CN108255885B - Song recommendation method and system - Google Patents
Song recommendation method and system Download PDFInfo
- Publication number
- CN108255885B CN108255885B CN201611251392.4A CN201611251392A CN108255885B CN 108255885 B CN108255885 B CN 108255885B CN 201611251392 A CN201611251392 A CN 201611251392A CN 108255885 B CN108255885 B CN 108255885B
- Authority
- CN
- China
- Prior art keywords
- search
- user
- list
- time
- song
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/63—Querying
- G06F16/638—Presentation of query results
- G06F16/639—Presentation of query results using playlists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/63—Querying
- G06F16/635—Filtering based on additional data, e.g. user or group profiles
- G06F16/637—Administration of user profiles, e.g. generation, initialization, adaptation or distribution
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A method of recommending songs, comprising: extracting search records of a song playing log to obtain a search list of a user, wherein the search list comprises search words input by the user and the time of clicking search; calculating the frequency of the search word of the user according to the search list to obtain a search word frequency list of the user; extracting the search time of the search word in the song playing log to obtain a search word time list, wherein the search word time list comprises the moment when the user clicks the song; screening according to the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search, and obtaining an effective search list; and obtaining a recommendation list according to the search word frequency number list and the effective search list. The method and the device can recommend songs to the user according to the logs searched and clicked on the music software; the method not only can accurately recommend popular songs to the user, but also can be simple and convenient in algorithm and high in calculation efficiency.
Description
Technical Field
The embodiment of the invention relates to the technical field of music analysis, in particular to a song recommending method and system.
Background
Currently, there are a variety of music software on the market, each of which generates a large number of search logs and click logs every day. The logs contain a large amount of implicit user behavior data, and mining analysis of the data is necessary.
In addition, the search behavior has a strong purpose, and if a keyword is searched and a click is performed in the search result music, the keyword and the clicked music can be considered to have a strong correlation. Thereby, a correlation pair between the keyword and the music can be obtained.
Therefore, how to utilize the search log and the click log to display songs similar to the search terms each time the user searches for the keyword becomes a problem to be solved urgently.
Disclosure of Invention
According to the method and the device, the implicit association of a large number of logs searched and clicked by users on music software can be analyzed, so that songs are recommended to the users.
A first aspect of the present application provides a method for recommending songs, where the method includes: extracting search records of a song playing log to obtain a search list of a user, wherein the search list comprises search words input by the user and the time of clicking search; calculating the frequency of the search word of the user according to the search list to obtain a search word frequency list of the user; extracting the search time of the search word in the song playing log to obtain a search word time list, wherein the search word time list comprises the moment when the user clicks the song; screening according to the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search, and obtaining an effective search list; and obtaining a recommendation list according to the search word frequency number list and the effective search list.
In one possible implementation, the search listing includes a behavior time when the user enters a search term and clicks on a search, a user ID, a search term, and a song ID.
In one possible implementation, the search term frequency number list includes a search term, a song ID, and a play number.
In one possible implementation, the search term time list includes a search time, a user ID, a search term, and a next search term.
In one possible implementation, the all-valid search list includes search terms, song IDs, and valid association frequency.
In one possible implementation, the recommendation list includes song IDs, search terms, and play times.
In a possible implementation manner, the screening is performed according to a difference between a moment when the user clicks a song and a moment when the user inputs a search word and clicks a search, so as to obtain an effective search list, and the method includes: comparing the difference value between the time when the user clicks the song and the time when the user inputs the search word and clicks the search with a preset effective time value; when the difference value is smaller than or equal to the preset effective time value, the search of the user is effective search; and when the difference value is larger than the preset effective time value, the search of the user is invalid search.
In one possible implementation, the preset valid time value is 5 minutes.
The second aspect of the application provides a song recommendation system, which comprises an extraction unit and a processing unit, wherein the extraction unit is used for extracting search records of a song playing log to obtain a search list of a user, and the search list comprises a moment when the user inputs a search word and clicks for searching; the processing unit is used for calculating the frequency of the search words of the user according to the search list to obtain a search word frequency list of the user; extracting the search time of the search word in the song playing log to obtain a search word time list, wherein the search word time list comprises the moment when the user clicks the song; screening according to the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search, and obtaining an effective search list; and obtaining a recommendation list according to the search word frequency number list and the effective search list.
In one possible implementation, the system further includes a comparison unit; the comparison unit compares the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search with a preset time value; when the difference value is smaller than or equal to the preset time value, the search of the user is effective search; and when the difference value is larger than the preset time value, the search of the user is invalid search.
According to the method and the device, the implicit association of the logs can be analyzed according to the logs searched and clicked on the music software, so that songs can be recommended to the user; the method not only can accurately recommend popular songs to the user, but also can be simple and convenient in algorithm and high in calculation efficiency.
Drawings
Fig. 1 is a schematic flowchart of a song recommendation method according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a search list of songs according to an embodiment of the present invention;
FIG. 3 is a search term frequency sequence representation intent of a song provided by an embodiment of the present invention;
FIG. 4 is a diagram illustrating a search term time list of songs according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of an effective search list of songs provided by an embodiment of the present invention;
FIG. 6 is a schematic diagram of a song recommendation list according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a song recommendation system according to an embodiment of the present invention.
Detailed Description
The technical solutions of the embodiments of the present invention are further described in detail with reference to the accompanying drawings and embodiments. The list of songs used hereinafter is described by way of example and for convenience of description only and should not be construed as limiting the embodiments of the present invention, as described by those skilled in the art.
Fig. 1 is a flowchart illustrating a song recommendation method according to an embodiment of the present invention, and as shown in fig. 1, the method includes steps S101 to S105.
S101, extracting search records of the song playing logs to obtain a search list of the user, wherein the search list comprises the time when the user inputs search words and clicks for searching.
Examples of song play logs are as follows:
in the above step, the PSRC field is searched for a search result- > result list "and a" FROM ═ song library ", and the user search behavior time is obtained according to the TM field. The user search behavior time refers to a time when the user inputs a search word and clicks a search.
It should be noted that PSRC is the song source field, TM is the time field, and there is a "search result- > result list" representation that this song play is from the search action. In the embodiment of the present invention, only the search of the current day is considered, and only the playtime > duration is considered to be 0.95, that is, the song has a playing time 0.95 times of the song complete time.
In this step, the search list includes a behavior time when the user inputs a search word and clicks a search, a user ID, a search word, and a song ID. As shown in fig. 2, fig. 2 is a schematic diagram of a search list of songs according to an embodiment of the present invention. The user ID identifies the user in the music database and the song ID identifies the song in the music database.
S102, calculating the frequency of the search words of the user according to the search list to obtain a search word frequency list of the user.
In this step, since a plurality of users may search with the same search word, a plurality of songs may appear, and a plurality of users may click on the same song. In the above search list, there may be a case where the user ID, the search word, and the song ID are duplicated, and thus, the search list needs to be deduplicated. And calculating the playing times of the song ID corresponding to the same search word of the user.
Fig. 3 is a search term frequency sequence representation intention of a song according to an embodiment of the present invention. As shown in fig. 3, the search word frequency count list includes a search word, a song ID, and a play number.
S103, extracting the search time of the search word in the song playing log to obtain a search word time list, wherein the search word time list comprises the time when the user clicks the song.
In this step, after the user inputs the search word and clicks the search, the music play may not be clicked. The song is not clicked. That is, the time of the user clicking the search is invalid, and the search time needs to be removed, so that the user can more accurately know that the user clicked the songs through the search word.
At this time, the effective search time of the search term in the song playing log, that is, the time when the user clicks the song, needs to be extracted. Fig. 4 is a schematic diagram of a search term time list of songs according to an embodiment of the present invention, where as shown in fig. 4, the search term time list includes a search time, a user ID, a search term, and a next search term.
S104, screening according to the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search, and obtaining an effective search list.
In the step, comparing the difference value between the time when the user clicks the song and the time when the user inputs the search word and clicks the search with a preset effective time value; when the difference value is smaller than or equal to the preset effective time value, the search of the user is effective search; and when the difference value is larger than the preset effective time value, the search of the user is invalid search.
In one example, the predetermined validity time value is 5 minutes. And the implicit association between the search terms and the song playing is embodied by presetting the effective time value. As shown in fig. 5, fig. 5 is a schematic diagram of an effective search list of songs provided by an embodiment of the present invention, where the effective search list includes search terms, song IDs, and effective association frequency.
S105, obtaining a recommendation list according to the search word frequency number list and the effective search list.
In this step, the following are taken into account: 1, effective association between the search term and the song ID, namely effective search; and 2, playing times. And combining the search word frequency number list and the effective search list to obtain a recommendation list.
Fig. 6 is a schematic diagram of a recommendation list of songs according to an embodiment of the present invention, and as shown in fig. 6, the recommendation list includes song IDs, search terms, and play times.
It should be noted that each search term in the recommendation list may be sorted according to the play times, or the recommendation list may be filtered, and only the play times list with the top rank of 10 is left as the recommendation list.
When a user searches, the method of the embodiment of the invention can recommend the song 10 before the playing times of the search word input by the user, and the song recommendation is completed.
The following describes a song recommendation system. Fig. 7 is a schematic structural diagram of a song recommendation system according to an embodiment of the present invention. As shown in fig. 7, the recommendation system includes an extraction unit 701, a processing unit 702, and a comparison unit 703.
The extracting unit 701 is configured to extract a search record of the song playing log to obtain a search list of the user, where the search list includes a time when the user inputs a search word and clicks a search.
The processing unit 702 is configured to calculate the frequency of the search word of the user according to the search list, so as to obtain a search word frequency list of the user; extracting the search time of the search word in the song playing log to obtain a search word time list, wherein the search word time list comprises the moment when the user clicks the song; screening according to the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search, and obtaining an effective search list; and obtaining a recommendation list according to the search word frequency number list and the effective search list.
The system further includes a comparing unit 703 for comparing a difference between a time when the user clicks the song and a time when the user inputs the search word and clicks the search with a preset time value; when the difference value is smaller than or equal to the preset time value, the search of the user is effective search; and when the difference value is larger than the preset time value, the search of the user is invalid search.
The embodiment of the present invention is not detailed, and is described in detail in fig. 1 and the text portion of fig. 1, which is not described herein again.
According to the method and the device, the implicit association of the logs can be analyzed according to the logs searched and clicked on the music software, so that songs can be recommended to the user; the method not only can accurately recommend popular songs to the user, but also can be simple and convenient in algorithm and high in calculation efficiency.
The above-mentioned embodiments are intended to illustrate the objects, technical solutions and advantages of the present invention in further detail, and it should be understood that the above-mentioned embodiments are merely exemplary embodiments of the present invention, and are not intended to limit the scope of the present invention, and any modifications, equivalent substitutions, improvements and the like made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method for recommending songs, the method comprising:
extracting search records of a song playing log to obtain a search list of a user, wherein the search list comprises search words input by the user and the time of clicking search;
calculating the frequency of the search word of the user according to the search list to obtain a search word frequency list of the user;
extracting the search time of the search word in the song playing log to obtain a search word time list, wherein the search word time list comprises the moment when the user clicks the song;
screening according to the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search, and obtaining an effective search list;
and obtaining a recommendation list according to the search word frequency number list and the effective search list.
2. The method of claim 1, wherein the search listing comprises a behavior time for a user to enter a search term and click to search, a user ID, a search term, and a song ID.
3. The method of claim 1, wherein the search term frequency number list comprises search terms, song IDs, and play numbers.
4. The method of claim 1, wherein the search term time list comprises a search time, a user ID, a search term, and a next search term.
5. The method of claim 1, wherein all valid search listings include search terms, song IDs, and valid association frequencies.
6. The method of claim 1, wherein the recommendation list comprises song IDs, search terms, and play times.
7. The method of claim 1, wherein the screening according to the difference between the time when the user clicks the song and the time when the user inputs the search word and clicks the search to obtain the effective search list comprises:
comparing the difference value between the time when the user clicks the song and the time when the user inputs the search word and clicks the search with a preset effective time value;
when the difference value is smaller than or equal to the preset effective time value, the search of the user is effective search;
and when the difference value is larger than the preset effective time value, the search of the user is invalid search.
8. The method of claim 7, wherein the predetermined effective time value is 5 minutes.
9. A system for recommending songs, characterized in that it comprises an extraction unit and a processing unit, wherein,
the extraction unit is used for extracting the search records of the song playing logs to obtain a search list of the user, wherein the search list comprises the time when the user inputs search words and clicks for searching;
the processing unit is used for calculating the frequency of the search words of the user according to the search list to obtain a search word frequency list of the user; extracting the search time of the search word in the song playing log to obtain a search word time list, wherein the search word time list comprises the moment when the user clicks the song; screening according to the difference value between the moment when the user clicks the song and the moment when the user inputs the search word and clicks the search, and obtaining an effective search list; and obtaining a recommendation list according to the search word frequency number list and the effective search list.
10. The system of claim 9, further comprising a comparison unit, wherein,
the comparison unit compares the difference value between the time when the user clicks the song and the time when the user inputs the search word and clicks the search with a preset time value; when the difference value is smaller than or equal to the preset time value, the search of the user is effective search; and when the difference value is larger than the preset time value, the search of the user is invalid search.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611251392.4A CN108255885B (en) | 2016-12-29 | 2016-12-29 | Song recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611251392.4A CN108255885B (en) | 2016-12-29 | 2016-12-29 | Song recommendation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108255885A CN108255885A (en) | 2018-07-06 |
| CN108255885B true CN108255885B (en) | 2020-11-06 |
Family
ID=62721270
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611251392.4A Active CN108255885B (en) | 2016-12-29 | 2016-12-29 | Song recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108255885B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109189977A (en) * | 2018-08-10 | 2019-01-11 | 北京微播视界科技有限公司 | Background music processing method, device and electronic equipment |
| CN116186322A (en) * | 2023-05-04 | 2023-05-30 | 四川酷乐科技有限公司 | Music recommendation method, device, equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1761961A (en) * | 2003-03-19 | 2006-04-19 | Nhn株式会社 | Method and apparatus for detecting invalid clicks on the internet search engine |
| CN102479366A (en) * | 2010-11-25 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Commodity recommending method and system |
| CN102591948A (en) * | 2011-12-27 | 2012-07-18 | 厦门市美亚柏科信息股份有限公司 | Method and system for improving search results based on user behavior analysis |
| CN103631801A (en) * | 2012-08-23 | 2014-03-12 | 阿里巴巴集团控股有限公司 | Method and device for providing commodity information |
| CN105808621A (en) * | 2014-12-31 | 2016-07-27 | 北京奇虎科技有限公司 | Method and device for calculating response time of search time |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8639570B2 (en) * | 2008-06-02 | 2014-01-28 | Microsoft Corporation | User advertisement click behavior modeling |
-
2016
- 2016-12-29 CN CN201611251392.4A patent/CN108255885B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1761961A (en) * | 2003-03-19 | 2006-04-19 | Nhn株式会社 | Method and apparatus for detecting invalid clicks on the internet search engine |
| CN102479366A (en) * | 2010-11-25 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Commodity recommending method and system |
| CN102591948A (en) * | 2011-12-27 | 2012-07-18 | 厦门市美亚柏科信息股份有限公司 | Method and system for improving search results based on user behavior analysis |
| CN103631801A (en) * | 2012-08-23 | 2014-03-12 | 阿里巴巴集团控股有限公司 | Method and device for providing commodity information |
| CN105808621A (en) * | 2014-12-31 | 2016-07-27 | 北京奇虎科技有限公司 | Method and device for calculating response time of search time |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108255885A (en) | 2018-07-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5540079B2 (en) | Knowledge base construction method and apparatus | |
| EP3005168B1 (en) | Natural language search results for intent queries | |
| AU2009234120B2 (en) | Search results ranking using editing distance and document information | |
| CN105653537B (en) | Paging query method and device for database application system | |
| CN107180093B (en) | Information searching method and device and timeliness query word identification method and device | |
| JP5616444B2 (en) | Method and system for document indexing and data querying | |
| JP2010067175A (en) | Hybrid content recommendation server, recommendation system, and recommendation method | |
| US9165058B2 (en) | Apparatus and method for searching for personalized content based on user's comment | |
| AU2014212510A1 (en) | Systems and methods for indentifying documents based on citation history | |
| CN102867042A (en) | Method and device for searching multimedia file | |
| CN104462085A (en) | Method and device for correcting search keywords | |
| CN111008321A (en) | Recommendation method and device based on logistic regression, computing equipment and readable storage medium | |
| KR101651780B1 (en) | Method and system for extracting association words exploiting big data processing technologies | |
| US10353966B2 (en) | Dynamic attributes for searching | |
| CN105159938A (en) | Retrieval method and apparatus | |
| CN105447169A (en) | Document normalization method, document searching method and corresponding apparatus | |
| JP5952711B2 (en) | Prediction server, program and method for predicting future number of comments in prediction target content | |
| CN110555108B (en) | Event context generation method, device, equipment and storage medium | |
| CN104657376A (en) | Searching method and searching device for video programs based on program relationship | |
| CN108255885B (en) | Song recommendation method and system | |
| CN105930423A (en) | Multimedia similarity determination method and apparatus as well as multimedia recommendation method | |
| CN106202440B (en) | Data processing method, device and equipment | |
| CN103226601A (en) | Method and device for image search | |
| KR101557960B1 (en) | Device for selecting core kyword, method for selecting core kyword, and method for providing search service using the same | |
| Kruit et al. | Extracting N-ary facts from wikipedia table clusters |
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 |