CN108920512A - A kind of recommended method based on Games Software scene - Google Patents
A kind of recommended method based on Games Software scene Download PDFInfo
- Publication number
- CN108920512A CN108920512A CN201810546648.7A CN201810546648A CN108920512A CN 108920512 A CN108920512 A CN 108920512A CN 201810546648 A CN201810546648 A CN 201810546648A CN 108920512 A CN108920512 A CN 108920512A
- Authority
- CN
- China
- Prior art keywords
- word
- frequency
- games
- games software
- picture
- 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
- 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
Abstract
The invention discloses a kind of recommended methods based on Games Software scene, acquire marking and comment of the user to Games Software, to in comment text and picture evaluate, comprehensive marking is carried out in conjunction with the result of user's marking, text evaluation and picture evaluation, finally Games Software is pushed according to comprehensive marking.Comprehensive marking is carried out to Games Software in terms of the present invention gives a mark from user, text is evaluated and picture evaluates these three, is advantageously implemented the precision of software ranking.
Description
Technical field
The present invention relates to Games Software fields, more particularly to the recommended method based on Games Software scene.
Background technique
Comprehensive with network popularizes, and in order to improve the convenient degree of people's life, application software emerges one after another, is related to giving birth to
Various aspects in work, clothing, food, lodging and transportion -- basic necessities of life have corresponding software to support, how to find from a large amount of software and meet user's need
The software asked is more and more important.A large amount of appearance of application software, for a user, selection are more, from the application software of magnanimity
The middle suitable APP of selection is more difficult；In addition, how businessman finds potential user when promoting the application software of oneself, and
And hold the demand of public users and most important.Therefore, in the environment of increasingly dog-eat-dog, how enterprise produces full
The software of sufficient user demand, the software of design is effectively expanded, and it is more important to come.In addition, current software sales platform, mentions
The standards of grading of confession are also that score evaluation determines, have ignored text evaluation, the importance of picture evaluation.Developer is in exploitation software
During, the real demand of user can not be understood, also become a significant problem of software development.
Summary of the invention
Goal of the invention：Consider what user's marking, text evaluation and picture were evaluated simultaneously the object of the present invention is to provide a kind of
Recommended method based on Games Software scene.
Technical solution：Recommended method of the present invention based on Games Software scene acquires user to Games Software
Marking and comment, in comment text and picture evaluate, in conjunction with user marking, text evaluation and picture evaluation knot
Fruit carries out comprehensive marking, is finally pushed according to comprehensive marking to Games Software.
Further, before carrying out text evaluation and picture evaluation, advanced line number Data preprocess.
Further, the text evaluation includes the following steps：
S1.1：For a Games Software, the content for needing to carry out all comments of text evaluation is taken out；
S1.2：Word frequency and reverse document-frequency corresponding to each word in every comment are calculated, word is then calculated
The product of frequency and reverse document-frequency；
S1.3：A matrix is defined, the element in matrix is the word frequency of some word and reverse file frequency in certain comment
The product of rate；
S1.4：For the product of all word frequency and reverse document-frequency, remove less than the word frequency of K and reverse document-frequency
Product；Wherein, K is the whether significant threshold value of grammatical term for the character；
S1.5：The word frequency of same word and reverse document-frequency product are overlapped summation, then takes out and comes front
N number of word；
S1.6：Word frequency and reverse document-frequency product to the top n word of taking-up are summed, and the Games Software is obtained
Text evaluation score.
Further, the text evaluation further includes step S1.7 and S1.8：
S1.7：For other Games Softwares, the N number of word for coming front is also taken out according to the process of step S1.1-S1.5；
S1.8：The top n word of other Games Softwares taking-up is traversed with the top n word that first Games Software takes out,
Obtain multiplying for the top n word of first Games Software word frequency corresponding in other Games Softwares and reverse document-frequency
Product, and the product of the word frequency and reverse document-frequency to same word in other Games Softwares is ranked up, according to sequence
Other Games Softwares successively are recommended to user.
Further, the text evaluation further includes step S1.9 and S1.10：
S1.9：For other Games Softwares, the N number of word for coming front is also taken out according to the process of step S1.1-S1.5；
S1.10：The relevance for calculating the top n word of different Games Softwares obtains demand of the user to Games Software.
Further, the picture evaluation includes the following steps：
S2.1：Reference picture is set；
S2.2：Significant region is arranged in reference picture, and picture to be evaluated is compareed with the significant region of setting,
It is identified；
S2.3：If can not identify, it is not belonging to the scene of this game, goes to step S2.5；If identification obtains, into
Row step S2.4；
S2.4：One score threshold is set, scores this picture recognition result, then makes the following judgment：Such as
Fruit scoring is lower than score threshold, then by the score of text evaluation multiplied by 1；If scoring is not less than score threshold, text is commented
The score of valence is multiplied by weight n, and 0<n<1；
S2.5：Classification storage is carried out according to image content, is supplied to Games Software developer.
Further, the reference picture in the step S2.1 is the surface chart of the scene figure in game or game.
Further, in the step S2.4, if scoring is lower than score threshold, also the interior of preservation text evaluation is dissolved in data
In library.
Further, in the step S2.2, before being identified, first picture to be evaluated is pre-processed, including figure
As acquisition, image enhancement, image restoration, image coding and compression, image segmentation.
Beneficial effect：The invention discloses a kind of recommended methods based on Games Software scene, comment from user's marking, text
Valence and picture evaluate these three aspects and carry out comprehensive marking to Games Software, are advantageously implemented the precision of software ranking.
Detailed description of the invention
Fig. 1 is the overall construction drawing in the specific embodiment of the invention；
Fig. 2 is the fining scoring process in the specific embodiment of the invention；
Fig. 3 is the flow chart of the text evaluation in the specific embodiment of the invention；
Fig. 4 is the pretreated flow chart of picture in the specific embodiment of the invention；
Fig. 5 is the flow chart of the picture evaluation in the specific embodiment of the invention.
Specific embodiment
Present embodiment discloses a kind of recommended method based on Games Software scene, acquires user to Games Software
Marking and comment, carry out data prediction, remove unworthy data, so that obtained data are truer, appropriateness is in trip
Opera script body.As depicted in figs. 1 and 2, then in comment text and picture evaluate, in conjunction with user's marking, text comments
Valence and the result of picture evaluation carry out comprehensive marking, are finally pushed according to comprehensive marking to Games Software.
As shown in figure 3, text evaluation includes the following steps：
S1.1：For a Games Software, the content for needing to carry out all comments of text evaluation is taken out；
S1.2：Word frequency and reverse document-frequency corresponding to each word in every comment are calculated, word is then calculated
The product of frequency and reverse document-frequency；
S1.3：A matrix is defined, the element in matrix is the word frequency of some word and reverse file frequency in certain comment
The product of rate；
S1.4：For the product of all word frequency and reverse document-frequency, remove less than the word frequency of K and reverse document-frequency
Product；Wherein, K is the whether significant threshold value of grammatical term for the character；
S1.5：The word frequency of same word and reverse document-frequency product are overlapped summation, then takes out and comes front
N number of word；
S1.6：Word frequency and reverse document-frequency product to the top n word of taking-up are summed, and the Games Software is obtained
Text evaluation score；
S1.7：For other Games Softwares, the N number of word for coming front is also taken out according to the process of step S1.1-S1.5；
S1.8：The top n word of other Games Softwares taking-up is traversed with the top n word that first Games Software takes out,
Obtain multiplying for the top n word of first Games Software word frequency corresponding in other Games Softwares and reverse document-frequency
Product, and the product of the word frequency and reverse document-frequency to same word in other Games Softwares is ranked up, according to sequence
Other Games Softwares successively are recommended to user；
S1.9：The relevance that the top n word of different Games Softwares is calculated using Apriori association rules method, is obtained
Demand of the user to Games Software.
Wherein, word frequency can be indicated with TF,Wherein m indicates the number that certain word occurs in this comment, M
Indicate total vocabulary number in this comment.Inversely document-frequency can be indicated with IDF,U is
General comment number, u are the comment number comprising a certain word.
As shown in figure 5, picture evaluation includes the following steps：
S2.1：Reference picture is set；Reference picture is the surface chart of the scene figure or game in game；
S2.2：Significant region is arranged in reference picture, pre-processes to picture to be evaluated, as shown in figure 4, include with
Lower process：Image Acquisition：Image content is extracted from comment content；Image enhancement：User submit comment need by
Processes, the quality of image such as acquisition, transmission, duplication more or less will cause certain degeneration, interested in image in order to protrude
Part, make the main structure of image definitely, it is necessary to improve to image, image enhancement improves the clear of image
The quality of degree, image, is more clear the profile of the object in image, details is more obvious；Image restoration：Image restoration is also referred to as
Image restores, and the image caused by the influence, movement of ambient noise when obtaining image obscures, makes due to the power of light etc.
It is fuzzy to obtain image, in order to which clearly image needs to restore image extraction comparison, image restores mainly to use filtering method,
Restore original graph from the image to degrade；Image coding and compression：The distinguishing feature of digital picture is that data volume is huge, needs to occupy
Sizable memory space, but the mass storage of computer based network bandwidth sum can not carry out the place of data image
Reason, storage, transmission, in order to quickly and easily transmit image or video in a network environment, then must be compiled to image
Code and compression, image compression technology can reduce the amount of redundant data and memory capacity, raising image transmitting speed of image
Degree shortens the processing time；Image segmentation：Image segmentation is to divide the image into some sons not overlapped and have respective feature
Region, each region are a continuums of pixel, and characteristic here can be color, shape, gray scale and texture of image etc.,
Image segmentation is that further image recognition, analysis and understanding are laid a good foundation.
Then picture to be evaluated is compareed with the significant region of setting, is identified；
S2.3：If can not identify, it is not belonging to the scene of this game, goes to step S2.5；If identification obtains, into
Row step S2.4；
S2.4：One score threshold is set, scores this picture recognition result, then makes the following judgment：Such as
Fruit scoring is lower than score threshold, then by the score of text evaluation multiplied by 1, and saves the interior of text evaluation and be dissolved in database, make
For the scheme of software change；If scoring is not less than score threshold, by the score of text evaluation multiplied by weight n, 0<n<1；
S2.5：Classification storage is carried out according to image content, is supplied to Games Software developer, developer can send out as needed
Effective information in existing picture.
As it can be seen that present embodiment has beneficial effect below：
1, score by rules is refined, convenient for the scoring of the more careful understanding software of user, provides player's download games more
Add safe guarantee.
2, it will be seen that the deficiency of game by evaluation content, understanding player is really idea, and integration game at this stage should
The function having, when being convenient for enterprise development game, Reasonable Orientation.
3, according to hobby rationally push, and it is not limited to the feature of game, but is intuitively felt according to game bring
Feel that, to push, in this way with more personalization, and the type of recommended games is also very much.
Claims (9)
1. a kind of recommended method based on Games Software scene, it is characterised in that：User is acquired to the marking of Games Software and is commented
By, in comment text and picture evaluate, in conjunction with user marking, text evaluation and picture evaluation result carry out it is comprehensive
Marking is closed, finally Games Software is pushed according to comprehensive marking.
2. the recommended method according to claim 1 based on Games Software scene, it is characterised in that：Carry out text evaluation and
Before picture evaluation, advanced line number Data preprocess.
3. the recommended method according to claim 1 based on Games Software scene, it is characterised in that：The text evaluation packet
Include following steps：
S1.1：For a Games Software, the content for needing to carry out all comments of text evaluation is taken out；
S1.2：Calculate word frequency and reverse document-frequency corresponding to each word in every comment, then calculate word frequency and
The product of reverse document-frequency；
S1.3：A matrix is defined, the element in matrix is the word frequency of some word and reverse document-frequency in certain comment
Product；
S1.4：For the product of all word frequency and reverse document-frequency, remove multiplying less than the word frequency of K and reverse document-frequency
Product；Wherein, K is the whether significant threshold value of grammatical term for the character；
S1.5：The word frequency of same word and reverse document-frequency product are overlapped summation, then takes out and comes the N number of of front
Word；
S1.6：Word frequency and reverse document-frequency product to the top n word of taking-up are summed, and the text of the Games Software is obtained
The score of this evaluation.
4. the recommended method according to claim 3 based on Games Software scene, it is characterised in that：The text evaluation is also
Including step S1.7 and S1.8：
S1.7：For other Games Softwares, the N number of word for coming front is also taken out according to the process of step S1.1-S1.5；
S1.8：The top n word that the taking-up of other Games Softwares is traversed with the top n word that first Games Software takes out, obtains
The product of the top n word of first Games Software word frequency corresponding in other Games Softwares and reverse document-frequency, and
The product of word frequency and reverse document-frequency of the same word in other Games Softwares is ranked up, according to the successive right of sequence
User recommends other Games Softwares.
5. the recommended method according to claim 3 based on Games Software scene, it is characterised in that：The text evaluation is also
Including step S1.9 and S1.10：
S1.9：For other Games Softwares, the N number of word for coming front is also taken out according to the process of step S1.1-S1.5；
S1.10：The relevance for calculating the top n word of different Games Softwares obtains demand of the user to Games Software.
6. the recommended method according to claim 1 based on Games Software scene, it is characterised in that：The picture evaluation packet
Include following steps：
S2.1：Reference picture is set；
S2.2：Significant region is arranged in reference picture, and picture to be evaluated is compareed with the significant region of setting, carries out
Identification；
S2.3：If can not identify, it is not belonging to the scene of this game, goes to step S2.5；If identification obtains, walked
Rapid S2.4；
S2.4：One score threshold is set, scores this picture recognition result, then makes the following judgment：If commented
Divide and be lower than score threshold, then by the score of text evaluation multiplied by 1；If scoring is not less than score threshold, by text evaluation
Score is multiplied by weight n, and 0<n<1；
S2.5：Classification storage is carried out according to image content, is supplied to Games Software developer.
7. the recommended method according to claim 6 based on Games Software scene, it is characterised in that：In the step S2.1
Reference picture be game in scene figure or game surface chart.
8. the recommended method according to claim 6 based on Games Software scene, it is characterised in that：The step S2.4
In, if scoring is lower than score threshold, also the interior of preservation text evaluation is dissolved in database.
9. the recommended method according to claim 6 based on Games Software scene, it is characterised in that：The step S2.2
In, before being identified, first picture to be evaluated is pre-processed, including Image Acquisition, image enhancement, image restoration, figure
As coding and compression, 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 true CN108920512A (en) | 2018-11-30 |
| CN108920512B 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) |
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 |
|---|
| 杨颖等: ""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"", 《心理学探新》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108920512B (en) | 2021-12-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764141B (en) | Game scene description method, device, equipment and storage medium thereof | |
| CN109033408B (en) | Information pushing method and device, computer readable storage medium and electronic equipment | |
| CN103744928B (en) | A kind of network video classification method based on history access record | |
| CN108492290A (en) | Image evaluation method and system | |
| CN106777282B (en) | The sort method and device of relevant search | |
| CN111729310B (en) | Method and device for sorting game props and electronic equipment | |
| CN111429341B (en) | Video processing method, device and computer readable storage medium | |
| CN110598700B (en) | Object display method and device, storage medium and electronic device | |
| CN111783712A (en) | Video processing method, device, equipment and medium | |
| CN111143613A (en) | Method, system, electronic device and storage medium for selecting video cover | |
| CN111597446B (en) | Content pushing method and device based on artificial intelligence, server and storage medium | |
| CN111737473B (en) | Text classification method, device and equipment | |
| CN112183588A (en) | Video processing method and device, electronic equipment and storage medium | |
| CN111432206A (en) | Video definition processing method and device based on artificial intelligence and electronic equipment | |
| CN112052352B (en) | Video ordering method, device, server and storage medium | |
| CN108920512A (en) | A kind of recommended method based on Games Software scene | |
| CN116992067A (en) | Non-heritable heritage digital display system and method | |
| CN107529067A (en) | The recommendation method and apparatus of video | |
| CN112565902A (en) | Video recommendation method and device and electronic equipment | |
| CN115965439A (en) | Data recall method, data recall device, electronic equipment and storage medium | |
| CN109299378A (en) | Methods of exhibiting, device, terminal and the storage medium of search result | |
| CN110287348A (en) | A kind of GIF format picture searching method based on machine learning | |
| CN115272057A (en) | Training of cartoon sketch image reconstruction network and reconstruction method and equipment thereof | |
| CN109376298B (en) | Data processing method and device, terminal equipment and computer storage medium | |
| CN103440342B (en) | Information-pushing method based on type of webpage and device |
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 |