CN108920512B - Game software scene-based recommendation method - Google Patents
Game software scene-based recommendation method Download PDFInfo
- Publication number
- CN108920512B CN108920512B CN201810546648.7A CN201810546648A CN108920512B CN 108920512 B CN108920512 B CN 108920512B CN 201810546648 A CN201810546648 A CN 201810546648A CN 108920512 B CN108920512 B CN 108920512B
- Authority
- CN
- China
- Prior art keywords
- game software
- frequency
- game
- picture
- word
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention discloses a recommendation method based on a game software scene, which comprises the steps of collecting scores and comments of a user on game software, evaluating texts and pictures in the comments, comprehensively scoring according to results of the scores, the text evaluations and the picture evaluations of the user, and finally pushing the game software according to the comprehensive scores. The method comprehensively scores the game software from three aspects of user scoring, text evaluation and picture evaluation, and is favorable for realizing the accuracy of software ranking.
Description
Technical Field
The invention relates to the field of game software, in particular to a recommendation method based on game software scenes.
Background
With the comprehensive popularization of networks, in order to improve the convenience of life of people, application software is in endless, and various aspects of life are related, clothes and eating houses are supported by corresponding software, and how to find software meeting the requirements of users from a large amount of software is more and more important. Due to the large amount of application software, for a user, the more choices, the more difficult it is to select a proper APP from a large amount of application software; in addition, when the merchant is promoting own application software, how to find potential users and grasp the needs of the public users are also important. Therefore, under the increasingly competitive environment, how enterprises produce software meeting the requirements of users is more important to effectively develop the designed software. In addition, the scoring standard provided by the current software sales platform is also a score evaluation decision, and the importance of text evaluation and picture evaluation is ignored. Developers cannot know the real requirements of users in the process of developing software, and the method also becomes a significant problem of software development.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a recommendation method based on game software scenes and considering user scoring, text evaluation and picture evaluation at the same time.
The technical scheme is as follows: the recommendation method based on the game software scene collects the scores and comments of the users on the game software, evaluates the texts and pictures in the comments, comprehensively scores the results of the scores, the text evaluations and the picture evaluations of the users, and finally pushes the game software according to the comprehensive scores.
Further, data preprocessing is performed before text evaluation and picture evaluation.
Further, the text evaluation comprises the steps of:
s1.1: for one game software, taking out the contents of all comments needing text evaluation;
s1.2: calculating the word frequency and the reverse file frequency corresponding to each word in each comment, and then calculating the product of the word frequency and the reverse file frequency;
s1.3: defining a matrix, wherein elements in the matrix are the product of the word frequency of a certain word in a certain comment and the frequency of a reverse file;
s1.4: removing the product of the word frequency smaller than K and the reverse file frequency from the product of all the word frequencies and the reverse file frequencies; k is a threshold value for judging whether the words are meaningful or not;
s1.5: superposing and summing the product of the word frequency of the same word and the frequency of the reverse file, and then taking out N words arranged in the front;
s1.6: and summing the product of the word frequency of the first N words and the reverse file frequency to obtain the score of the text evaluation of the game software.
Further, the text evaluation further comprises steps S1.7 and S1.8:
s1.7: for other game software, taking out N words arranged in the front according to the processes of the steps S1.1-S1.5;
s1.8: traversing the first N words taken out by the first game software by the first N words taken out by the other game software to obtain the product of the word frequency and the reverse file frequency of the first N words of the first game software in the other game software, sequencing the product of the word frequency and the reverse file frequency of the same word in the other game software, and recommending the other game software to the user according to the sequencing order.
Further, the text evaluation further comprises steps S1.9 and S1.10:
s1.9: for other game software, taking out N words arranged in the front according to the processes of the steps S1.1-S1.5;
s1.10: and calculating the relevance of the first N words of different game software to obtain the requirement of the user on the game software.
Further, the picture evaluation comprises the following steps:
s2.1: setting a reference picture;
s2.2: setting a symbolic region by referring to the picture, comparing the picture to be evaluated with the set symbolic region, and identifying;
s2.3: if not, the game does not belong to the scene of the game, and the step S2.5 is carried out; if so, performing step S2.4;
s2.4: setting a score threshold value, scoring the image recognition result, and then judging as follows: multiplying the score of the text rating by 1 if the score is below a score threshold; multiplying the score of the text rating by a weight n if the score is not below a score threshold, 0< n < 1;
s2.5: and classifying and storing according to the picture content, and providing the picture content for game software developers.
Further, the reference picture in the step S2.1 is a scene graph in the game or an interface graph of the game.
Further, in step S2.4, if the score is lower than the score threshold, the content of the text evaluation is also saved in the database.
Further, in step S2.2, before the identification, the picture to be evaluated is preprocessed, including image acquisition, image enhancement, image restoration, image coding and compression, and image segmentation.
Has the advantages that: the invention discloses a recommendation method based on game software scenes, which comprehensively scores game software from three aspects of user scoring, text evaluation and picture evaluation and is beneficial to realizing the accuracy of software ranking.
Drawings
FIG. 1 is a general block diagram of an embodiment of the present invention;
FIG. 2 is a refinement scoring process in accordance with an embodiment of the present invention;
FIG. 3 is a flow diagram of text evaluation in an embodiment of the present invention;
FIG. 4 is a flow chart of image pre-processing in an embodiment of the present invention;
fig. 5 is a flowchart of picture evaluation in the embodiment of the present invention.
Detailed Description
The specific embodiment discloses a recommendation method based on a game software scene, which is used for collecting scores and comments of a user on game software, preprocessing data and removing worthless data, so that the obtained data is more real and is close to a game. As shown in fig. 1 and fig. 2, the text and the picture in the comment are evaluated, the comprehensive scoring is performed according to the results of the user scoring, the text evaluation and the picture evaluation, and finally, the game software is pushed according to the comprehensive scoring.
As shown in fig. 3, the text evaluation includes the following steps:
s1.1: for one game software, taking out the contents of all comments needing text evaluation;
s1.2: calculating the word frequency and the reverse file frequency corresponding to each word in each comment, and then calculating the product of the word frequency and the reverse file frequency;
s1.3: defining a matrix, wherein elements in the matrix are the product of the word frequency of a certain word in a certain comment and the frequency of a reverse file;
s1.4: removing the product of the word frequency smaller than K and the reverse file frequency from the product of all the word frequencies and the reverse file frequencies; k is a threshold value for judging whether the words are meaningful or not;
s1.5: superposing and summing the product of the word frequency of the same word and the frequency of the reverse file, and then taking out N words arranged in the front;
s1.6: summing the product of the word frequency and the reverse file frequency of the first N words to obtain the score of the text evaluation of the game software;
s1.7: for other game software, taking out N words arranged in the front according to the processes of the steps S1.1-S1.5;
s1.8: traversing the first N words taken out by the first game software by the first N words taken out by other game software to obtain the product of the word frequency and the reverse file frequency of the first N words of the first game software in other game software, sequencing the product of the word frequency and the reverse file frequency of the same word in other game software, and recommending other game software to the user according to the sequencing order;
s1.9: and calculating the relevance of the first N words of different game software by using an Apriori association rule method to obtain the requirement of the user on the game software.
Wherein, the word frequency can be expressed by TF,
where M represents the number of occurrences of a word in the comment and M represents the total number of words in the comment. The inverse file frequency may be represented by an IDF,
u is the total number of comments, and U is the number of comments that contain a word.
As shown in fig. 5, the picture evaluation includes the following steps:
s2.1: setting a reference picture; the reference picture is a scene picture in the game or an interface picture of the game;
s2.2: setting a landmark region with reference to the picture, and preprocessing the picture to be evaluated, as shown in fig. 4, the method includes the following processes: image acquisition: extracting the picture content from the comment content; image enhancement: in order to highlight interesting parts in the images and make the main body structures of the images more definite, the images must be improved, the image enhancement improves the definition and the quality of the images, and the outlines of objects in the images are clearer and the details are more obvious; image restoration: image restoration is also called image restoration, wherein the image is blurred due to the influence of environmental noise, image blurring caused by movement, light intensity and other reasons when the image is acquired, the image needs to be restored for extracting a clearer image, and the image restoration mainly adopts a filtering method to restore an original image from a degraded image; image coding and compression: the digital image has the remarkable characteristics that the data volume is huge, a quite large storage space is required to be occupied, but the network bandwidth and the large-capacity memory based on a computer cannot process, store and transmit the data image, and in order to transmit the image or video in a network environment quickly and conveniently, the image must be coded and compressed, and the image coding and compression technology can reduce the redundant data volume and the memory capacity of the image, improve the image transmission speed and shorten the processing time; image segmentation: the image segmentation is to divide the image into a plurality of sub-regions which are not overlapped and have respective characteristics, each region is a continuous set of pixels, the characteristics can be the color, the shape, the gray scale, the texture and the like of the image, and the image segmentation lays a foundation for further image recognition, analysis and understanding.
Then, comparing the picture to be evaluated with the set symbolic region, and identifying;
s2.3: if not, the game does not belong to the scene of the game, and the step S2.5 is carried out; if so, performing step S2.4;
s2.4: setting a score threshold value, scoring the image recognition result, and then judging as follows: if the score is lower than the score threshold value, multiplying the score of the text evaluation by 1, and storing the content of the text evaluation in a database as a scheme for software modification; multiplying the score of the text rating by a weight n if the score is not below a score threshold, 0< n < 1;
s2.5: and classifying and storing the pictures according to the contents of the pictures, and providing the pictures for game software developers, wherein the developers can find effective information in the pictures according to needs.
It can be seen that the present embodiment has the following beneficial effects:
1. and the fine scoring is convenient for the user to know the scoring of the software more carefully, and the safer guarantee for the player to download the game is provided.
2. The deficiency of the game can be known through evaluating the content, the true idea of the player can be known, the functions which the game should have at the present stage are integrated, and the enterprise can conveniently develop the game and can reasonably position the game.
3. The game is pushed reasonably according to the preference, not only limited to the characteristics of the game, but also pushed according to the visual feeling brought by the game, so that the game is more personalized, and the types of recommended games are various.
Claims (6)
1. A recommendation method based on game software scenes is characterized in that: collecting scores and comments of a user on the game software, evaluating texts and pictures in the comments, comprehensively scoring according to results of the user scores, the text evaluations and the picture evaluations, and finally pushing the game software according to the comprehensive scores;
wherein the text evaluation comprises the steps of:
s1.1: for one game software, taking out the contents of all comments needing text evaluation;
s1.2: calculating the word frequency and the reverse file frequency corresponding to each word in each comment, and then calculating the product of the word frequency and the reverse file frequency;
s1.3: defining a matrix, wherein elements in the matrix are the product of the word frequency of a certain word in a certain comment and the frequency of a reverse file;
s1.4: removing the product of the word frequency smaller than K and the reverse file frequency from the product of all the word frequencies and the reverse file frequencies; k is a threshold value for judging whether the words are meaningful or not;
s1.5: superposing and summing the product of the word frequency of the same word and the frequency of the reverse file, and then taking out N words arranged in the front;
s1.6: summing the product of the word frequency and the reverse file frequency of the first N words to obtain the score of the text evaluation of the game software;
s1.7: for other game software, taking out N words arranged in the front according to the processes of the steps S1.1-S1.5;
s1.8: traversing the first N words taken out by the first game software by the first N words taken out by other game software to obtain the product of the word frequency and the reverse file frequency of the first N words of the first game software in other game software, sequencing the product of the word frequency and the reverse file frequency of the same word in other game software, and recommending other game software to the user according to the sequencing order;
the picture evaluation comprises the following steps:
s2.1: setting a reference picture;
s2.2: setting a symbolic region by referring to the picture, comparing the picture to be evaluated with the set symbolic region, and identifying;
s2.3: if not, the game does not belong to the scene of the game, and the step S2.5 is carried out; if so, performing step S2.4;
s2.4: setting a score threshold value, scoring the image recognition result, and then judging as follows: multiplying the score of the text rating by 1 if the score is below a score threshold; multiplying the score of the text rating by a weight n if the score is not below a score threshold, 0< n < 1;
s2.5: and classifying and storing according to the picture content, and providing the picture content for game software developers.
2. The game software scene-based recommendation method according to claim 1, wherein: and performing data preprocessing before text evaluation and picture evaluation.
3. The game software scene-based recommendation method according to claim 1, wherein: the text evaluation further comprises steps S1.9 and S1.10:
s1.9: for other game software, taking out N words arranged in the front according to the processes of the steps S1.1-S1.5;
s1.10: and calculating the relevance of the first N words of different game software to obtain the requirement of the user on the game software.
4. The game software scene-based recommendation method according to claim 1, wherein: the reference picture in the step S2.1 is a scene graph in a game or an interface graph of the game.
5. The game software scene-based recommendation method according to claim 1, wherein: in step S2.4, if the score is lower than the score threshold, the content of the text rating is also saved in the database.
6. The game software scene-based recommendation method according to claim 1, wherein: in step S2.2, before the identification, the picture to be evaluated is preprocessed, including image acquisition, image enhancement, image restoration, image encoding and compression, and image segmentation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810546648.7A CN108920512B (en) | 2018-05-31 | 2018-05-31 | Game software scene-based recommendation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810546648.7A CN108920512B (en) | 2018-05-31 | 2018-05-31 | Game software scene-based recommendation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108920512A CN108920512A (en) | 2018-11-30 |
| CN108920512B true CN108920512B (en) | 2021-12-28 |
Family
ID=64419855
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810546648.7A Active CN108920512B (en) | 2018-05-31 | 2018-05-31 | Game software scene-based recommendation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108920512B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114202389A (en) * | 2021-10-27 | 2022-03-18 | 杭州拼便宜网络科技有限公司 | User evaluation control method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105279148A (en) * | 2015-10-19 | 2016-01-27 | 昆明理工大学 | User review consistency judgment method of APP (Application) software |
| CN105975453A (en) * | 2015-12-01 | 2016-09-28 | 乐视网信息技术（北京）股份有限公司 | Method and device for comment label extraction |
| CN106202519A (en) * | 2016-07-22 | 2016-12-07 | 桂林电子科技大学 | A kind of combination user comment content and the item recommendation method of scoring |
| CN107180021A (en) * | 2016-03-09 | 2017-09-19 | 北京京东尚科信息技术有限公司 | A kind of data processing method, system and its server |
| CN107993126A (en) * | 2017-11-30 | 2018-05-04 | 武汉理工大学 | It is a kind of that the improvement collaborative filtering method for correcting user's scoring is commented on based on excavation |
-
2018
- 2018-05-31 CN CN201810546648.7A patent/CN108920512B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105279148A (en) * | 2015-10-19 | 2016-01-27 | 昆明理工大学 | User review consistency judgment method of APP (Application) software |
| CN105975453A (en) * | 2015-12-01 | 2016-09-28 | 乐视网信息技术（北京）股份有限公司 | Method and device for comment label extraction |
| CN107180021A (en) * | 2016-03-09 | 2017-09-19 | 北京京东尚科信息技术有限公司 | A kind of data processing method, system and its server |
| CN106202519A (en) * | 2016-07-22 | 2016-12-07 | 桂林电子科技大学 | A kind of combination user comment content and the item recommendation method of scoring |
| CN107993126A (en) * | 2017-11-30 | 2018-05-04 | 武汉理工大学 | It is a kind of that the improvement collaborative filtering method for correcting user's scoring is commented on based on excavation |
Non-Patent Citations (1)
| Title |
|---|
| "无图无真相？图片和文字网络评论对服务产品消费者态度的影响";杨颖等;《心理学探新》;20140228;论文第1，3节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108920512A (en) | 2018-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Ma et al. | No-reference retargeted image quality assessment based on pairwise rank learning | |
| CN110019943B (en) | Video recommendation method and device, electronic equipment and storage medium | |
| CN110807757B (en) | Image quality evaluation method and device based on artificial intelligence and computer equipment | |
| CN112749608A (en) | Video auditing method and device, computer equipment and storage medium | |
| CN110751649B (en) | Video quality evaluation method and device, electronic equipment and storage medium | |
| US20230353828A1 (en) | Model-based data processing method and apparatus | |
| Su et al. | Image inpainting for random areas using dense context features | |
| Siahaan et al. | Semantic-aware blind image quality assessment | |
| CN111783712A (en) | Video processing method, device, equipment and medium | |
| CN111177470A (en) | Video processing method, video searching method and terminal equipment | |
| CN111078940A (en) | Image processing method, image processing device, computer storage medium and electronic equipment | |
| CN115080865B (en) | E-commerce data operation management system based on multidimensional data analysis | |
| CN113761253A (en) | Video tag determination method, device, equipment and storage medium | |
| CN111432206A (en) | Video definition processing method and device based on artificial intelligence and electronic equipment | |
| CN114588633B (en) | Content recommendation method | |
| CN112950579A (en) | Image quality evaluation method and device and electronic equipment | |
| CN112132766A (en) | Image restoration method and device, storage medium and electronic device | |
| CN108920512B (en) | Game software scene-based recommendation method | |
| CN115294162B (en) | Target identification method, device, equipment and storage medium | |
| CN116416436A (en) | Video and audio feature extraction method and processing system based on neural network | |
| CN116261009A (en) | Video detection method, device, equipment and medium for intelligently converting video audience | |
| CN114782720A (en) | Method, device, electronic device, medium, and program product for determining matching of document | |
| Chang et al. | Image Quality Evaluation Based on Gradient, Visual Saliency, and Color Information | |
| CN113592765A (en) | Image processing method, device, equipment and storage medium | |
| Tan et al. | Multi-scale attentive residual network for single image deraining |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |