CN103294795A - Method for adjusting film recommending diversity by utilizing users' characters - Google Patents
Method for adjusting film recommending diversity by utilizing users' characters Download PDFInfo
- Publication number
- CN103294795A CN103294795A CN2013101989219A CN201310198921A CN103294795A CN 103294795 A CN103294795 A CN 103294795A CN 2013101989219 A CN2013101989219 A CN 2013101989219A CN 201310198921 A CN201310198921 A CN 201310198921A CN 103294795 A CN103294795 A CN 103294795A
- Authority
- CN
- China
- Prior art keywords
- film
- user
- diversity
- personality
- multifarious
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a method for adjusting film recommending diversity by utilizing users' characters. The method includes the following steps: a, acquiring characters of users and preferences of the users to films; b, calculating and analyzing relevance of the characters and film diversity; c, determining requirements of the users on diversity of a recommending film list; d, formulating a diversity adjusting list; e, performing content-based film recommending according to the preferences of the users to the films and the diversity adjusting list. Character factors of the users are taken into consideration, and personalized recommending various in diversity level is provided for each user according to influences of the character factors on film recommending diversity, so that requirements of the users on recommending diversity are grasped more accurately and satisfaction degree of the users with a recommending system is increased.
Description
Technical field
The present invention relates to a kind of cinematic data recommendation service, specifically a kind of user's of utilization nature factor is regulated film and is recommended multifarious method.
Background technology
Along with infotech and Internet development, people have suffered from the challenge of increasing information overload, browse a large amount of irrelevant information and products the consumer is constantly run off.In order to address these problems, commending system arises at the historic moment.Yet good commending system is the accurately behavior of predictive user not only, and the visual field that can extending user, and they may be interested to help the user to find those, but the thing of not finding so easily.Therefore, the diversity of recommendation, as an important indicator estimating commending system, more and more can not be out in the cold.
There are many researchs to be devoted to improve the diversity of recommendation list at present, are still an insoluble problem but how between the precision of recommending and diversity, to obtain perfect balance.Most of method that exists goes to control diversity the recommendation list from the angle of algorithm, account for 40% such as diversity, and precision accounts for 60% etc., though by the parameter that obtains after calculating certain rationality is arranged, the diversity degree that often finally offers each user's recommendation list is identical, fixing.And in fact, these class methods have been ignored some oneself factors (such as personality etc.) of user inherence for the influence of recommending the diversity demand, different users is for recommending the diversity demand to be not quite similar, the user who has likes various, various recommendation, and what have then has a preference for certain single, specific recommends etc. really.
Summary of the invention
The objective of the invention is at the recommendation list diversity degree fixed single that offers each user in the prior art, and a kind of user's of utilization personality that ignoring user's personality provides problem such as the influence of recommending the diversity demand is regulated film and is recommended multifarious method, for each user provides the diversity degree different personalized recommendation.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of user's of utilization personality is regulated film and is recommended multifarious method, and this method comprises the steps:
A) obtain user's character trait and its preference to film; Specifically comprise:
⑴ carry out big five properties lattice test by survey to the user, obtains user's five dimension personality scores;
⑵ obtain the user to the preference information of film by user's survey; Wherein, described user ten films that the preference information of film is comprised ten films that the user watched and wants to see; The favorite a kind of film types of user, a movie director, a film performer, a film place of production and a shadow are shown the time period; The a certain dimension attribute of paying close attention to most when the user selects film;
B) the multifarious correlativity of computational analysis personality and film; Specifically comprise:
⑴ calculate the diversity of film one-dimensional attribute;
⑵ calculate the whole diversity of film;
⑶ calculate the correlativity between personality and the film one-dimensional attribute diversity;
⑷ calculate the correlativity between personality and the whole diversity of film;
C) determine that the user is to recommending the multifarious demand of movie listings; Specifically comprise:
⑴ determine that according to the described correlativity of ⑶ in user's personality and the step b) user is to the multifarious demand of film one-dimensional attribute;
⑵ determine that according to the described correlativity of ⑷ in user's personality and the step b) user is to the whole multifarious demand of film;
D) formulate the diversity reconciliation statement; Specifically comprise:
Consider simultaneously the user to the multifarious demand of film one-dimensional attribute and user to the whole multifarious demand of film;
⑵ regulate various film number in the recommendation list;
E) according to preference and the diversity reconciliation statement of user to film, carry out content-based film and recommend; Specifically comprise:
⑴ recommend the various film (3≤n≤7) of n portion according to preference information and the diversity reconciliation statement of user to film;
⑵ residue m portion's film (m=10-n) is recommended according to the hobby situation to film that the user provides.
Compare with background technology, the present invention has following advantage:
The present invention has considered that user's self nature factor is for the influence of recommending the diversity demand when the diversity of balance recommendation list and precision.According to the correlativity between user's personality and the film diversity, the user who analyzes different personality recommends diversity that different demands is arranged for film, thereby has more personalized diversified recommendation for each user provides.
The present invention not only considered the user for the multifarious demand of film one-dimensional attribute (such as liking seeing various types of films, but only like certain director), also considered the user for the whole multifarious demand of film (such as liking various films, different performers is arranged, the director, the place of production etc., film types is not limit yet), thereby can more fully satisfy the user for the needs of the various recommendation of personalization.
Adopted content-based recommendation, increasing the multifarious while, can not produce very big influence to recommending precision, the evaluation index that checks and balance at diversity and precision bidimensional has found a better balance, has improved the precision recommended and user's satisfaction simultaneously.
Description of drawings
Fig. 1 is schematic flow sheet of the present invention.
Embodiment
The present invention determines that according to user's personality the user recommends multifarious demand for film, with this diversity degree of adjusting recommendation list, makes different users can obtain personalized diversified recommendation; By considering that simultaneously the user for the whole diversity of film and the multifarious demand of film one-dimensional attribute, more fully improves the precision of recommendation and user's satisfaction; Process of the present invention is described in detail in detail below:
The first step: carry out big five properties lattice tests (big-five) by survey for the user, obtain user's five dimension personality marks, five dimension personality comprise morality, sharp his property, opening, sociability and adaptability;
Second step: by survey, obtain the user to the preference information of film, ten films (totally 20 ones) that comprise ten films that 1. users watched and want to see are the favorite a kind of film types of user 2., a movie director, a film performer, a film place of production and a film show the time period that (example: favorite film types is action movie, favorite director is the Si Pier Burger, favorite performer is the enlightening Caprio, the favorite film place of production is the U.S., it is favorite that to show the time period be the nineties in 20th century) a certain dimension attribute (example: the director), be designated as Attribute paid close attention to most when 3. the user selects film
Favor;
The 3rd step: by following formula, calculate the diversity of each one-dimensional attribute of each user-selected film, comprise type diversity, director's diversity, performer's diversity, place of production diversity and show the time diversity.
Wherein: above-mentioned formula is example with performer's diversity, the multifarious computing formula of other film one-dimensional attributes by that analogy, i in the formula, j represent i and j portion film, n is the film sum.Utilize the Jaccard coefficient come two films in the computing formula similarity Sim (i, j), the Jaccard computing formula is as follows:
The 4th step: by following formula, calculate the whole diversity of each user-selected film.
Wherein: W
iBe weight at random, 0＜W
i＜1, ∑
iW
i=1, Diversity (attr
i) be the classification diversity calculated in the 3rd step, director's diversity, performer's diversity, place of production diversity and show multifarious value of time;
The 5th step: use the SPSS statistical software, five dimension personality scores and film one-dimensional attribute diversity score are carried out the analysis of Spearman (Spearman) related coefficient, and the significant correlation item that obtains (degree of confidence sig＜0.05 is significant correlation) put into personality-film one-dimensional attribute diversity correlation table, and indicate that both sides relation belongs to positive correlation or negative correlation.
In like manner use the whole multifarious correlativity of identical methods analyst personality and film, the significant correlation item (degree of confidence sig＜0.05) that obtains is put into the whole diversity correlation table of personality-film, and indicate both positive and negative incidence relations.
The 6th step: according to user's personality and the personality-film one-dimensional attribute diversity correlation table in the 5th step, determine that the user is to the multifarious desirability of film one-dimensional attribute in the recommendation list.Demand is divided into 3 classes altogether: height, in, low }.
The one-dimensional attribute refers in particular to this dimension attribute Attribute that pays close attention to most when the user selects film in second step among the present invention
Favor
The 7th one: according to user's personality and the whole diversity correlation table of the personality-film in the 5th step, determine that the user is to the whole multifarious desirability of film in the recommendation list.Demand is divided into 3 classes altogether: height, in, low };
The 8th step: comprehensive user formulates the diversity reconciliation statement for the multifarious demand of film one-dimensional attribute with for the whole multifarious demand of film, determines film number n(3≤n≤7 various in the recommendation list);
The diversity reconciliation statement
The 9th step: go on foot preference and the diversity reconciliation statement in the 8th step for film that provides in second according to the user, the content-based personalized recommendation tabulation that comprises 10 films is provided for the user, wherein n portion film all has diversity (the shared weight maximum of a most important dimension attribute) on five dimension attributes, and residue m portion's film (m=10-n) is recommended according to the hobby situation to film that the user provides.
Claims (3)
1. one kind is utilized user's personality adjusting film to recommend multifarious method, it is characterized in that, comprises the steps:
A) obtain user's character trait and its preference to film; Specifically comprise:
⑴ carry out big five properties lattice test by survey to the user, obtains user's five dimension personality scores;
⑵ obtain the user to the preference information of film by user's survey;
B) the multifarious correlativity of computational analysis personality and film; Specifically comprise:
⑴ calculate the diversity of film one-dimensional attribute;
⑵ calculate the whole diversity of film;
⑶ calculate the correlativity between personality and the film one-dimensional attribute diversity;
⑷ calculate the correlativity between personality and the whole diversity of film;
C) determine that the user is to recommending the multifarious demand of movie listings; Specifically comprise:
⑴ determine that according to the described correlativity of ⑶ in user's personality and the step b) user is to the multifarious demand of film one-dimensional attribute;
⑵ determine that according to the described correlativity of ⑷ in user's personality and the step b) user is to the whole multifarious demand of film;
D) formulate the diversity reconciliation statement; Specifically comprise:
Consider simultaneously the user to the multifarious demand of film one-dimensional attribute and user to the whole multifarious demand of film;
⑵ regulate various film number in the recommendation list;
E) according to preference and the diversity reconciliation statement of user to film, carry out content-based film and recommend; Specifically comprise:
⑴ recommend various film 3≤n≤7 of n portion according to preference information and the diversity reconciliation statement of user to film;
⑵ residue
mPortion's film is recommended according to the hobby situation to film that the user provides; Wherein: m=10-n.
2. method according to claim 1 is characterized in that ten films that described user comprises ten films that the user watched and wants to see the preference information of film; The favorite a kind of film types of user, a movie director, a film performer, a film place of production and a film are shown the time period; The a certain dimension attribute of paying close attention to most when the user selects film.
3. method according to claim 1 is characterized in that described five dimension personality comprise morality, sharp his property, opening, sociability and adaptability.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101989219A CN103294795A (en) | 2013-05-24 | 2013-05-24 | Method for adjusting film recommending diversity by utilizing users' characters |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101989219A CN103294795A (en) | 2013-05-24 | 2013-05-24 | Method for adjusting film recommending diversity by utilizing users' characters |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103294795A true CN103294795A (en) | 2013-09-11 |
Family
ID=49095657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013101989219A Pending CN103294795A (en) | 2013-05-24 | 2013-05-24 | Method for adjusting film recommending diversity by utilizing users' characters |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103294795A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462454A (en) * | 2014-12-17 | 2015-03-25 | 上海斐讯数据通信技术有限公司 | Character analyzing method |
| CN104462385A (en) * | 2014-12-10 | 2015-03-25 | 山东科技大学 | Personalized movie similarity calculation method based on user interest model |
| CN109963509A (en) * | 2016-08-05 | 2019-07-02 | V·尤利亚诺通讯公司 | For the system and method to content consumer recommendation service |
| CN112395496A (en) * | 2020-10-22 | 2021-02-23 | 上海众源网络有限公司 | Information recommendation method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102360358A (en) * | 2011-09-28 | 2012-02-22 | 百度在线网络技术（北京）有限公司 | Keyword recommendation method and system |
| CN102411754A (en) * | 2011-11-29 | 2012-04-11 | 南京大学 | Personalized recommendation method based on commodity property entropy |
| US20120180107A1 (en) * | 2011-01-07 | 2012-07-12 | Microsoft Corporation | Group-associated content recommendation |
-
2013
- 2013-05-24 CN CN2013101989219A patent/CN103294795A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120180107A1 (en) * | 2011-01-07 | 2012-07-12 | Microsoft Corporation | Group-associated content recommendation |
| CN102360358A (en) * | 2011-09-28 | 2012-02-22 | 百度在线网络技术（北京）有限公司 | Keyword recommendation method and system |
| CN102411754A (en) * | 2011-11-29 | 2012-04-11 | 南京大学 | Personalized recommendation method based on commodity property entropy |
Non-Patent Citations (2)
| Title |
|---|
| LI CHEN 等: "How Personality Influences Users’Needs for Recommendation Diversity", 《CHI"13 EXTENDED ABSTRACTS ON HUMAN FACTORS IN COMPUTING SYSTEMS》 * |
| WEN WU 等: "Using personality to adjust diversity in recommender systems", 《PROCEEDINGS OF THE 24TH ACM CONFERENCE ON HYPERTEXT AND SOCIAL MEDIA》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104462385A (en) * | 2014-12-10 | 2015-03-25 | 山东科技大学 | Personalized movie similarity calculation method based on user interest model |
| CN104462454A (en) * | 2014-12-17 | 2015-03-25 | 上海斐讯数据通信技术有限公司 | Character analyzing method |
| CN109963509A (en) * | 2016-08-05 | 2019-07-02 | V·尤利亚诺通讯公司 | For the system and method to content consumer recommendation service |
| US11197059B2 (en) | 2016-08-05 | 2021-12-07 | V. Juliano Communications, LLC | System and method for recommending a content service to a content consumer |
| US11985383B2 (en) | 2016-08-05 | 2024-05-14 | V. Juliano Communications, LLC | System and method for recommending a content service to a content consumer |
| CN112395496A (en) * | 2020-10-22 | 2021-02-23 | 上海众源网络有限公司 | Information recommendation method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104935963B (en) | A kind of video recommendation method based on timing driving | |
| Kuo et al. | Building and evaluating a location-based service recommendation system with a preference adjustment mechanism | |
| WO2020048084A1 (en) | Resource recommendation method and apparatus, computer device, and computer-readable storage medium | |
| Chung et al. | A general consumer preference model for experience products: application to internet recommendation services | |
| US8880516B2 (en) | Endorsing local search results | |
| Yin et al. | Silence is also evidence: interpreting dwell time for recommendation from psychological perspective | |
| US20130097176A1 (en) | Method and system for data mining of social media to determine an emotional impact value to media content | |
| CN103744917B (en) | Method and system are recommended in mixing | |
| CN101482884A (en) | Cooperation recommending system based on user predilection grade distribution | |
| US20160283952A1 (en) | Ranking information providers | |
| RU2012100619A (en) | MACHINE READABLE MEDIA FOR USE IN COLLECTIVE KNOWLEDGE SYSTEMS | |
| US10970296B2 (en) | System and method for data mining and similarity estimation | |
| US20090144226A1 (en) | Information processing device and method, and program | |
| KR102131791B1 (en) | Method to provide recommended contents and associated contents | |
| CN107025606B (en) | Project recommendation method combining evaluation data and trust relationship in social network | |
| CN105741077A (en) | Occupational information providing method and system | |
| CN109309847A (en) | A kind of video display entity temperature comprehensive estimation method | |
| CN104008193B (en) | A kind of information recommendation method based on group of typical user discovery technique | |
| CN103294795A (en) | Method for adjusting film recommending diversity by utilizing users' characters | |
| KR101712291B1 (en) | System for recommending a user-customized famous place based on opinion mining and Method of the Same | |
| US20170323019A1 (en) | Ranking information providers | |
| KR101318975B1 (en) | A Personalized Intelli gent Method for Recommending Mobile Applications | |
| US20130332440A1 (en) | Refinements in Document Analysis | |
| KR101542417B1 (en) | Method and apparatus for learning user preference | |
| US20240112210A1 (en) | Self-learning valuation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130911 |