CN105426514A - Personalized mobile APP recommendation method - Google Patents
Personalized mobile APP recommendation method Download PDFInfo
- Publication number
- CN105426514A CN105426514A CN201510867018.6A CN201510867018A CN105426514A CN 105426514 A CN105426514 A CN 105426514A CN 201510867018 A CN201510867018 A CN 201510867018A CN 105426514 A CN105426514 A CN 105426514A
- Authority
- CN
- China
- Prior art keywords
- app
- user
- recommendation
- model
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a personalized mobile APP recommendation method. The method comprises: obtaining information of a user and an APP from an application market and preprocessing the information; by utilizing a sentiment-aspect-region (SAR) model, inputting a preprocessed document to obtain potential preferences of the user to emotion-aspect-region of the APP; predicting the value of a probability of selecting one APP by the user; processing and converting the value into an APP index file and a user index file; by utilizing a co-correlative topic model, obtaining a recommendation score matrix of User-App; linearly combining a probability value obtained by the SAR model and a recommendation score obtained by a CTM model; and distributing a weight to obtain a final recommendation value. According to the method, the defect of only considering a single element in a conventional recommendation method is overcome. The potential preferences of the user are discovered by comprehensively considering aspects, sentiments, types and regions in comments, the actual demands of the user are better met, the preference degree of the user to each attribute of the APP is explored, user demands and APP features are better known, and the cold-start problem is solved.
Description
Technical field
The present invention proposes a kind of Mobile solution recommend method, the Mobile solution APP recommend method of particularly personalization.
Background technology
It is more convenient that the development of mobile phone A PP gives user, facilitates the life of user.But, APP countless and various in stylely also cause some problems to user.Research finds only by browse and simply inquiry is found the useful and APP of user preference and become quite difficulty, information excessive to a certain extent means poor information, therefore just need certain instrument to find rapidly required for user and the information of preference carrys out aid decision making, prevent user from losing.So, there is many APP recommend methods.
Before the present invention makes, traditional recommend method such as collaborative filtering (CF) is devoted to by finding the user having similar interests with designated user in customer group, these similar users comprehensive are to the evaluation of a certain information, and formation system recommends APP to the fancy grade prediction of this designated user to this information.But, business application along with personalization is extended to the every aspect of user's life information stream, personalized recommendation technology is also at development with rapid changepl. never-ending changes and improvements, be similar to the such earlier technique of collaborative filtering can not meet new environment under requirement, such as when user and commodity more see increase, the performance of system can be more and more lower also or when user is very sparse to the evaluation of commodity, the similarity between the user obtained based on the evaluation of user like this may the inaccurate commodity that even cause not recommended.In addition, current most of technology only considers individual element, but while user has higher demand to commodity, APP is recommended, except function, the impact of its APP attribute, kind, geographic position and user feeling also should be considered, such as in geographic position, such as U.S. group, popular comment, Google Maps are like this based on the software in geographic position, and the area size involved by them can have influence on whether this APP of this recommendation.
Summary of the invention
The object of the invention is to overcome above-mentioned defect, develop a kind of APP recommend method of personalization.
Technical scheme of the present invention is:
Personalized Mobile solution APP recommend method, is characterized in that step is as follows:
Step 1). Data Collection: the information obtaining user and Mobile solution and APP from application market, comprises functional description and review information;
Step 2). carry out pre-service to avoid occurring cold start-up problem to the original APP data obtained;
Step 3). utilize emotion-aspect-area and SAR model, using the review information of APP as input document, obtain the territory of use of user for the emotion of APP, the aspect of APP and APP respectively, obtain user thus to the potential preference of APP different attribute and predict that user selects the probable value of APP;
Step 4). to step 2) data that obtain are for further processing, and are converted to APP index file and user index file respectively;
Step 5). utilize association's related subject and CTM model, input abovementioned steps 4) two parts of files, obtain the recommender score matrix of User-App;
Step 6). the probable value that SAR model and CTM model obtain respectively is linearly combined with recommender score, then sorts by the online proposed algorithm of Top-N, the higher APP of prediction scoring is recommended corresponding user.
Described step 1) in, in application shop GooglePlay, user to he the grading of used APP be openly visible, once the ID obtaining user just can see user all APP of commenting on, by climbing data tool, all raw data are retrieved thus.
Described step 2) in the preprocessing process of raw data comprise:
A) remove write be less than 2 comment users and filter user after without any comment APP;
B) hold in the palm willingization: remove punctuation mark, remove numeral;
C) stop words is removed: remove English stop words, comprise preposition, pronoun, article;
D) stemmed: the prototype each word being converted into it, past tense is converted into prototype, and present progressive tense is converted into prototype.
Described step 3) in the computing formula of calculated recommendation probability:
Namely represent that user u likes APPt and the probability of grading to it, wherein, t, s
+, u, r, a, c
trepresent APP respectively, positive emotion, user, area, the aspect of APP and the kind of APP.
Described step 4) in data processing, it is divided into following step:
A) to step 2) in all APP of obtaining be numbered, be followed successively by 0,1,2,3,4 ..., n, each numbering its APP information corresponding, is step 3) in SAR model filter the information obtained;
B) to step 2) in all users of obtaining be numbered, be followed successively by 0,1,2,3,4 ..., n, each numbering its user profile corresponding, is step 3) in SAR model filter the information obtained;
C) collected data preparation is become a user index input file, call format: the information of a behavior user, the coding+1, the second that row head is user be classified as user the APP quantity of grading, remainder is classified as all APP numberings that user graded;
D) collected data preparation is become a APP index input file, call format: the information of a behavior APP, the coding+1, the second that row head is APP is classified as and carries out the number of users of grading to this APP, and remainder is classified as all to the Customs Assigned Number of APP grading;
E) by above two parts of files input CTM model, obtain a User-App recommender score matrix, recommendation is that regular representation can be recommended, the more large more worth recommendation of value, on the contrary recommendation is bear then to represent that this APP is do not recommend to be worth to this user, matrix behavior APP, is classified as user.
7. APP many attributes recommend method according to claim 1, is characterized in that step 6) in Probability p (t, s that SAR model is obtained
+| u) and the recommendation r that obtains of GTM model
utlinear junction close computing formula, as follows: to set two parameter alpha, β, then merging recommender score Score be:
Score=αp(t,s
+|u)+βr
ut
Wherein, α, β are weighing input parameters.
Advantage of the present invention and effect are to consider the object that aspect in comment, emotion, kind and area reach personalized recommendation.Mainly contain following advantages:
1. this recommend method considers aspect in comment, emotion, kind and area to find the potential preference of user, more meets the actual demand of user.
2. this recommend method can solve traditional collaborative filtering and the insurmountable cold start-up problem of CF.
3. conventional recommendation method such as CF does not use the content of APP, it has the user of parallel pattern to recommend based on selected APP, and this recommend method is content-based and user's grading simultaneously, prediction of result is content-based or user's grading depends on that how many users grade to APP.
4. this recommend method is a kind of personalized recommendation, and the application recommended can consider the interest, their location etc. of user.
5. this recommend method proposes the brand-new sorting technique to APP attribute, obtains more detailed by the Attribute transposition of APP, explores user thus to the preference of each attribute of APP, understands user's request and APP feature better.
Accompanying drawing explanation
Fig. 1---overall procedure schematic diagram of the present invention.
The functional circuit information schematic diagram of Fig. 2---GoogleMaps of the present invention.
Fig. 3---user Sarah of the present invention is to the review information schematic diagram of YellowPages.
Fig. 4---user Sarah of the present invention is to the review information schematic diagram of CommanderCompassLite.
The functional circuit information schematic diagram of Fig. 5---the GoogleMaps after the present invention is pretreated.
Fig. 6---the user Sarah after the present invention is pretreated is to the review information schematic diagram of YellowPages.
Fig. 7---the user Sarah after the present invention is pretreated is to the review information schematic diagram of CommanderCompassLite.
Fig. 8---the preliminary Output rusults schematic diagram of emotion-aspect-regional model of the present invention.
Fig. 9---the present invention assists the call format schematic diagram of the user index input file of correlation model.
Figure 10---the present invention assists the call format schematic diagram of the APP index input file of correlation model.
Figure 11---the present invention assists the output format schematic diagram of correlation model User-App recommender score matrix.
Embodiment
Technical thought of the present invention is:
The present invention considers in conjunction with the impact on recommended technology of aspect, emotion, kind and area, detailed classification is compared to APP attribute, the ratio of such as interface, geographic position, function menu, discharging quantity and activation amount, setting, understand user in further detail to the requirement of APP different attribute and preference with this, thus make recommendation effect better.Also utilize this personalized recommendation of association's related subject model, to ensure that this recommend method can be used widely.
The present invention carries out modeling in conjunction with CTM model and SAR model to user comment information, finds the potential preference of user and carry out detailed recommendation with this.
Below in conjunction with accompanying drawing, technical scheme of the present invention is described in detail:
Step 1).
Data Collection: in application shop as in GooglePlay, the grading of user to he the used APP of institute is openly visible, the ID once acquisition user just can see user all APP of commenting on.Thus can by climbing data tool by all raw data as user comment and functional description are retrieved.Be exemplified below: suppose that user Sarah wants to look for a navigation APP, user used YellowPages and CommanderCompassLite two to navigate APP before this, Fig. 2 is the functional circuit information of GoogleMaps, this APP is that user Sarah is original, Fig. 3 is the user comment information of Sarah to YellowPages, and Fig. 4 is the user comment information of Sarah to CommanderCompassLite.
Step 2).
Carry out pre-service to avoid occurring cold start-up problem to the original APP data obtained.First, remove write be less than 2 comment users and filter user after without any comment APP.Secondly, carry out asking willingization to be tokenization, respectively removal punctuation mark as " ", "? " and remove numeral as " 1 ", " 23 ".Then stop words and stopping, such as preposition " for " is removed, " to "; Pronoun " it ", " he "; Article " a ", " an " " the ".Finally stemmed i.e. stemming, each word is converted into its prototype, such as " fixed " is past tense, is converted into prototype " fix "; " moving " is present progressive tense, is converted into prototype " move ".The algorithm selected time stemmed is Porterstemmingalgorithm.After three step operations, produce the pretreated document removing noise.After process, result is as shown in Fig. 5/6/7.
Step 3).
Utilize SAR model, using the functional description of APP and user comment information as input document, obtain following information respectively:
A) the function keyword of GoogleMaps: nayigate, fast, easy, comprehensive, accurate, voice-guided, report, reroute, detailed and weak point: not-all-regions, battery-loss
Supplementary notes: comparatively length language need be converted into compound word.
B) this user is for a) emotion of used YellowPages and sentiment vocabulary: best, love; B) aspect and aspect vocabulary: navigation, price (free), view; C) territory of use and region:not-all-regions and this user are for a) the sentiment vocabulary of CommanderCompassLite: well, great; B) aspect vocabulary: accurate, tech, price, display and view.
As shown in Figure 8, obtain from above user on APP navigation (navigation), the free aspect such as (free) and view (view) potential preference and geographic position impact, finally obtain according to formula prediction probability value p (t, the s that user selects APP
+| be u) 5.91267063568302.
Step 4).
To step 2) data that obtain do to arrange further, are converted to APP index file and user index file respectively.First, obtained all APP are numbered, are followed successively by 0,1,2,3,4 ..., n, each numbering its APP information corresponding, is step 3) in SAR model filter the information obtained, Customs Assigned Number is the same.Arrange GoogleMaps be numbered 3, Customs Assigned Number is 6.Secondly, as shown in Figure 9, collected data preparation is become a user index input file, the information of an one behavior user, the coding+1, the second that row head is user be classified as user the APP quantity of grading, remainder is classified as all APP numberings that user graded.In addition, as shown in Figure 10, collected data preparation is become a APP index input file, the information of an one behavior APP, the coding+1, the second that row head is APP is classified as and carries out the number of users of grading to this APP, and remainder is classified as all to the Customs Assigned Number of APP grading.
Step 5).
As shown in figure 11, the file input CTM model that above two parts of arrangements are obtained, obtain a User-App recommender score matrix, recommendation is that regular representation can be recommended, the more large more worth recommendation of value, otherwise, recommendation is bear then to represent that this APP is do not recommend to be worth to this user, this matrix behavior APP, is classified as user, by scheming APP3 is 5.14723419271134 to the recommendation of user 6.
Step 6).
By probable value p (t, s that SAR model and CTM model obtain respectively
+| u) with recommender score r
ut is linearin conjunction with, as follows: to set two weighing input parameters α, β, its weights can distribute voluntarily according to self preference of user, such as, if the recommendation results of user's more preference SAR model, then can by α, β respectively assignment be 60%, 40%, obtaining the merging recommender score Score of GoogleMaps to user 6 is thus 3.477934629648461, in addition, also have other navigation type APP to the recommender score of user 6, then sort by the online proposed algorithm of Top-N, the higher APP of prediction scoring is recommended user 6, reaches the object of recommendation.
Special instruction:
The online proposed algorithm of Top-N is conventional a kind of means of carrying out personalized information push directly to user, and it effectively can shorten the time of calculating, thus improves counting yield.Suppose existence array, merging recommender score obtained above is the element in array, and what Top-N recommended to refer to online is exactly from this array existed, and finds out front n maximum or minimum element.Object due to this recommend method is that the APP finding recommender score higher recommends to user, and therefore Top-N herein refers to and finds front n maximum element, is implemented as follows:
A) take out front n element of array, establishment length is the most rickle of n.
The surplus element of the array that B) circulates from n, if element a is larger than the root node of most rickle, is arranged to the root node of most rickle by a, and allows the characteristic of piling and minimally piling.
C) after having circulated, all elements in most rickle is exactly maximum n the element needing to look for, and is the highest n of a recommender score APP, sorts thus and effectively recommends.
Claims (6)
1. the Mobile solution APP recommend method of personalization, is characterized in that step is as follows:
Step 1). Data Collection: the information obtaining user and Mobile solution and APP from application market, comprises functional description and review information;
Step 2). carry out pre-service to avoid occurring cold start-up problem to the original APP data obtained;
Step 3). utilize emotion-aspect-area and SAR model, using the review information of APP as input document, obtain the territory of use of user for the emotion of APP, the aspect of APP and APP respectively, obtain user thus to the potential preference of APP different attribute and predict that user selects the probable value of APP;
Step 4). to step 2) data that obtain are for further processing, and are converted to APP index file and user index file respectively;
Step 5). utilize association's related subject and CTM model, input abovementioned steps 4) two parts of files, obtain the recommender score matrix of User-App;
Step 6). the probable value that SAR model and CTM model obtain respectively is linearly combined with recommender score, then sorts by the online proposed algorithm of Top-N, the higher APP of prediction scoring is recommended corresponding user.
2. the Mobile solution APP recommend method of personalization according to claim 1, it is characterized in that step 1) in, in application shop GooglePlay, user to he the grading of used APP be openly visible, once obtain the ID of user just can see user all APP of commenting on, by climbing data tool, all raw data are retrieved thus.
3. the Mobile solution APP recommend method of personalization according to claim 1, is characterized in that step 2) in the preprocessing process of raw data comprise:
A) remove write be less than 2 comment users and filter user after without any comment APP;
B) hold in the palm willingization: remove punctuation mark, remove numeral;
C) stop words is removed: remove English stop words, comprise preposition, pronoun, article;
D) stemmed: the prototype each word being converted into it, past tense is converted into prototype, and present progressive tense is converted into prototype.
4. the Mobile solution APP recommend method of personalization according to claim 1, is characterized in that step 3) in the computing formula of calculated recommendation probability:
Namely represent that user u likes APPt and the probability of grading to it, wherein, t, s
+, u, r, a, c
trepresent APP respectively, positive emotion, user, area, the aspect of APP and the kind of APP.
5. the Mobile solution APP recommend method of personalization according to claim 1, is characterized in that step 4) in data processing, it is divided into following step:
A) to step 2) in all APP of obtaining be numbered, be followed successively by 0,1,2,3,4 ..., n, each numbering its APP information corresponding, is step 3) in SAR model filter the information obtained;
B) to step 2) in all users of obtaining be numbered, be followed successively by 0,1,2,3,4 ..., n, each numbering its user profile corresponding, is step 3) in SAR model filter the information obtained;
C) collected data preparation is become a user index input file, call format: the information of a behavior user, the coding+1, the second that row head is user be classified as user the APP quantity of grading, remainder is classified as all APP numberings that user graded;
D) collected data preparation is become a APP index input file, call format: the information of a behavior APP, the coding+1, the second that row head is APP is classified as and carries out the number of users of grading to this APP, and remainder is classified as all to the Customs Assigned Number of APP grading;
E) by above two parts of files input CTM model, obtain a User-App recommender score matrix, recommendation is that regular representation can be recommended, the more large more worth recommendation of value, on the contrary recommendation is bear then to represent that this APP is do not recommend to be worth to this user, matrix behavior APP, is classified as user.
6. the Mobile solution APP recommend method of personalization according to claim 1, is characterized in that step 6) in Probability p (t, s that SAR model is obtained
+| u) and the recommendation r that obtains of CTM model
utlinear junction close computing formula, as follows: to set two parameter alpha, β, then merging recommender score Score be:
Score=αp(t,s
+|u)+βr
ut
Wherein, α, β are weighing input parameters.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510867018.6A CN105426514B (en) | 2015-11-30 | 2015-11-30 | Personalized mobile application APP recommended method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510867018.6A CN105426514B (en) | 2015-11-30 | 2015-11-30 | Personalized mobile application APP recommended method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105426514A true CN105426514A (en) | 2016-03-23 |
CN105426514B CN105426514B (en) | 2019-03-01 |
Family
ID=55504726
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510867018.6A Active CN105426514B (en) | 2015-11-30 | 2015-11-30 | Personalized mobile application APP recommended method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105426514B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106095776A (en) * | 2016-05-25 | 2016-11-09 | 中山大学 | A kind of method that user is carried out Topics Crawling and application recommendation |
CN106250146A (en) * | 2016-07-28 | 2016-12-21 | 维沃移动通信有限公司 | A kind of application demand treating method and apparatus |
CN106547922A (en) * | 2016-12-07 | 2017-03-29 | 广州优视网络科技有限公司 | A kind of sort method of application program, device and server |
CN106569687A (en) * | 2016-10-19 | 2017-04-19 | 北京三快在线科技有限公司 | Icon arrangement method and apparatus for virtual buttons, and terminal |
CN106844240A (en) * | 2017-01-25 | 2017-06-13 | 百度在线网络技术(北京)有限公司 | Information displaying method and device |
CN106846094A (en) * | 2016-12-29 | 2017-06-13 | 广州优视网络科技有限公司 | A kind of method and apparatus for recommending application message based on application has been installed |
CN106952129A (en) * | 2017-02-23 | 2017-07-14 | 广东小天才科技有限公司 | Application recommendation method and device for application store |
CN107220303A (en) * | 2017-05-10 | 2017-09-29 | 努比亚技术有限公司 | Recommendation method, device and the computer-readable medium of a kind of application |
CN107329770A (en) * | 2017-07-04 | 2017-11-07 | 扬州大学 | The personalized recommendation method repaired for software security BUG |
CN107861753A (en) * | 2017-06-26 | 2018-03-30 | 平安普惠企业管理有限公司 | APP generations index, search method and system and readable storage medium storing program for executing |
CN108307039A (en) * | 2017-12-21 | 2018-07-20 | 维沃移动通信有限公司 | A kind of application message methods of exhibiting and mobile terminal |
CN108563758A (en) * | 2018-04-17 | 2018-09-21 | 广州虎牙信息科技有限公司 | Number of users measuring method, device, equipment and storage medium |
CN108710680A (en) * | 2018-05-18 | 2018-10-26 | 哈尔滨理工大学 | It is a kind of to carry out the recommendation method of the film based on sentiment analysis using deep learning |
CN108763258A (en) * | 2018-04-03 | 2018-11-06 | 平安科技(深圳)有限公司 | Document subject matter parameter extracting method, Products Show method, equipment and storage medium |
CN109189887A (en) * | 2018-09-07 | 2019-01-11 | 江苏瑞康安全装备有限公司 | A kind of micro-blog information recommended method of facing moving terminal |
US10885568B2 (en) | 2016-11-15 | 2021-01-05 | Industrial Technology Research Institute | Interactive recommendation system and method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20130059738A (en) * | 2011-11-29 | 2013-06-07 | 에스케이플래닛 주식회사 | System and method for recommending application using contents analysis |
CN103559262A (en) * | 2013-11-04 | 2014-02-05 | 北京邮电大学 | Community-based author and academic paper recommending system and recommending method |
CN104461558A (en) * | 2014-12-22 | 2015-03-25 | 扬州大学 | Software-fix-orientated individualized task recommendation method based on co-correlativity topic model |
-
2015
- 2015-11-30 CN CN201510867018.6A patent/CN105426514B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20130059738A (en) * | 2011-11-29 | 2013-06-07 | 에스케이플래닛 주식회사 | System and method for recommending application using contents analysis |
CN103559262A (en) * | 2013-11-04 | 2014-02-05 | 北京邮电大学 | Community-based author and academic paper recommending system and recommending method |
CN104461558A (en) * | 2014-12-22 | 2015-03-25 | 扬州大学 | Software-fix-orientated individualized task recommendation method based on co-correlativity topic model |
Non-Patent Citations (1)
Title |
---|
KAIQI ZHAO 等: "SAR: A sentiment-aspect-region model for user preference analysis in geo-tagged reviews", 《2015 IEEE 31ST INTERNATIONAL CONFERENCE ON DATA ENGINEERING (ICDE)》 * |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106095776B (en) * | 2016-05-25 | 2019-10-01 | 中山大学 | The method that a kind of couple of user carries out Topics Crawling and application is recommended |
CN106095776A (en) * | 2016-05-25 | 2016-11-09 | 中山大学 | A kind of method that user is carried out Topics Crawling and application recommendation |
CN106250146A (en) * | 2016-07-28 | 2016-12-21 | 维沃移动通信有限公司 | A kind of application demand treating method and apparatus |
CN106250146B (en) * | 2016-07-28 | 2021-03-12 | 维沃移动通信有限公司 | Application demand processing method and device |
CN106569687A (en) * | 2016-10-19 | 2017-04-19 | 北京三快在线科技有限公司 | Icon arrangement method and apparatus for virtual buttons, and terminal |
US10885568B2 (en) | 2016-11-15 | 2021-01-05 | Industrial Technology Research Institute | Interactive recommendation system and method |
CN106547922A (en) * | 2016-12-07 | 2017-03-29 | 广州优视网络科技有限公司 | A kind of sort method of application program, device and server |
CN106547922B (en) * | 2016-12-07 | 2020-08-25 | 阿里巴巴(中国)有限公司 | Application program sorting method and device and server |
CN106846094A (en) * | 2016-12-29 | 2017-06-13 | 广州优视网络科技有限公司 | A kind of method and apparatus for recommending application message based on application has been installed |
CN106844240A (en) * | 2017-01-25 | 2017-06-13 | 百度在线网络技术(北京)有限公司 | Information displaying method and device |
CN106952129A (en) * | 2017-02-23 | 2017-07-14 | 广东小天才科技有限公司 | Application recommendation method and device for application store |
CN107220303A (en) * | 2017-05-10 | 2017-09-29 | 努比亚技术有限公司 | Recommendation method, device and the computer-readable medium of a kind of application |
CN107861753A (en) * | 2017-06-26 | 2018-03-30 | 平安普惠企业管理有限公司 | APP generations index, search method and system and readable storage medium storing program for executing |
CN107861753B (en) * | 2017-06-26 | 2020-12-11 | 平安普惠企业管理有限公司 | APP generation index, retrieval method and system and readable storage medium |
CN107329770A (en) * | 2017-07-04 | 2017-11-07 | 扬州大学 | The personalized recommendation method repaired for software security BUG |
CN108307039A (en) * | 2017-12-21 | 2018-07-20 | 维沃移动通信有限公司 | A kind of application message methods of exhibiting and mobile terminal |
WO2019192122A1 (en) * | 2018-04-03 | 2019-10-10 | 平安科技(深圳)有限公司 | Document topic parameter extraction method, product recommendation method and device, and storage medium |
CN108763258A (en) * | 2018-04-03 | 2018-11-06 | 平安科技(深圳)有限公司 | Document subject matter parameter extracting method, Products Show method, equipment and storage medium |
CN108763258B (en) * | 2018-04-03 | 2023-01-10 | 平安科技(深圳)有限公司 | Document theme parameter extraction method, product recommendation method, device and storage medium |
CN108563758A (en) * | 2018-04-17 | 2018-09-21 | 广州虎牙信息科技有限公司 | Number of users measuring method, device, equipment and storage medium |
CN108710680A (en) * | 2018-05-18 | 2018-10-26 | 哈尔滨理工大学 | It is a kind of to carry out the recommendation method of the film based on sentiment analysis using deep learning |
CN109189887A (en) * | 2018-09-07 | 2019-01-11 | 江苏瑞康安全装备有限公司 | A kind of micro-blog information recommended method of facing moving terminal |
Also Published As
Publication number | Publication date |
---|---|
CN105426514B (en) | 2019-03-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105426514A (en) | Personalized mobile APP recommendation method | |
US10217058B2 (en) | Predicting interesting things and concepts in content | |
US11176142B2 (en) | Method of data query based on evaluation and device | |
CN101364239B (en) | Method for auto constructing classified catalogue and relevant system | |
CN106339502A (en) | Modeling recommendation method based on user behavior data fragmentation cluster | |
JP4637969B1 (en) | Properly understand the intent of web pages and user preferences, and recommend the best information in real time | |
CN107844565B (en) | Commodity searching method and device | |
CN104252456B (en) | A kind of weight method of estimation, apparatus and system | |
WO2018040069A1 (en) | Information recommendation system and method | |
US9344507B2 (en) | Method of processing web access information and server implementing same | |
TW201337814A (en) | Product information publishing method and device | |
CN103838756A (en) | Method and device for determining pushed information | |
CN108021715B (en) | Heterogeneous label fusion system based on semantic structure feature analysis | |
CN101833560A (en) | Manufacturer public praise automatic sequencing system based on internet | |
KR20120101233A (en) | Method for providing sentiment information and method and system for providing contents recommendation using sentiment information | |
US8204872B2 (en) | Method and system for instantly expanding a keyterm and computer readable and writable recording medium for storing program for instantly expanding keyterm | |
CN104111925A (en) | Item recommendation method and device | |
KR101088710B1 (en) | Method and Apparatus for Online Community Post Searching Based on Interactions between Online Community User and Computer Readable Recording Medium Storing Program thereof | |
CN106354867A (en) | Multimedia resource recommendation method and device | |
CN105468649A (en) | Method and apparatus for determining matching of to-be-displayed object | |
JP2008234338A (en) | Season degree analysis system, in-season degree analysis method, and season degree analysis program | |
CN102929975A (en) | Recommending method based on document tag characterization | |
CN116882414B (en) | Automatic comment generation method and related device based on large-scale language model | |
CN103942233B (en) | The lobby page recognition methods of directory type web and device | |
CN105975508A (en) | Personalized meta-search engine searched result merging and sorting method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |