method of claim 5, wherein calculating the value score for each sight point in the travel route based on the group preference is specifically:
the value score for each sight is calculated according to the following formula:
fg(gm,li)＝gm·li；
in the formula (f)gAs a scoring function, representing a value score of the attraction, gmRepresenting aggregated group preferences,/iIndicating a sight.
7. A group dynamic travel route recommendation system based on user multi-intent, comprising:
a user representation construction module for constructing a user representation based on multi-view representation learning, the user representation comprising a user base representation and a content representation, the user preference representation being established based on the user base representation and the content representation;
the group preference representation module is used for constructing a group dynamic preference model according to the difference of individual influences, and maximally meeting the preference requirements of group users;
and the tourism route recommendation module is used for screening and recommending the tourism routes obtained by deep clustering analysis based on the multi-intention preference and the group preference of the user.
8. The system of claim 7, wherein the travel route recommendation module comprises:
the deep clustering analysis unit is used for acquiring the tour route information and generating a tour route set through deep clustering analysis;
the route filtering unit is used for constructing representation of user intention characteristics based on the user multi-intention information and filtering the tour route according to the user multi-intention;
the score calculating unit is used for calculating the value scores of all the scenic spots in the tour route based on the group preference;
and the route recommending unit is used for taking the tourism routes with a certain number of top ranks as the group recommended routes according to the size of the scene point values in the tourism routes.
9. An electronic device comprising a processor, a memory, and a computer program stored in and executed on the memory, wherein the processor when executing the computer program implements the steps of the group dynamic travel route recommendation method based on user multi-intent according to any of claims 1-6.
10. A computer-readable storage medium, having a computer program stored thereon, wherein the computer program, when being executed by a processor, implements the steps of the group dynamic travel route recommendation method based on user multi-intent according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110589140.7A CN113515697A (en) | 2021-05-27 | 2021-05-27 | Group dynamic tour route recommendation method and system based on multiple intentions of user |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110589140.7A CN113515697A (en) | 2021-05-27 | 2021-05-27 | Group dynamic tour route recommendation method and system based on multiple intentions of user |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113515697A true CN113515697A (en) | 2021-10-19 |
Family
ID=78065099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110589140.7A Pending CN113515697A (en) | 2021-05-27 | 2021-05-27 | Group dynamic tour route recommendation method and system based on multiple intentions of user |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113515697A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110378798A (en) * | 2019-06-03 | 2019-10-25 | 中移互联网有限公司 | Isomery social networks construction method, group recommending method, device and equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150019469A1 (en) * | 2012-03-08 | 2015-01-15 | Thomson Licensing | Method of recommending items to a group of users |
| CN109948055A (en) * | 2019-03-19 | 2019-06-28 | 中森云链(成都)科技有限责任公司 | A kind of recommended method based on group's preference |
| CN110263256A (en) * | 2019-06-21 | 2019-09-20 | 西安电子科技大学 | Personalized recommendation method based on multi-modal heterogeneous information |
| CN110502704A (en) * | 2019-08-12 | 2019-11-26 | 山东师范大学 | A kind of group recommending method and system based on attention mechanism |
| CN111445308A (en) * | 2020-03-26 | 2020-07-24 | 四川旅游学院 | Tourist attraction recommendation method based on user portrait |
| CN112149010A (en) * | 2020-11-01 | 2020-12-29 | 云境商务智能研究院南京有限公司 | Group tour route recommendation method based on attention mechanism |
| CN112307332A (en) * | 2020-10-16 | 2021-02-02 | 四川大学 | Collaborative filtering recommendation method and system based on user portrait clustering and storage medium |
-
2021
- 2021-05-27 CN CN202110589140.7A patent/CN113515697A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150019469A1 (en) * | 2012-03-08 | 2015-01-15 | Thomson Licensing | Method of recommending items to a group of users |
| CN109948055A (en) * | 2019-03-19 | 2019-06-28 | 中森云链(成都)科技有限责任公司 | A kind of recommended method based on group's preference |
| CN110263256A (en) * | 2019-06-21 | 2019-09-20 | 西安电子科技大学 | Personalized recommendation method based on multi-modal heterogeneous information |
| CN110502704A (en) * | 2019-08-12 | 2019-11-26 | 山东师范大学 | A kind of group recommending method and system based on attention mechanism |
| CN111445308A (en) * | 2020-03-26 | 2020-07-24 | 四川旅游学院 | Tourist attraction recommendation method based on user portrait |
| CN112307332A (en) * | 2020-10-16 | 2021-02-02 | 四川大学 | Collaborative filtering recommendation method and system based on user portrait clustering and storage medium |
| CN112149010A (en) * | 2020-11-01 | 2020-12-29 | 云境商务智能研究院南京有限公司 | Group tour route recommendation method based on attention mechanism |
Non-Patent Citations (5)
| Title |
|---|
| PEIPEI WANG: "Socially-driven multi-interaction attentive group representation learning for group recommendation", 《PATTERN RECOGNITION LETTERS》 * |
| PEIPEI WANG等: "Socially-driven multi-interaction attentive group representation learning for group recommendation" * |
| 夏立新;杨金庆;程秀峰;: "移动环境下融合情境信息的群组推荐模型研究――基于用户APP行为数据的实证分析" * |
| 李方文: "基于用户生成空间数据的群组旅游推存研究", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 胡川;孟祥武;张玉洁;杜雨露;: "一种改进的偏好融合组推荐方法" * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110378798A (en) * | 2019-06-03 | 2019-10-25 | 中移互联网有限公司 | Isomery social networks construction method, group recommending method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107729444B (en) | Knowledge graph-based personalized tourist attraction recommendation method | |
| CN110555112B (en) | Interest point recommendation method based on user positive and negative preference learning | |
| US8577962B2 (en) | Server apparatus, client apparatus, content recommendation method, and program | |
| CN110795571B (en) | Cultural travel resource recommendation method based on deep learning and knowledge graph | |
| Subramaniyaswamy et al. | Intelligent travel recommendation system by mining attributes from community contributed photos | |
| CN112149010B (en) | Group travel route recommendation method based on attention mechanism | |
| CN107220365A (en) | Accurate commending system and method based on collaborative filtering and correlation rule parallel processing | |
| CN109947987B (en) | Cross collaborative filtering recommendation method | |
| CN113343127B (en) | Tourism route recommendation method, system, server and storage medium | |
| KR101543780B1 (en) | System and method for expert search by dynamic profile and social network reliability | |
| CN110276008A (en) | A kind of recommending scenery spot method and device based on user's travelling decision process | |
| CN110889450B (en) | Super-parameter tuning and model construction method and device | |
| CN106776701B (en) | Problem determination method and device for item recommendation | |
| CN106021456B (en) | Interest point recommendation method fusing text and geographic information in local collaborative arrangement | |
| CN113806630B (en) | Attention-based multi-view feature fusion cross-domain recommendation method and device | |
| CN104239496A (en) | Collaborative filtering method based on integration of fuzzy weight similarity measurement and clustering | |
| Chen et al. | Next POI recommendation based on location interest mining with recurrent neural networks | |
| CN110110225A (en) | Online education recommended models and construction method based on user behavior data analysis | |
| CN107391582A (en) | The information recommendation method of user preference similarity is calculated based on context ontology tree | |
| CN113515697A (en) | Group dynamic tour route recommendation method and system based on multiple intentions of user | |
| KR102403681B1 (en) | System and Method for Inducing Intelligent Fuzzy Based on User Loacation and Keyword | |
| CN116433269B (en) | Method and device for charging parking lot of zone type unmanned vehicle based on big data | |
| CN110309424A (en) | A kind of socialization recommended method based on Rough clustering | |
| CN115936939A (en) | Project recommendation method and system based on multiple preference degrees | |
| CN114707066A (en) | Scenic spot recommendation method and system based on community perception and self-adaptive random walk |
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: 20211019 |