CN113127674B - Song list recommendation method, device, electronic equipment and computer storage medium - Google Patents

Abstract

The embodiment of the invention discloses a song list recommending method, a song list recommending device, electronic equipment and a computer storage medium. The song list recommending method comprises the following steps: acquiring song behavior data of a user in a first period and song behavior data of the user in a second period; wherein, the first period of time to which the song behavior data of the first period of time of the user belongs is after the second period of time to which the song behavior data of the second period of time of the user belongs; determining a target song list by using the song behavior data of the first period of time of the user and the song behavior data of the second period of time of the user; and sending the target song list to the user terminal. According to the embodiment of the invention, the accuracy of song list recommendation can be improved.

Description

Song list recommendation method, device, electronic equipment and computer storage medium

technical field

The invention belongs to the technical field of data processing, and in particular relates to a song list recommendation method, device, electronic equipment and computer storage medium.

Background technique

At present, in addition to the communication function of the mobile terminal, the music playing function has become an indispensable additional function. However, in the face of today's massive song resource library, users are faced with a new problem: how to quickly find songs that suit them. In this regard, the traditional online music push method generally classifies songs, such as folk, rock, jazz, etc., and then provides them to the user for selection. If the user selects one or more classifications, the corresponding songs under the selected classification will be pushed. Although the traditional online music push method through classification can save users from the process of searching, selecting and playing, it is still difficult for users to hear the songs they want to listen to.

Therefore, how to improve the accuracy of song list recommendation is a technical problem urgently needed to be solved by those skilled in the art.

Contents of the invention

Embodiments of the present invention provide a song list recommendation method, device, equipment and computer storage medium, which can improve the accuracy of song list recommendation.

In the first aspect, a song list recommendation method is provided, including:

Acquiring the user's song behavior data in the first period and the user's song behavior data in the second period; wherein, the first period of the user's song behavior data in the first period is after the second period in which the user's song behavior data in the second period belongs;

Determine the target song list by using the song behavior data of the user in the first period and the song behavior data of the user in the second period;

Send the target song list to the user terminal.

Optionally, using the user's song behavior data in the first period and the user's song behavior data in the second period, determine the target song list, including:

Based on the weight scoring algorithm, use the user's song behavior data in the first period to determine the first playlist and the first score corresponding to each first playlist;

Based on the cosine similarity algorithm, using the user's song behavior data in the second period, determine the second playlist and the second score corresponding to the second playlist;

The target playlist is determined according to the first score of the first playlist and the second score of the second playlist.

Optionally, based on the weight scoring algorithm, the first song list and the first score corresponding to each first song list are determined by using the user's song behavior data in the first period, including:

According to the user's song behavior data in the first period, determine the first song corresponding to the user's first period song behavior data; wherein, the user's first period song behavior type corresponding to the user's first period song behavior data includes: user's first period song download, user's first period song collection and user's first period song trial listening; each first song corresponds to at least one user's first period song behavior;

Determining the first playlist containing the first song;

Determine the first score of the first song list according to the number of first songs in the first song list and the song behavior of the user corresponding to each first song in the first time period.

Optionally, based on the cosine similarity algorithm, the second song list and the second score corresponding to the second song list are determined by using the user's song behavior data in the second period, including:

According to the song behavior data of the user's second period, determine the second song corresponding to the song behavior data of the user's second period;

Determine the label of the second song according to the preset song label information, and perform deduplication processing on the label of the second song, and use the deduplicated song label as the user label;

Calculate the cosine similarity between the song tags of the user tag and each second song respectively to obtain the cosine similarity value;

According to the order of each cosine similarity value from large to small, determine the second song corresponding to the preset number of cosine similarity values with the largest value;

Based on the second song corresponding to the preset number of cosine similarity values with the largest value, the second song list and the second score are determined.

Optionally, the label of the second song is determined according to the preset song label information, and the label of the second song is deduplicated, and the deduplicated song label is used as the user label, including:

According to the song identification number of each second song, determine the copyright information corresponding to the second song;

Determine the song tag corresponding to each copyright information according to the preset copyright tag mapping table;

The tags of the second song are deduplicated, and the deduplicated song tags are used as user tags.

Optionally, acquiring the song behavior data of the user in the first period and the song behavior data of the user in the second period, including:

According to the preset data source table, the song behavior data of the user in the first period and the song behavior data of the user in the second period are determined.

In the second aspect, a song list recommendation device is provided, including:

The acquisition module is used to obtain the song behavior data of the user in the first period and the song behavior data of the user in the second period; wherein, the first period of the song behavior data of the user in the first period is after the second period of the song behavior data of the user in the second period;

Determining module, for utilizing user's song behavior data in the first period and the user's song behavior data in the second period to determine the target song list;

The sending module is used to send the target song list to the user terminal.

Optionally, identify modules, including:

The first song list determination sub-module is used to determine the first song list and the first score corresponding to each first song list by using the song behavior data of the user in the first period based on the weight scoring algorithm;

The second song list determination sub-module is used to determine the second song list and the second score corresponding to the second song list by using the user's song behavior data in the second period based on the cosine similarity algorithm;

The target playlist determination submodule is used to determine the target playlist according to the first score of the first playlist and the second score of the second playlist.

Optionally, the first song list determines submodules, including:

The first song determining unit is used to determine the first song corresponding to the user's first period song behavior data according to the user's first period song behavior data; wherein, the user's first period song behavior type corresponding to the user's first period song behavior data includes: user first period song download, user first period song collection and user first period song trial listening; each first song corresponds to at least one user's first period song behavior;

The first song list determination unit is used to determine the first song list containing the first song;

The first score determination unit is configured to determine the first score of the first song list according to the number of first songs in the first song list and the user's song behavior in the first period corresponding to each first song.

Optionally, the second song list determines submodules, including:

The second song determination unit is used to determine the second song corresponding to the user's second period song behavior data according to the user's second period song behavior data;

A user label determining unit is used to determine the label of the second song according to the preset song label information, and perform deduplication processing on the label of the second song, and use the deduplicated song label as the user label;

A cosine similarity calculation unit is used to calculate the cosine similarity between the user label and the song label of each second song respectively, to obtain the cosine similarity value;

The second song determination unit is used to determine the second song corresponding to the preset number of cosine similarity values with the largest value according to the order of each cosine similarity value from large to small;

The second song list determination unit is configured to determine the second song list and the second score based on the second song corresponding to the preset number of cosine similarity values with the largest value.

Optionally, the user label determination unit includes:

The copyright information determination subunit is used to determine the copyright information corresponding to the second song according to the song identification number of each second song;

The song label determination subunit is used to determine the song label corresponding to each copyright information according to the preset copyright label mapping table;

The user label determination subunit is configured to perform deduplication processing on the label of the second song, and use the deduplicated song label as the user label.

Optionally, get modules, including:

The acquisition sub-module is used to determine the song behavior data of the user in the first period and the song behavior data of the user in the second period according to the preset data source table.

In a third aspect, an electronic device is provided, and the electronic device includes: a processor and a memory storing computer program instructions;

When the processor executes the computer program instructions, the play list recommendation method in the first aspect or any optional implementation manner of the first aspect is implemented.

In a fourth aspect, a computer storage medium is provided. Computer program instructions are stored on the computer storage medium. When the computer program instructions are executed by a processor, the method for recommending playlists in the first aspect or in any optional implementation manner of the first aspect is implemented.

The song list recommendation method, device, electronic equipment and computer storage medium of the embodiments of the present invention can improve the accuracy of song list recommendation. The song list recommendation method first obtains the user's song behavior data in the first period and the user's song behavior data in the second period. Since the user's song behavior data in the first period and the user's song behavior data in the second period can reflect the user's song preference attribute characteristics, using the user's song behavior data in the first period and the user's song behavior data in the second period to determine the target playlist can improve the accuracy of song list recommendation.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative work.

Fig. 1 is a schematic flow chart of a song list recommendation method provided by an embodiment of the present invention;

Fig. 2 is a schematic flow chart of another song list recommendation method provided by an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a song list recommendation device provided by an embodiment of the present invention;

Fig. 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

Detailed ways

The characteristics and exemplary embodiments of various aspects of the present invention will be described in detail below. In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described here are only configured to explain the present invention, not to limit the present invention. It will be apparent to one skilled in the art that the present invention may be practiced without some of these specific details. The following description of the embodiments is only to provide a better understanding of the present invention by showing examples of the present invention.

It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements but also other elements not expressly listed or which are inherent to such process, method, article or apparatus. Without further limitations, an element defined by the statement "comprising..." does not exclude the presence of additional same elements in the process, method, article or device comprising said element.

At present, in addition to the communication function of the mobile terminal, the music playing function has become an indispensable additional function. However, in the face of today's massive song resource library, users are faced with a new problem: how to quickly find songs that suit them. In this regard, the traditional online music push method generally classifies songs, such as folk, rock, jazz, etc., and then provides them to the user for selection. If the user selects one or more classifications, the corresponding songs under the selected classification will be pushed. Although the traditional online music push method through classification can save users from the process of searching, selecting and playing, it is still difficult for users to hear the songs they want to listen to.

In order to solve the problems in the prior art, embodiments of the present invention provide a playlist recommendation method, device, electronic equipment, and computer storage medium. The method for recommending a song list provided by the embodiment of the present invention is firstly introduced below.

Fig. 1 is a schematic flowchart of a song list recommendation method provided by an embodiment of the present invention. As shown in Figure 1, the song list recommendation method includes:

S101. Obtain the song behavior data of the user in the first period and the song behavior data of the user in the second period; wherein, the first period to which the user's song behavior data belongs in the first period is after the second period to which the user's song behavior data in the second period belongs.

The duration of the first period and the second period may be the same or different, and are not specifically limited. In one embodiment, the song behavior data of the user in the first period is the user's recent song behavior data, and the user's song behavior data in the second period is the user's long-term song behavior data.

In order to obtain more accurate song behavior data of the user in the first period and the song behavior data of the user in the second period, in one embodiment, obtaining the song behavior data of the user in the first period and the song behavior data of the user in the second period may generally include: determining the song behavior data of the user in the first period and the song behavior data of the user in the second period according to the preset data source table.

In one embodiment, in order to make the recommended playlist not only conform to the user's recent behavior, but also conform to the user's long-term preference, the user's recent song behavior data and user's long-term song behavior data are respectively obtained. The user’s recent song behavior data is the user’s recently downloaded songs, user’s favorite songs, and user’s audiovisual songs, and search for playlists containing the above songs; the user’s long-term song behavior data includes user tags and song tags obtained from logs; the data source table is shown in Table 1, and the table header includes the data source name and the HIVE (data warehouse tool) table.

Table 1

According to the data source table, obtain the user's recent song behavior data to determine the song list. Among them, the song list screening rules include:

(1) Screen playlist playlists, filter conditions:

a) For playlists created after a certain date, the date is no more than 15 days away from the current date.

b) Playlist name setting.

c) The number of valid songs in the playlist cannot be less than 5.

(2) Identify the edited playlist among the playlists collected daily.

S102. Using the song behavior data of the user in the first period and the song behavior data of the user in the second period, determine a target playlist.

S103. Send the target playlist to the user terminal.

In order to recommend more accurate playlists to the user, in one embodiment, using the user’s song behavior data in the first period and the user’s song behavior data in the second period to determine the target playlist usually includes: based on a weight scoring algorithm, using the user’s song behavior data in the first period to determine the first song list and the first score corresponding to each first playlist;

In order to determine a more accurate first song list and the first score, in one embodiment, based on the weight scoring algorithm, using the user’s song behavior data in the first period, determine the first song list and the first score corresponding to each first song list, usually including: According to the user’s song behavior data in the first period, determine the first song corresponding to the user’s song behavior data in the first period; wherein, the user’s song behavior type corresponding to the user’s song behavior data in the first period includes: song downloading in the user’s first period, song collection in the user’s first period, and song trial listening in the user’s first period; each first song corresponds to at least one song behavior in the user’s first period; The first song list of the song; according to the quantity of the first song in the first song list and the user's song behavior in the first period corresponding to each first song, determine the first score of the first song list.

In one embodiment, the user's recent song behavior data is processed using a weight scoring method, and the data source table and the weight scoring table are used to determine all the first playlists and the first score of each first playlist. The weight scoring table is shown in Table 2.

Table 2 Weight scoring table

In one embodiment, determining the first score of each first playlist according to the weight scoring table specifically includes determining the score of the playlist according to the following behaviors:

(1) For user download behavior:

During the recommendation period, according to the songs downloaded by the user, search for a playlist containing the song. For every song in it, the playlist will get 4 points, and the sum will be accumulated according to different songs.

(2) User collection behavior:

During the recommendation cycle, search for a playlist containing the song based on the user's favorite songs. For each song in it, the playlist will get 3 points, and the sum will be accumulated for different songs.

(3) Audition behavior for users:

During the recommendation period, according to the songs that the user auditioned (the audition time is greater than 180s), search for the playlist containing the song. For each song in it, the playlist will get 1 point, and the sum will be accumulated according to different songs.

In order to determine a more accurate second song list and second score, in one embodiment, based on the cosine similarity algorithm, using the user's second period song behavior data to determine the second song list and the second score corresponding to the second song list, usually can include: According to the user's second period song behavior data, determine the second song corresponding to the user's second period song behavior data; determine the label of the second song according to the preset song label information, and perform deduplication processing on the label of the second song, and use the deduplicated song label as the user label; respectively calculate the cosine similarity between the user label and the song labels of each second song, Obtain the cosine similarity value; according to the order of each cosine similarity value from large to small, determine the second song corresponding to the preset number of cosine similarity values with the largest value; determine the second song list and the second score based on the second song corresponding to the preset number of cosine similarity values with the largest value.

In order to determine a more accurate user label, in one embodiment, determine the label of the second song according to the preset song label information, and perform deduplication processing on the label of the second song, and use the song label after deduplication as the user label, which may generally include: determining the copyright information corresponding to the second song according to the song identification number of each second song;

In one embodiment, the cosine similarity algorithm is used to process the user's long-term song behavior data, which specifically includes:

The calculation method of cosine similarity is used to calculate the similarity between user tags and song list tags. The formula for calculating cosine similarity is as follows:

For two-dimensional space, according to the vector dot product formula, it is obvious that:

Among them, a and b in formula (1) represent two vectors, and θ is the angle between them.

Assume that the coordinate representations of vectors a and b are (x ₁ , y ₁ ), (x ₂ , y ₂ ) respectively. but:

Extended to multi-dimensional, let the vector A=(A ₁ ,A ₂ ,…A _n ),B=(B ₁ ,B ₂ ,…B _n ), then:

Among them, n represents the dimension of vector A, A _i represents the i-th A vector, and B _i represents the i-th B vector.

An example calculation is as follows:

The set of user tags is {0,1,0,…,1}, and the set of song tags is {1,1,0,…,1}. Putting the vector into the formula can calculate the similarity score. Among them, the total dimension of tags is dynamically generated every day according to the data preprocessing, and the value is the union of user tags and song list tags.

The processing logic of the user's long-term song behavior data is to find out the tag of the song according to the copyright-tag (copyright_tag) relationship table, and then deduplicate the song tag as the user tag. Finally, the user tags and song tags are used to calculate the cosine similarity, and the TOP50 songs are taken as a playlist, and the score of the playlist is recorded as 8 points.

Combining two methods (weight scoring algorithm and cosine similarity algorithm) to summarize scores for song list recommendation, including:

(1) Calculate the final score of the playlist according to users and different scoring sources, and arrange them in descending order of score.

(2) The playlists recommended for 5 days are not repeated, and a playlist recommendation history table is generated.

(3) Insert the songs with the TOP1 score in the daily playlist into the history table and the recommendation table at the same time.

The above content is described below with a specific embodiment. As shown in FIG. 2 , the user song data is first collected to obtain the user's recent song behavior data and the user's long-term song behavior data. Then, on the one hand, based on the user’s recent song behavior data, perform screening and weighted summation according to the user’s downloading, collection, and audition behaviors, and determine at least one first playlist and the first score corresponding to each first playlist; Finally, based on the first score corresponding to each first playlist and the second score of the second playlist, a summary score recommendation operation is performed to determine a target playlist. This embodiment has the following beneficial effects:

1. Use the weight scoring method to process the user's recent behavior data, and use the cosine similarity algorithm to process the user's long-term behavior data. The combination of the two algorithms is used for data processing. Compared with the processing method of single weight scoring, the data processing is more accurate and perfect.

2. When the cosine similarity algorithm processes user long-term behavior data, the total dimension of tags is dynamically generated according to the daily data situation, and the value is the union of user tags and song tags, which makes data matching more accurate.

3. The user behavior data used includes the user's recent song behavior data and the user's long-term song behavior data, and the data collection is more accurate and comprehensive.

A playlist recommending device, electronic equipment, and computer storage medium provided in the embodiments of the present invention are introduced below. The playlist recommending device, electronic equipment, and computer storage medium described below and the playlist recommending method described above can be referred to in correspondence.

Fig. 3 is a schematic structural diagram of a song list recommending device provided by an embodiment of the present invention. As shown in Fig. 3, the song list recommending device may include:

Obtaining module 301, is used for acquiring user's song behavior data in the first period and user's song behavior data in the second period; wherein, the first period to which the user's first period song behavior data belongs is after the second period to which the user's second period song behavior data belongs;

A determining module 302, configured to determine the target song list by utilizing the song behavior data of the user in the first period and the song behavior data of the user in the second period;

The sending module 303 is configured to send the target playlist to the user terminal.

Optionally, in one embodiment, the determining module 302 includes:

The first song list determination sub-module is used to determine the first song list and the first score corresponding to each first song list by using the song behavior data of the user in the first period based on the weight scoring algorithm;

The second song list determination sub-module is used to determine the second song list and the second score corresponding to the second song list by using the user's song behavior data in the second period based on the cosine similarity algorithm;

The target playlist determination submodule is used to determine the target playlist according to the first score of the first playlist and the second score of the second playlist.

Optionally, in one embodiment, the first playlist determination submodule includes:

The first song determining unit is used to determine the first song corresponding to the user's first period song behavior data according to the user's first period song behavior data; wherein, the user's first period song behavior type corresponding to the user's first period song behavior data includes: user first period song download, user first period song collection and user first period song trial listening; each first song corresponds to at least one user's first period song behavior;

The first song list determination unit is used to determine the first song list containing the first song;

The first score determination unit is configured to determine the first score of the first song list according to the number of first songs in the first song list and the user's song behavior in the first period corresponding to each first song.

Optionally, in one embodiment, the second playlist determination submodule includes:

The second song determination unit is used to determine the second song corresponding to the user's second period song behavior data according to the user's second period song behavior data;

A user label determining unit is used to determine the label of the second song according to the preset song label information, and perform deduplication processing on the label of the second song, and use the deduplicated song label as the user label;

A cosine similarity calculation unit is used to calculate the cosine similarity between the user label and the song label of each second song respectively, to obtain the cosine similarity value;

The second song determination unit is used to determine the second song corresponding to the preset number of cosine similarity values with the largest value according to the order of each cosine similarity value from large to small;

The second song list determination unit is configured to determine the second song list and the second score based on the second song corresponding to the preset number of cosine similarity values with the largest value.

Optionally, in one embodiment, the user label determining unit includes:

The copyright information determination subunit is used to determine the copyright information corresponding to the second song according to the song identification number of each second song;

The song label determination subunit is used to determine the song label corresponding to each copyright information according to the preset copyright label mapping table;

The user label determination subunit is configured to perform deduplication processing on the label of the second song, and use the deduplicated song label as the user label.

Optionally, in one embodiment, the acquisition module includes:

The acquisition sub-module is used to determine the song behavior data of the user in the first period and the song behavior data of the user in the second period according to the preset data source table.

Each module in the song list recommending device provided in FIG. 3 has the function of realizing each step in the example shown in FIG. 1 , and achieves the same technical effect as the song list recommendation method shown in FIG. 1 .

Fig. 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

The electronic device may include a processor 401 and a memory 402 storing computer program instructions.

Specifically, the above-mentioned processor 401 may include a central processing unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured to implement one or more integrated circuits in the embodiments of the present invention.

Memory 402 may include mass storage for data or instructions. By way of example and not limitation, memory 402 may include a Hard Disk Drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Storage 402 may include removable or non-removable (or fixed) media, where appropriate. Under appropriate circumstances, the storage 402 can be inside or outside the comprehensive gateway disaster recovery device. In a particular embodiment, memory 402 is a non-volatile solid-state memory. In particular embodiments, memory 402 includes read-only memory (ROM). Where appropriate, the ROM may be mask programmed ROM, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), electrically rewritable ROM (EAROM) or flash memory or a combination of two or more of these.

The processor 401 reads and executes the computer program instructions stored in the memory 402 to implement any one of the song list recommendation methods in the above-mentioned embodiments.

In one example, the electronic device may further include a communication interface 403 and a bus 410 . Wherein, as shown in FIG. 4 , the processor 401 , the memory 402 , and the communication interface 403 are connected through a bus 410 to complete mutual communication.

The communication interface 403 is mainly used to realize the communication between various modules, devices, units and/or devices in the embodiments of the present invention.

The bus 410 includes hardware, software or both, and couples the components of the online data traffic charging device to each other. By way of example and not limitation, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infiniband Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus or other suitable bus or a combination of two or more of the above. Bus 410 may comprise one or more buses, where appropriate. Although embodiments of the invention describe and illustrate a particular bus, the invention contemplates any suitable bus or interconnect.

In addition, in combination with the song list recommendation method in the foregoing embodiments, the embodiment of the present invention may provide a computer storage medium for implementation. Computer program instructions are stored on the computer storage medium; when the computer program instructions are executed by a processor, the method for recommending playlists in the embodiment shown in FIG. 1 is realized.

It is to be understood that the invention is not limited to the specific arrangements and processes described above and shown in the drawings. For conciseness, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the sequence of steps after understanding the spirit of the present invention.

The functional blocks shown in the structural block diagrams described above may be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, it may be, for example, an electronic circuit, an application specific integrated circuit (ASIC), suitable firmware, a plug-in, a function card, or the like. When implemented in software, the elements of the invention are the programs or code segments employed to perform the required tasks. Programs or code segments can be stored in machine-readable media, or transmitted over transmission media or communication links by data signals carried in carrier waves. "Machine-readable medium" may include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and the like. Code segments may be downloaded via a computer network such as the Internet, an Intranet, or the like.

It should also be noted that the exemplary embodiments mentioned in the present invention describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps may be performed in the order mentioned in the embodiment, or may be different from the order in the embodiment, or several steps may be performed simultaneously.

The above is only a specific implementation of the present invention, and those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, module and unit can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto, and any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present invention, and these modifications or replacements should all be covered within the protection scope of the present invention.

Claims (9)

1. A song list recommendation method, comprising:

acquiring song behavior data of a user in a first period and song behavior data of the user in a second period; wherein, the first period of time to which the song behavior data of the first period of time of the user belongs is after the second period of time to which the song behavior data of the second period of time of the user belongs;

determining a target song list by using the song behavior data of the first period of time of the user and the song behavior data of the second period of time of the user;

sending the target song list to a user terminal;

the determining the target song list by using the song behavior data of the first period of time of the user and the song behavior data of the second period of time of the user comprises the following steps:

determining a first song list and a first score corresponding to each first song list by using song behavior data of the user in a first period based on a weight scoring algorithm;

based on a cosine similarity algorithm, determining a second song list and a second score corresponding to the second song list by using song behavior data of the user in a second period;

determining the target song list according to the first score of the first song list and the second score of the second song list;

the weight-based scoring algorithm, using the song behavior data of the user in the first period, determines a first song list and a first score corresponding to each first song list, including:

determining a first song corresponding to the song behavior data of the user in the first period according to the song behavior data of the user in the first period; the user first period song behavior type corresponding to the user first period song behavior data comprises: downloading songs in a first time period of a user, collecting songs in the first time period of the user and listening to the songs in the first time period of the user; each first song corresponds to at least one song behavior of a user in a first period;

determining the first song list including the first song;

and determining the first score of the first song list according to the number of the first songs in the first song list and the song behaviors of the user at the first time period corresponding to each first song.

2. The song list recommendation method according to claim 1, wherein the determining, based on the cosine similarity algorithm, a second song list and a second score corresponding to the second song list using the user second period song behavior data comprises:

determining a second song corresponding to the song behavior data of the user in the second period according to the song behavior data of the user in the second period;

determining a label of the second song according to preset song label information, performing duplication removal processing on the label of the second song, and taking the duplicated song label as a user label;

respectively calculating cosine similarity between the user tag and the song tags of the second songs to obtain cosine similarity values;

determining second songs corresponding to a preset number of cosine similarity values with the largest values according to the sequence of the cosine similarity values from large to small;

and determining the second song list and the second score based on the second songs corresponding to the maximum number of cosine similarity values.

3. The song list recommendation method according to claim 2, wherein determining the label of the second song according to the preset song label information, performing a duplication removal process on the label of the second song, and taking the duplication-removed song label as a user label, includes:

determining copyright information corresponding to each second song according to the song identification number of each second song;

determining song labels corresponding to each piece of copyright information according to a preset copyright label mapping table;

and performing de-duplication processing on the label of the second song, and taking the de-duplicated song label as the user label.

4. A song list recommendation method according to any one of claims 1 to 3, wherein said obtaining song behaviour data of a user in a first period of time and song behaviour data of a user in a second period of time comprises:

and determining the song behavior data of the first period of time of the user and the song behavior data of the second period of time of the user according to a preset data source table.

5. A song list recommendation apparatus, comprising:

the acquisition module is used for acquiring song behavior data of a first period of time of a user and song behavior data of a second period of time of the user; wherein, the first period of time to which the song behavior data of the first period of time of the user belongs is after the second period of time to which the song behavior data of the second period of time of the user belongs;

the determining module is used for determining a target song list by utilizing the song behavior data of the first period of the user and the song behavior data of the second period of the user;

the sending module is used for sending the target song list to the user terminal;

the determining module includes:

the first song list determining sub-module is used for determining a first song list and a first score corresponding to each first song list by utilizing the song behavior data of the user in a first period based on a weight scoring algorithm;

a second song list determining sub-module, configured to determine a second song list and a second score corresponding to the second song list by using the song behavior data of the user in a second period based on a cosine similarity algorithm;

a target song list determining submodule, configured to determine the target song list according to the first score of the first song list and the second score of the second song list;

the first song list determination submodule includes:

the first song determining unit is used for determining a first song corresponding to the song behavior data of the first period of the user according to the song behavior data of the first period of the user; the user first period song behavior type corresponding to the user first period song behavior data comprises: downloading songs in a first time period of a user, collecting songs in the first time period of the user and listening to the songs in the first time period of the user; each first song corresponds to at least one song behavior of a user in a first period;

a first song list determining unit configured to determine the first song list including the first song;

and the first score determining unit is used for determining the first score of the first song list according to the number of the first songs in the first song list and the song behaviors of the user at the first time period corresponding to each first song.

6. The song order recommendation apparatus of claim 5, wherein said second song order determination sub-module comprises:

a second song determining unit, configured to determine a second song corresponding to the second period song behavior data of the user according to the second period song behavior data of the user;

the user tag determining unit is used for determining the tag of the second song according to preset song tag information, performing duplication removal processing on the tag of the second song, and taking the duplication-removed song tag as a user tag;

the cosine similarity calculation unit is used for calculating cosine similarity between the user tag and the song tags of the second songs respectively to obtain cosine similarity values;

the second song determining unit is used for determining a second song corresponding to a preset number of cosine similarity values with the largest value according to the sequence from the high value to the low value of each cosine similarity value;

and the second song list determining unit is used for determining the second song list and the second score based on the second songs corresponding to the cosine similarity values with the largest preset number.

7. The song order recommendation apparatus according to claim 6, wherein said user tag determination unit comprises:

a copyright information determining subunit, configured to determine copyright information corresponding to each second song according to a song identifier of the second song;

a song label determining subunit, configured to determine a song label corresponding to each piece of copyright information according to a preset copyright label mapping table;

and the user tag determination subunit is used for carrying out duplication removal processing on the tags of the second songs and taking the duplication-removed song tags as the user tags.

8. An electronic device, the electronic device comprising: a processor and a memory storing computer program instructions;

the instructions of the computer program when executed by the processor implement a song recommendation method as claimed in any one of claims 1 to 4.

9. A computer storage medium having stored thereon computer program instructions which, when executed by a processor, implement the song recommendation method of any one of claims 1-4.