CN105447126A - Game prop personalized recommendation method - Google Patents
Game prop personalized recommendation method Download PDFInfo
- Publication number
- CN105447126A CN105447126A CN201510793393.0A CN201510793393A CN105447126A CN 105447126 A CN105447126 A CN 105447126A CN 201510793393 A CN201510793393 A CN 201510793393A CN 105447126 A CN105447126 A CN 105447126A
- Authority
- CN
- China
- Prior art keywords
- data
- player
- stage property
- recommendation method
- personalized recommendation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Abstract
The present invention relates to a game prop personalized recommendation method, comprising: carrying out extraction, cleaning and processing on data; mapping the data into a low-dimensional space from an input space; classifying the data to partition players into different categories; and recommending props which are most possibly bought by each player for the player. According to the game prop personalized recommendation method disclosed by the present invention, a calculated amount of a recommendation algorithm is greatly reduced, prop recommendation accuracy is significantly improved, and a user experience effect of a network game is promoted.
Description
Technical field
The present invention relates to field of network game, particularly a kind of game item personalized recommendation method.
Background technology
In existing network game, there is numerous game items.Three kinds of technology below usual employing realize game item personalized recommendation: (1) Data Dimensionality Reduction technology; (2) sorting algorithm; (3) proposed algorithm.Each algorithmic technique all plays most crucial effect in game item proposed algorithm, and three kinds of algorithms are together to form an algorithmic system, indispensable.
Wherein, what Data Dimensionality Reduction technology will solve is the complicated problem of game data.Because game data itself has some non-traditional features, such as: game data source is complicated, data type is varied, data dimension is high (hundreds of or thousands of), data magnanimity.Therefore, before use proposed algorithm, both needed reduced data, ensure that the information comprised in data can not be lost again as far as possible, this just needs Data Dimensionality Reduction technology.The current comparative maturity of Data Dimensionality Reduction technology, improvable space is little, and usage factor analytic approach carries out Data Dimensionality Reduction herein.
Data after sorting algorithm uses dimensionality reduction are classified to game user.If do not classified to player before use proposed algorithm, calculated amount can be made to be doubled and redoubled on the one hand; On the other hand, because the hobby matrix difference of dissimilar player is huge, when not carrying out classification in advance to player, the difference between the fuzzy dissimilar player of meeting, thus the accuracy that impact is recommended.Sorting algorithm is also the technology of comparative maturity, and improvable space is little, and due to game data magnanimity, used herein is K-Means clustering algorithm.
The effect of proposed algorithm is that suitable game item is recommended suitable player.Proposed algorithm comprises again the multiple technologies such as association analysis, collaborative filtering, matrix decomposition, and proposed algorithm is at present still in development improves.That comparatively popular is project-based collaborative filtering (ItemCF) at present.ItemCF algorithm comes into vogue from (about calendar year 2001) after the paper and Patent Publication of Amazon.But because the stage property of online game is all virtual objects, the data characteristics of online game is very different with the data of electric business again, the proposed algorithm of electric business is copyed everything mechanically on data of network game and must cause discomfort.Disclosed ItemCF algorithm is all the article (Item) bought based on user at present, first the distance matrix between article and article is calculated, then new article (Item) are recommended to each user, this computing method are applied directly on data of network game and there is following drawback: the stage property kind of (1) massively multiplayer game is extremely many, distance matrix between the article of structure and article is also inevitable huge, and this can greatly increase later calculated amount.(2) there is the function of a lot of stage property identical in game and can phase trans-substitution, the player buying this kind of stage property should be extremely similar, the stage property recommended to them also should be similar, but calculate according to current disclosed method, the distance between stage property that these functions are identical can be made on the contrary very large, thus give the stage property very different that the player buying this stage property recommends.Therefore, need to improve existing ItemCF algorithm, to solve above-mentioned drawback.
Summary of the invention
The object of the invention is to solve above-mentioned two kinds of drawbacks ItemCF algorithm being applied directly to data of network game and producing, the stage property for network game game is recommended to provide a kind of suitable recommend method.
To achieve these goals, game item personalized recommendation method of the present invention, comprises the following steps:
(1) data extracted, clean, process;
(2) data are mapped to lower dimensional space from the input space;
(3) data are classified, player is divided into different classes of;
(4) for each player recommends the stage property of his most probable purchase.
Wherein, extracting data in described step (1) is from data warehouse, extract player attribute data, logon data, playing method data, transaction data, load value data, consumption data.
Wherein, in described step (1) to data clean be to extract data be described statistics, that finds out data repeats record, missing values, exceptional value, and replaces accordingly or deletion action.
Wherein, carrying out processing to data in step (1) is change data.
Wherein, to data carry out conversion comprise data are normal standardized, data are asked logarithm.
Wherein, described step (2) comprises usage factor analytic approach further and carries out dimensionality reduction to the attribute data of player after processing, logon data, playing method data, transaction data, load value data and consumption data.
Wherein, described step (3) comprises further:
1) use K-Means algorithm, the data after factorial analysis dimensionality reduction are classified, player is divided into several classifications;
2) use K-Means algorithm, the consumption data after processing is classified, therefrom extracts several main purchasing model;
3) intersected by the classification results of above-mentioned two kinds of K-Means algorithms, whole user can be divided into several colonies, the attribute of the player of each colony inside, game behavior, purchasing model are more similar.
Wherein, described step (4) comprises further:
1) each extraction user group from several colonies in order, carries out the following steps operation to this intragroup player;
2) calculate the distance matrix between the stage property class of player and stage property detail, this distance can be measured by Jaccard similarity;
3) by the known value of each player in each stage property class, be multiplied by distance matrix, obtain the preference degree of each player to each detailed stage property;
4) determine the stage property recommended, each player is sorted, to the stage property that this player recommends rank forward according to the fancy grade of each detailed stage property.
Adopt game item personalized recommendation method of the present invention, can the whole computation process of definite network game item proposed algorithm; Compared with existing ItemCF algorithm, significantly reduce the calculated amount of proposed algorithm; For this specific industry of online game, compared with existing ItemCF algorithm, the accuracy that stage property is recommended can be significantly improved.
Other features and advantages of the present invention will be set forth in the following description, and, partly become apparent from instructions, or understand by implementing the present invention.
Accompanying drawing explanation
Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions, and together with embodiments of the present invention, for explaining the present invention, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is according to game item personalized recommendation method process flow diagram of the present invention;
Fig. 2 is according to game item personalized recommendation method embodiment 1 workflow diagram of the present invention.
Embodiment
Below in conjunction with accompanying drawing, embodiments of the invention are described, should be appreciated that preferred embodiment described herein is only for instruction and explanation of the present invention, is not intended to limit the present invention.
Fig. 1 is according to game item personalized recommendation method process flow diagram of the present invention.Below with reference to Fig. 1, game item personalized recommendation method of the present invention is described in detail.
Step 101, extracts data, cleans, processes, and becomes to be applicable to the form analyzed by data preparation.Wherein, be from data warehouse, extract required data to the extraction of data, comprise the attribute data of player, logon data, playing method data, transaction data, load value data, consumption data; To the cleaning of data be to extract data be described statistics, that finds out data repeats record, missing values, exceptional value, and replaces accordingly or deletion action; The conversions to data to the processing of data, such as, data are normal standardized, data are asked logarithm, specifically adopt that data mart modeling mode to be determined by adopted algorithm.
Data are mapped to a lower dimensional space from the input space by linear or nonlinear transformation by step 102, thus an acquisition low-dimensional of compacting about former data set represents.Usage factor analytic approach, dimensionality reduction is carried out to the player attributes data after processing, logon data, playing method data, transaction data, load value data, high-dimensional data variable is comprehensively become a few factor, and this few factor contains most information of raw data.
Step 103, is divided into inside same class by similar player, makes the player in the same group similar as far as possible, and the player's difference between different groups is remarkable as far as possible.Comprise following three steps:
(1) use K-Means algorithm, classify, player is divided into several classifications to the data after factorial analysis dimensionality reduction, for the ease of later explanation, hypothesis is divided into M class herein;
(2) use K-Means algorithm, classify, therefrom extract several main purchasing model to the consumption data after processing, for the ease of later explanation, hypothesis is divided into N class herein;
(3) intersected by the classification results of above-mentioned two kinds of K-Means algorithms, whole user can be divided into M*N colony, the attribute of the player of each colony inside, game behavior, purchasing model are more similar.
Step 104, utilizes the ItemCF algorithm after improvement, to the stage property that each player recommends his most probable to buy.
Wherein, be mainly reflected in adjust the distance in the improvement of matrix to the improvement of ItemCF algorithm, the enforcement of the ItemCF algorithm after improvement specifically comprises the following steps:
(1) each extraction user group from M*N colony in order, carries out the following steps operation to this intragroup player.
(2) calculate the distance matrix between the stage property class of player and stage property detail, this distance can be measured by Jaccard similarity, and the form of matrix is as follows:
Table 1: the distance matrix between stage property class and stage property detail
Stage property 1 | Stage property 2 | Stage property 3 | Stage property 4 | Stage property 5 | Stage property 6 | …… | |
Stage property class 1 | d 11 | d 12 | d 13 | d 14 | d 15 | d 16 | …… |
Stage property class 2 | d 21 | d 22 | d 23 | d 24 | d 25 | d 26 | …… |
Stage property class 3 | d 31 | d 32 | d 33 | d 34 | d 35 | d 36 | …… |
Stage property class 4 | d 41 | d 42 | d 43 | d 44 | d 45 | d 46 | …… |
…… | …… | …… | …… | …… | …… | …… | …… |
The difference of the distance matrix of this matrix and conventional ItemCF algorithm is: the distance matrix of conventional ItemCF algorithm is the distance matrix between detailed stage property and detailed stage property, suppose there are 5000 kinds of stage properties, so conventional ItemCF algorithm just requires the matrix of a 5000*5000; And herein, suppose stage property identical for functions a large amount of in game to be merged into 30 classes, so, the matrix of a 30*5000 only need be calculated.This calculates each player in each detailed stage property during fancy grade below, can reduce calculated amount in a large number, and can significantly improve the accuracy of recommendation.
(3) by the known value of each player in each stage property class, be multiplied by distance matrix above, obtain the preference degree of each player to each detailed stage property.
(4) determine the stage property recommended, each player is sorted, to the stage property that this player recommends rank forward according to the fancy grade of each detailed stage property.
Embodiment 1
Below in conjunction with embodiment 1, game item personalized recommendation method of the present invention is described in detail:
Step 201, extracts data,
(1) load value data: player's recharge amount of nearest 6 months.
(2) logon data: player's login number of days of nearest 3 months, online total duration.
(3) level data: role hierarchy, life caste that player is current
(4) playing method data: player participates in the number of times of often kind of playing method for nearest 3 months
(5) transaction data: nearest 3 months, the transaction count in game between player and transaction official silver
(6) consumption data: player buys quantity and the amount of money of often kind of stage property for nearest 3 months
Step 202, cleans (only for supplement with money, log in, grade, playing method, transaction data) to data,
(1) record of N_EXT_ID=-1 is removed.
(2) record of N_USER_ID=0 is removed.
(3) player logged in less than 5 days is removed in 3 months.
(4) remove grade lower than 6 player.
(5) districts have multiple role's, and retain the role that the highest grade, remaining falls clearly.
(6) missing values replaces with 0.
Step 203, cleans (only for consumption data) consumption data,
(1) remove and repeat record
(2) the basic stage property that everybody buys is removed
(3) stage property that basic no one buys is removed
Step 204, makes data normalization,
(1) to supplementing with money, log in, playing method, transaction data ask logarithm.Grade and consumption data remain unchanged.
(2) above-mentioned data transformations is become standardized normal distribution.
Step 205, extracts the player characteristic factor,
By factor analysis, to processed above supplement with money, log in, grade, playing method, transaction data carry out factorial analysis, obtains 6 player characteristic factors.
Step 206, to the classification of player behavioural characteristic,
Do K-Means algorithm by the factor of 6 above to classify to player, player is polymerized to 8 classes, be equivalent to player is divided into 8 kinds of behavior patterns.After original data, add row, preserve the playing method classification belonging to each player, and output class center.
Step 207, to consumption behaviour classification,
Use K-Means algorithm, use consumption data that player is polymerized to 50 classes, be equivalent to the consumer behavior of player to be divided into 50 kinds of consumption modes, player similar for consumer behavior is divided at same class.After original data, add row, preserve the consumption classification belonging to each player, and output class center.
Step 208, to users classification,
The classification of above-mentioned player behavioural characteristic and consumer behavior classification are intersected, obtains the combination of 8*50=400 kind altogether, be equivalent to player is divided into 400 colonies.The game behavioural characteristic of the player in the same group is all very similar with consumer behavior feature.
Step 209, stage property is recommended,
(1) the 1st, the 2nd is read respectively ..., the 400th groups of users, carries out following calculating respectively to the player of each group, for the player of the 1st group.
(2) take out the stage property record that the player in this group buys, the stage property of existing detail in this record, also has the classification to stage property.
(3) the Jaccard similarity matrix between stage property class and stage property detail is calculated.(matrix form is in table 1)
(4) by the known value of each player in each stage property class, be multiplied by Jaccard similarity matrix above, obtain the preference degree of each player to each detailed stage property.
(5) each player is sorted to the fancy grade of each detailed stage property according to him, to 10 kinds of stage properties that this player recommends rank forward.
Game item personalized recommendation method of the present invention, by the improvement to collaborative filtering, the stage property making it be more suitable for online game is recommended, algorithm after improvement, significantly can not only reduce calculated amount, the accuracy that online game stage property is recommended can also be significantly improved, have remarkable result for the experience improving network gaming user.
One of ordinary skill in the art will appreciate that: the foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, although with reference to previous embodiment to invention has been detailed description, for a person skilled in the art, it still can be modified to the technical scheme that foregoing embodiments is recorded, or carries out equivalent replacement to wherein portion of techniques feature.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (8)
1. a game item personalized recommendation method, is characterized in that, comprises the following steps:
(1) data extracted, clean, process;
(2) data are mapped to lower dimensional space from the input space;
(3) data are classified, player is divided into different classes of;
(4) for each player recommends the stage property of his most probable purchase.
2. game item personalized recommendation method according to claim 1, it is characterized in that, extracting data in described step (1) is from data warehouse, extract player attribute data, logon data, playing method data, transaction data, load value data, consumption data.
3. game item personalized recommendation method according to claim 1, it is characterized in that, in described step (1) to data clean be to extract data be described statistics, that finds out data repeats record, missing values, exceptional value, and replaces accordingly or deletion action.
4. game item personalized recommendation method according to claim 1, is characterized in that, carrying out processing to data in step (1) is change data.
5. game item personalized recommendation method according to claim 4, is characterized in that, to data carry out conversion comprise data are normal standardized, data are asked logarithm.
6. game item personalized recommendation method according to claim 1, it is characterized in that, described step (2) comprises usage factor analytic approach further and carries out dimensionality reduction to the attribute data of player after processing, logon data, playing method data, transaction data, load value data and consumption data.
7. game item personalized recommendation method according to claim 1, is characterized in that, described step (3) comprises further:
1) use K-Means algorithm, the data after factorial analysis dimensionality reduction are classified, player is divided into several classifications;
2) use K-Means algorithm, the consumption data after processing is classified, therefrom extracts several main purchasing model;
3) intersected by the classification results of above-mentioned two kinds of K-Means algorithms, whole user can be divided into several colonies, the attribute of the player of each colony inside, game behavior, purchasing model are more similar.
8. game item personalized recommendation method according to claim 1, is characterized in that, described step (4) comprises further:
1) each extraction user group from several colonies in order, carries out the following steps operation to this intragroup player;
2) calculate the distance matrix between the stage property class of player and stage property detail, this distance can be measured by Jaccard similarity;
3) by the known value of each player in each stage property class, be multiplied by distance matrix, obtain the preference degree of each player to each detailed stage property;
4) determine the stage property recommended, each player is sorted, to the stage property that this player recommends rank forward according to the fancy grade of each detailed stage property.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510793393.0A CN105447126A (en) | 2015-11-17 | 2015-11-17 | Game prop personalized recommendation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510793393.0A CN105447126A (en) | 2015-11-17 | 2015-11-17 | Game prop personalized recommendation method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105447126A true CN105447126A (en) | 2016-03-30 |
Family
ID=55557303
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510793393.0A Pending CN105447126A (en) | 2015-11-17 | 2015-11-17 | Game prop personalized recommendation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105447126A (en) |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105894350A (en) * | 2016-03-29 | 2016-08-24 | 南京大学 | Method for recommending mobile game props based on multi-example multiple labeling learning |
CN105939360A (en) * | 2016-06-24 | 2016-09-14 | 北京奇虎科技有限公司 | Equipping method and device of game equipment |
CN106411846A (en) * | 2016-08-30 | 2017-02-15 | 江苏名通信息科技有限公司 | Connecting system and method for network game and e-commerce |
CN106512405A (en) * | 2016-12-06 | 2017-03-22 | 腾讯科技(深圳)有限公司 | Method and device for acquiring plug-in resource of virtual object |
CN106779933A (en) * | 2016-12-06 | 2017-05-31 | 腾讯科技(深圳)有限公司 | A kind of virtual item recommends method and client |
CN106730846A (en) * | 2016-11-10 | 2017-05-31 | 北京像素软件科技股份有限公司 | The data processing method and device of one attribute stage property |
CN107154930A (en) * | 2017-03-31 | 2017-09-12 | 武汉斗鱼网络科技有限公司 | A kind of method and system for testing leak |
CN108090800A (en) * | 2017-11-27 | 2018-05-29 | 珠海金山网络游戏科技有限公司 | A kind of game item method for pushing and device based on player's consumption potentiality |
WO2018103516A1 (en) * | 2016-12-06 | 2018-06-14 | 腾讯科技(深圳)有限公司 | Method of acquiring virtual resource of virtual object, and client |
WO2018126885A1 (en) * | 2017-01-04 | 2018-07-12 | 腾讯科技(深圳)有限公司 | Game data processing method |
CN110052033A (en) * | 2019-04-26 | 2019-07-26 | 腾讯科技(深圳)有限公司 | A kind of game article requesting method, device, server and storage medium |
CN110465086A (en) * | 2018-05-11 | 2019-11-19 | 腾讯科技(深圳)有限公司 | A kind of game articles methods of exhibiting, device and storage medium |
CN110559664A (en) * | 2019-09-19 | 2019-12-13 | 湘潭大学 | game hero outgoing recommendation method and system based on multi-objective optimization |
CN110807150A (en) * | 2019-10-14 | 2020-02-18 | 腾讯科技(深圳)有限公司 | Information processing method and device, electronic equipment and computer readable storage medium |
CN111061949A (en) * | 2019-12-03 | 2020-04-24 | 深圳市其乐游戏科技有限公司 | Prop recommendation method, recommendation device and computer-readable storage medium |
CN111111210A (en) * | 2019-12-23 | 2020-05-08 | 北京像素软件科技股份有限公司 | Article sorting method and device for online game |
CN111672122A (en) * | 2020-05-29 | 2020-09-18 | 腾讯科技(深圳)有限公司 | Interface display method, device, terminal and storage medium |
CN111695941A (en) * | 2020-06-15 | 2020-09-22 | 广州探途网络技术有限公司 | Commodity transaction website data analysis method and device and electronic equipment |
CN111729301A (en) * | 2020-06-15 | 2020-10-02 | 北京智明星通科技股份有限公司 | Method and device for recommending props in breakthrough game and game terminal |
CN113450181A (en) * | 2021-05-27 | 2021-09-28 | 广州三七极耀网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101183378A (en) * | 2006-11-14 | 2008-05-21 | 国际商业机器公司 | Method and system for cleansing sequence-based data at query time |
CN101487892A (en) * | 2009-02-23 | 2009-07-22 | 北京航空航天大学 | High-spectrum data dimensionality reduction method based on factor analysis model |
US20090271417A1 (en) * | 2008-04-25 | 2009-10-29 | John Toebes | Identifying User Relationships from Situational Analysis of User Comments Made on Media Content |
CN102226905A (en) * | 2011-05-26 | 2011-10-26 | 北京交通大学 | Statistical analysis and evaluation model for railway emergency management system |
CN102236783A (en) * | 2010-04-29 | 2011-11-09 | 索尼公司 | Method and equipment for detecting abnormal actions and method and equipment for generating detector |
CN102609523A (en) * | 2012-02-10 | 2012-07-25 | 上海视畅信息科技有限公司 | Collaborative filtering recommendation algorithm based on article sorting and user sorting |
CN103279552A (en) * | 2013-06-06 | 2013-09-04 | 浙江大学 | Collaborative filtering recommendation method based on user interest groups |
CN103473375A (en) * | 2013-09-29 | 2013-12-25 | 方正国际软件有限公司 | Data cleaning method and data cleaning system |
CN104408150A (en) * | 2014-12-03 | 2015-03-11 | 天津南大通用数据技术股份有限公司 | Data import/ export method and device adapted to a plurality of data formats of databases |
-
2015
- 2015-11-17 CN CN201510793393.0A patent/CN105447126A/en active Pending
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101183378A (en) * | 2006-11-14 | 2008-05-21 | 国际商业机器公司 | Method and system for cleansing sequence-based data at query time |
US20090271417A1 (en) * | 2008-04-25 | 2009-10-29 | John Toebes | Identifying User Relationships from Situational Analysis of User Comments Made on Media Content |
CN101487892A (en) * | 2009-02-23 | 2009-07-22 | 北京航空航天大学 | High-spectrum data dimensionality reduction method based on factor analysis model |
CN102236783A (en) * | 2010-04-29 | 2011-11-09 | 索尼公司 | Method and equipment for detecting abnormal actions and method and equipment for generating detector |
CN102226905A (en) * | 2011-05-26 | 2011-10-26 | 北京交通大学 | Statistical analysis and evaluation model for railway emergency management system |
CN102609523A (en) * | 2012-02-10 | 2012-07-25 | 上海视畅信息科技有限公司 | Collaborative filtering recommendation algorithm based on article sorting and user sorting |
CN103279552A (en) * | 2013-06-06 | 2013-09-04 | 浙江大学 | Collaborative filtering recommendation method based on user interest groups |
CN103473375A (en) * | 2013-09-29 | 2013-12-25 | 方正国际软件有限公司 | Data cleaning method and data cleaning system |
CN104408150A (en) * | 2014-12-03 | 2015-03-11 | 天津南大通用数据技术股份有限公司 | Data import/ export method and device adapted to a plurality of data formats of databases |
Non-Patent Citations (3)
Title |
---|
(美)里奇: "《推荐系统技术、评估及高效算法》", 31 July 2015 * |
孙慧峰: "基于协同过滤的个性化Web推荐", 《中国博士学位论文全文数据库 信息科技辑》 * |
毛基业: "《管理信息系统基础、应用于方法》", 28 February 2011 * |
Cited By (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105894350A (en) * | 2016-03-29 | 2016-08-24 | 南京大学 | Method for recommending mobile game props based on multi-example multiple labeling learning |
CN105939360B (en) * | 2016-06-24 | 2019-07-09 | 北京奇虎科技有限公司 | The outfit method and device of game article |
CN105939360A (en) * | 2016-06-24 | 2016-09-14 | 北京奇虎科技有限公司 | Equipping method and device of game equipment |
CN106411846A (en) * | 2016-08-30 | 2017-02-15 | 江苏名通信息科技有限公司 | Connecting system and method for network game and e-commerce |
CN106730846A (en) * | 2016-11-10 | 2017-05-31 | 北京像素软件科技股份有限公司 | The data processing method and device of one attribute stage property |
CN106512405A (en) * | 2016-12-06 | 2017-03-22 | 腾讯科技(深圳)有限公司 | Method and device for acquiring plug-in resource of virtual object |
CN106779933A (en) * | 2016-12-06 | 2017-05-31 | 腾讯科技(深圳)有限公司 | A kind of virtual item recommends method and client |
US11020664B2 (en) | 2016-12-06 | 2021-06-01 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for obtaining virtual resource of virtual object |
WO2018103516A1 (en) * | 2016-12-06 | 2018-06-14 | 腾讯科技(深圳)有限公司 | Method of acquiring virtual resource of virtual object, and client |
CN106512405B (en) * | 2016-12-06 | 2019-02-19 | 腾讯科技(深圳)有限公司 | A kind of method and device of the plug-in resource acquisition of virtual objects |
TWI658854B (en) * | 2016-12-06 | 2019-05-11 | 大陸商騰訊科技(深圳)有限公司 | Methord for obtaining virtual resource for virtual object and client terminal thereof |
WO2018126885A1 (en) * | 2017-01-04 | 2018-07-12 | 腾讯科技(深圳)有限公司 | Game data processing method |
CN107154930A (en) * | 2017-03-31 | 2017-09-12 | 武汉斗鱼网络科技有限公司 | A kind of method and system for testing leak |
CN108090800A (en) * | 2017-11-27 | 2018-05-29 | 珠海金山网络游戏科技有限公司 | A kind of game item method for pushing and device based on player's consumption potentiality |
CN110465086A (en) * | 2018-05-11 | 2019-11-19 | 腾讯科技(深圳)有限公司 | A kind of game articles methods of exhibiting, device and storage medium |
CN110052033A (en) * | 2019-04-26 | 2019-07-26 | 腾讯科技(深圳)有限公司 | A kind of game article requesting method, device, server and storage medium |
CN110052033B (en) * | 2019-04-26 | 2022-03-25 | 腾讯科技(深圳)有限公司 | Game equipment request method, device, server and storage medium |
CN110559664A (en) * | 2019-09-19 | 2019-12-13 | 湘潭大学 | game hero outgoing recommendation method and system based on multi-objective optimization |
CN110807150A (en) * | 2019-10-14 | 2020-02-18 | 腾讯科技(深圳)有限公司 | Information processing method and device, electronic equipment and computer readable storage medium |
CN111061949A (en) * | 2019-12-03 | 2020-04-24 | 深圳市其乐游戏科技有限公司 | Prop recommendation method, recommendation device and computer-readable storage medium |
CN111111210A (en) * | 2019-12-23 | 2020-05-08 | 北京像素软件科技股份有限公司 | Article sorting method and device for online game |
CN111111210B (en) * | 2019-12-23 | 2023-06-27 | 北京像素软件科技股份有限公司 | Article sorting method and device for online game |
CN111672122A (en) * | 2020-05-29 | 2020-09-18 | 腾讯科技(深圳)有限公司 | Interface display method, device, terminal and storage medium |
CN111695941A (en) * | 2020-06-15 | 2020-09-22 | 广州探途网络技术有限公司 | Commodity transaction website data analysis method and device and electronic equipment |
CN111729301A (en) * | 2020-06-15 | 2020-10-02 | 北京智明星通科技股份有限公司 | Method and device for recommending props in breakthrough game and game terminal |
CN113450181A (en) * | 2021-05-27 | 2021-09-28 | 广州三七极耀网络科技有限公司 | Virtual commodity recommendation method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105447126A (en) | Game prop personalized recommendation method | |
CN107562818A (en) | Information recommendation system and method | |
CN108090800B (en) | Game prop pushing method and device based on player consumption potential | |
CN103377250B (en) | Top k based on neighborhood recommend method | |
Barbier et al. | Elephants, economics and ivory | |
CN109583651A (en) | A kind of method and apparatus for insuring electric business platform user attrition prediction | |
Bandyopadhyay et al. | Product recommendation for e-commerce business by applying principal component analysis (PCA) and K-means clustering: benefit for the society | |
CN108153791B (en) | Resource recommendation method and related device | |
Hong et al. | Rapid identification of the optimal product configuration and its parameters based on customer-centric product modeling for one-of-a-kind production | |
CN102254028A (en) | Personalized commodity recommending method and system which integrate attributes and structural similarity | |
CN106503438A (en) | A kind of H RFM user modeling method and system for pharmacy member analysis | |
CN104424339A (en) | Method, device and system for data analysis | |
CN112232930A (en) | E-commerce platform customer segmentation method based on weighted RFM model | |
Dachyar et al. | Loyalty improvement of Indonesian local brand fashion customer based on customer lifetime value (CLV) segmentation | |
Ahn et al. | What makes the difference between popular games and unpopular games? analysis of online game reviews from steam platform using word2vec and bass model | |
CN104050298B (en) | A kind of subscriber segmentation method based on LDA topic model algorithms | |
CN110968670B (en) | Method, device, equipment and storage medium for acquiring attributes of popular commodities | |
CN110020918B (en) | Recommendation information generation method and system | |
CN116362236A (en) | Target word mining method and device and storage medium | |
CN111275459A (en) | Cigarette brand recommendation algorithm based on consumer modeling | |
CN103793504A (en) | Cluster initial point selection method based on user preference and project properties | |
Patil et al. | Study of Customer Segmentation Using K-Means Clustering and RFM Modelling | |
Cho et al. | Effective purchase pattern mining with weight based on FRAT analysis for recommender in e-commerce | |
Hanner et al. | Counting Customers in Mobile Business–The Case of Free to Play | |
Kaur | Customer segmentation using clustering algorithm |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160330 |
|
RJ01 | Rejection of invention patent application after publication |