CN109408665B - Information recommendation method and device and storage medium - Google Patents
Information recommendation method and device and storage medium Download PDFInfo
- Publication number
- CN109408665B CN109408665B CN201811642134.8A CN201811642134A CN109408665B CN 109408665 B CN109408665 B CN 109408665B CN 201811642134 A CN201811642134 A CN 201811642134A CN 109408665 B CN109408665 B CN 109408665B
- Authority
- CN
- China
- Prior art keywords
- song
- preset
- historical
- data
- scoring
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses an information recommendation method, an information recommendation device and a storage medium, wherein the method comprises the following steps: acquiring historical operation data of a target user for selecting songs aiming at the full amount of history in a preset statistical period; obtaining historical evaluation data of each historical selected song in the total historical selected songs based on the corresponding relation between preset operation data and operation scores and the historical operation data; the historical evaluation data is used for representing the likeness degree of the target user to the song; performing time attenuation processing on the historical evaluation data to obtain historical statistical evaluation data of each historical selected song; sorting the full-scale historical selected songs according to the historical statistical evaluation data to obtain a sorting result of the full-scale historical selected songs; and recommending the target song according to the sequencing result and a preset full-scale song list.
Description
Technical Field
The present invention relates to information processing technologies, and in particular, to an information recommendation method and apparatus, and a storage medium.
Background
With the development of terminal and internet technology, users generally listen to music through music application, music application recommends music meeting their personal preferences for users through some music recommendation algorithms, the currently used music recommendation algorithm is to form a group by a plurality of users with mutual interests, generate recommendation information by using information of interest and information of no interest of all users in the group, and recommend the recommendation information to all users, so that the obtained personal recommendation information is greatly influenced by other users in the group, if abnormal users exist in other users or the information of interest of other users is inaccurate, the problem that the deviation of the recommendation information and the personal preferences is large is caused, that is, the accuracy of the recommendation information is low.
Disclosure of Invention
The invention mainly aims to provide an information recommendation method, an information recommendation device and a storage medium, which can improve the accuracy of recommended information.
The technical scheme of the invention is realized as follows:
the embodiment of the invention provides an information recommendation method, which comprises the following steps:
acquiring historical operation data of a target user for selecting songs aiming at the full amount of history in a preset statistical period;
obtaining historical evaluation data of each historical selected song in the total historical selected songs based on the corresponding relation between preset operation data and operation scores and the historical operation data; the historical evaluation data is used for representing the likeness degree of the target user to the song;
performing time attenuation processing on the historical evaluation data to obtain historical statistical evaluation data of each historical selected song;
sorting the full-scale historical selected songs according to the historical statistical evaluation data to obtain a sorting result of the full-scale historical selected songs;
and recommending the target song according to the sequencing result and a preset full-scale song list.
In the above scheme, the preset statistical period includes at least one preset time period; the historical evaluation data comprises time interval evaluation data corresponding to each time interval in the at least one preset time interval;
the time attenuation processing is performed on the historical evaluation data to obtain the historical statistical evaluation data of each historical selected song, and the method comprises the following steps:
obtaining an attenuation coefficient corresponding to each time period according to the corresponding relation between a preset attenuation coefficient and time and the time information represented by each time period;
and summing the time interval evaluation data corresponding to the at least one preset time interval according to the attenuation coefficient to obtain the historical statistical evaluation data.
In the foregoing solution, the recommending a target song according to the sorting result and a preset full-scale menu includes:
according to the sorting result, determining a reference song from the songs selected from the full-scale history;
determining a reference song list from the preset total song list according to the reference song;
and recommending the target song according to the reference song list.
In the above scheme, the reference song list includes a sub-reference song list corresponding to each reference song in the reference songs;
the recommending the target song according to the reference song list comprises the following steps:
and determining the target song from the reference list according to the repeated times of each song in the sub-reference list, and recommending the target song.
In the foregoing solution, the determining the target song from the reference list according to the number of repetitions of each song in the sub-reference list includes:
sorting the historical statistical evaluation data of each reference song to obtain a data sorting sequence number of each reference song;
sequencing the repeated times of each song in the sub-reference song list to obtain a sequencing sequence number of the times of each song in the sub-reference song list;
determining the songs to be recommended corresponding to each reference song from the sub-reference song list according to the time sequencing sequence number to obtain at least one song to be recommended;
determining the favorite score of the song to be recommended according to the data sorting sequence number and the frequency sorting sequence number of the song to be recommended;
and determining the target song from the at least one song to be recommended according to the preference score.
In the foregoing solution, after the determining a reference menu from the preset full-volume menu according to the reference song, the method further includes:
according to preset song attributes, quantifying the songs of each of the reference singals to obtain reference attribute characteristics of each of the reference singals;
obtaining scoring data containing scoring results of each song in the reference song list according to a preset song list scoring model and the reference attribute characteristics; the preset singing bill grading model represents the corresponding relation between attribute characteristics and grading data;
and when first scoring data in the scoring data is smaller than a preset scoring threshold value, deleting the singing sheet corresponding to the first scoring data in the reference singing sheet from the reference singing sheet to obtain an updated reference singing sheet.
In the foregoing solution, before obtaining scoring data including a scoring result of each of the reference singing tickets according to a preset singing ticket scoring model and the reference attribute feature, the method further includes:
in a preset training period, obtaining sample operation data of a target user for a full amount of sample songs;
determining a positive sample song list and a negative sample song list from the preset full-volume song list according to the sample operation data;
acquiring positive scoring data corresponding to the positive sample song list, wherein the positive sample song list and the positive scoring data form a positive sample;
acquiring negative scoring data corresponding to the negative sample song list, wherein the negative sample song list and the negative scoring data form a negative sample;
training a preset initial singing bill scoring model by using the positive sample and the negative sample until the accuracy of the trained initial singing bill scoring model is not less than a preset accuracy threshold;
and taking the trained initial singing bill scoring model as the preset singing bill scoring model.
The embodiment of the invention provides an information recommendation device, which comprises: the device comprises an acquisition unit, a calculation unit, an attenuation processing unit and a recommendation unit; wherein the content of the first and second substances,
the acquisition unit is used for acquiring historical operation data of a target user for selecting songs aiming at the full amount of history in a preset statistical period;
the computing unit is used for obtaining historical evaluation data of each historical selected song in the total historical selected songs based on the corresponding relation between preset operation data and operation scores and the historical operation data; the historical evaluation data is used for representing the likeness degree of the target user to the song;
the attenuation processing unit is used for carrying out time attenuation processing on the historical evaluation data to obtain historical statistical evaluation data of each historical selected song;
the recommending unit is used for sequencing the full-amount historical selected songs according to the historical statistical evaluation data to obtain a sequencing result of the full-amount historical selected songs; and recommending the target song according to the sequencing result and the preset full-scale song list.
In the above scheme, the preset statistical period includes at least one preset time period; the historical evaluation data comprises time interval evaluation data corresponding to each time interval in the at least one preset time interval;
the attenuation processing unit is specifically configured to obtain an attenuation coefficient corresponding to each time period according to a corresponding relationship between a preset attenuation coefficient and time and the time information represented by each time period; and according to the attenuation coefficient, summing the time interval evaluation data corresponding to the at least one preset time interval to obtain the historical statistical evaluation data.
In the above scheme, the recommending unit is specifically configured to determine a reference song from the full-amount history selected songs according to the sorting result; determining a reference song list from the preset full-volume song list according to the reference song; and recommending the target song according to the reference song list.
In the above scheme, the reference song list includes a sub-reference song list corresponding to each reference song in the reference songs;
and the recommending unit is specifically used for determining the target song from the reference list according to the repetition times of each song in the sub-reference list and recommending the target song.
In the foregoing scheme, the recommending unit is specifically configured to sort the historical statistical evaluation data of each reference song to obtain a data sorting sequence number of each reference song; sequencing the repeated times of each song in the sub-reference song list to obtain a sequencing sequence number of the times of each song in the sub-reference song list; determining the songs to be recommended corresponding to each reference song from the sub-reference song list according to the time sequencing sequence number to obtain at least one song to be recommended; determining the favorite score of the song to be recommended according to the data sorting serial number and the frequency sorting serial number of the song to be recommended; and determining the target song from the at least one song to be recommended according to the preference score.
In the foregoing solution, the recommending unit is further configured to quantify, according to a preset song attribute, the song of each of the reference menus after the reference menu is determined from the preset total volume menu according to the reference song, so as to obtain a reference attribute feature of each of the reference menus; obtaining scoring data containing scoring results of each song in the reference song list according to a preset song list scoring model and the reference attribute characteristics; the preset singing bill grading model represents the corresponding relation between attribute characteristics and grading data; and when first scoring data in the scoring data is smaller than a preset scoring threshold value, deleting the singing sheet corresponding to the first scoring data in the reference singing sheet from the reference singing sheet to obtain an updated reference singing sheet.
In the above scheme, the apparatus further comprises: the model training unit is used for acquiring sample operation data of a target user for a full amount of sample songs in a preset training period before obtaining scoring data containing a scoring result of each song in the reference song list according to a preset song list scoring model and the reference attribute characteristics; determining a positive sample song list and a negative sample song list from the preset full-volume song list according to the sample operation data; acquiring positive scoring data corresponding to the positive sample song list, wherein the positive sample song list and the positive scoring data form a positive sample; acquiring negative scoring data corresponding to the negative sample song list, wherein the negative sample song list and the negative scoring data form a negative sample; training a preset initial singing bill scoring model by using the positive sample and the negative sample until the accuracy of the trained initial singing bill scoring model is not less than a preset accuracy threshold; and taking the trained initial singing bill scoring model as the preset singing bill scoring model.
The embodiment of the invention provides an information recommendation device, which comprises: a processor, a memory and a communication bus, the memory communicating with the processor through the communication bus, the memory storing one or more programs executable by the processor, the one or more programs, when executed, causing the processor to perform the steps of any of the information recommendation methods described above.
An embodiment of the present invention provides a computer-readable storage medium storing a program, which, when executed by at least one processor, causes the at least one processor to perform the steps of any one of the information recommendation methods described above.
The embodiment of the invention provides an information recommendation method, an information recommendation device and a storage medium, wherein the method comprises the following steps: acquiring historical operation data of a target user for selecting songs aiming at the full amount of history in a preset statistical period; obtaining historical evaluation data of each historical selected song in the total historical selected songs based on the corresponding relation between preset operation data and operation scores and the historical operation data; the historical evaluation data is used for representing the likeness degree of the target user to the song; performing time attenuation processing on the historical evaluation data to obtain historical statistical evaluation data of each historical selected song; sorting the full-scale historical selected songs according to the historical statistical evaluation data to obtain a sorting result of the full-scale historical selected songs; and recommending the target song according to the sequencing result and a preset full-scale song list. By adopting the technical implementation scheme, time attenuation is carried out on the historical evaluation data in the preset statistical period to obtain the historical statistical evaluation data after the time attenuation, further the sequencing result of the whole quantity of historically-selected songs is obtained, the target songs are recommended according to the sequencing result, and as the historical evaluation data show the song likeness degree of the target user in the preset time period, the time attenuation is carried out on the historical evaluation data in the preset statistical period, the historical statistical evaluation data showing the current preference of the user can be obtained, further the sequencing result of the historical statistical evaluation data of the whole quantity of historical songs is utilized, the obtained target songs are more in line with the current preference of the target user, and the accuracy of the recommended information is improved.
Drawings
Fig. 1 is a schematic structural diagram of an information recommendation system according to an embodiment of the present invention;
fig. 2 is a first flowchart of an information recommendation method according to an embodiment of the present invention;
fig. 3 is a second flowchart of an information recommendation method according to an embodiment of the present invention;
fig. 4 is a first schematic structural diagram of an information recommendation device according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an information recommendation apparatus according to an embodiment of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
In the following description, suffixes such as "module", "component", or "unit" used to denote elements are used only for facilitating the explanation of the present invention, and have no specific meaning in itself. Thus, "module", "component" or "unit" may be used mixedly.
As shown in fig. 1, which is a schematic structural diagram of an information recommendation system for implementing various embodiments of the present invention, the system 1 may include: a server 10 and a terminal 11; the server 10 may store data files of the application programs, user data of the application programs, and the like, and may also broadcast and synchronize the user data; when the terminal 11 runs the application program, it performs data interaction with the server 10, and may download a data file of the application program from the server 10 through a network, and may also upload operation data generated by a user operating the application program to the server 10, where the terminal 11 may be implemented in various forms, for example, may be a mobile terminal such as a mobile phone, a tablet computer, a notebook computer, a palm computer, or a fixed terminal such as a desktop computer; the program should include music-like applications.
Those skilled in the art will appreciate that the information recommendation system architecture shown in FIG. 1 does not constitute a limitation of the recommendation information recommendation system, and that the information recommendation system may include more or less components than shown, or combine certain components, or a different arrangement of components.
It should be noted that the embodiment of the present invention may be implemented based on the information recommendation system shown in fig. 1, and the information recommendation device may be the server 10.
Example one
An embodiment of the present invention provides an information recommendation method, as shown in fig. 2, the method includes:
s201: and acquiring historical operation data of the target user for selecting songs aiming at the full amount of history in a preset statistical period.
The information recommendation device acquires time interval operation data of a target user for selecting songs aiming at the whole amount of history according to preset time intervals from the terminal, and correspondingly stores the preset time intervals and the time interval operation data until the history operation data in a preset statistical period is acquired, wherein the preset statistical period comprises at least one preset time interval, and the history operation data comprises time interval operation data in each time interval in the at least one preset time interval; for example, the preset time period may be one day, one hour, or the like.
Illustratively, taking a music application program as an example, when a user uses the music application program on a terminal, the terminal records login account information (e.g., a user identification number (ID) of the user, the terminal also records operation data, the operation data is generated after the user performs a selection operation on a song in the music application program, the terminal correspondingly stores the user ID and the operation data, and sends the time interval operation data in each time interval to a server in combination with the user ID, and the server stores the time interval operation data of each user according to the user ID; wherein the selecting operation for the song comprises: the operation data includes song identification (e.g., song ID) of the song, and operation type of the selection operation.
S202: obtaining historical evaluation data of each historical selected song in the total historical selected songs based on the corresponding relation between preset operation data and operation scores and the historical operation data; the historical rating data is used to characterize the likeness of the song by the target user.
The information recommending device determines historical operation data of each user in the target users from the historical operation data, calculates operation scores of time interval operation data corresponding to each time interval in the historical operation data of each user according to the corresponding relation between preset operation data and the operation scores, obtains time interval evaluation data corresponding to each time interval, and further obtains historical evaluation data consisting of the time interval evaluation data corresponding to all the time intervals of each user.
For example, the preset time period is one day, the time period evaluation data corresponding to each time period is calculated, that is, the like degree of the user to different songs on a single day is calculated, the information recommendation apparatus may set a corresponding like score for each operation category according to the daily usage habit, the like score reflects the like degree of the user expressed by the selection operation of each operation category, and the preset operation data and the corresponding operation score are obtained as shown in table 1 below:
TABLE 1
| Operating data | Score value |
| Operation of cutting song | -0.5 |
| Marking favorite operations | 2 |
| Marking dislike operations | -2 |
| |
4 |
| Comment operation | 1 |
| Sharing operations | 2 |
| Operation of collection | 2 |
| Audition operations | 0.5 |
| |
4 |
| Set to color Ring operation | 3 |
| Search operations | 1 |
Illustratively, the information recommendation apparatus calculates history evaluation data of each of the total number of history selection songs on the basis of the correspondence of table 1 and the operation type of the selection operation in the history operation data, and records in the following form: "user ID-song ID-operation score-date", the date is time information corresponding to each time period in at least one preset time period.
Illustratively, after the server acquires the period operation data for each time period, the period evaluation data for each of the historically selected songs by each of the users is calculated using the period operation data, for example, the period operation data for the historically selected song a includes: in one day, the user B searches 5 times, shares 4 times, collects 1 time, reviews 2 times, downloads 2 times, and cuts songs 3 times, then according to the corresponding relationship of table 1, obtains formula (1), and calculates by using formula (1) that the time period evaluation data of the user B for historically selecting songs a in one day is 23.5:
5×1+4×2+1×2+2×1+2×4+3×(-0.5)＝23.5 (1)
when the time information corresponding to the day is 2018, 10, month and 11, the time period evaluation data of the history selection song a can be recorded in the following form: user B-song a-23.5-20181011.
S203: and performing time attenuation processing on the historical evaluation data to obtain historical statistical evaluation data of each historical selected song.
The historical evaluation data of each user acquired by the information recommendation device is composed of time period evaluation data corresponding to each time period, and time attenuation processing is carried out on the time period evaluation data corresponding to all the time periods to obtain historical statistical evaluation data corresponding to each user.
It should be noted that, because the preference of the user for the song may be dynamically changed, the current preference of the user often cannot be accurately determined only according to the time period evaluation data within a preset time period, so that the information recommendation device determines the interest and hobbies of the user according to the historical evaluation data within a preset statistical period; for example, a preset statistical period of 90 days is used to determine the total favorite score of the user for each history selection song within 90 days according to 90 single-day favorite scores of the user for each history selection song.
In some embodiments, the information recommendation device obtains an attenuation coefficient corresponding to each time period according to a corresponding relation between a preset attenuation coefficient and time and the time information represented by each time period; and according to the attenuation coefficient, summing the time interval evaluation data corresponding to at least one preset time interval to obtain historical statistical evaluation data.
Illustratively, because the preference of the user to the song is dynamically changed, the recent period evaluation data of the user often reflects the current preference of the user to the song more accurately than the early period evaluation data, that is, the reflecting accuracy of the recent preference of the user is gradually attenuated by the period evaluation data along with the increase of the time interval with the current time, so that the information recommendation device determines the attenuation coefficient corresponding to each time period according to the time interval of the initial time corresponding to the time information corresponding to each time period and the preset statistical period at the end time of the preset statistical period, and performs weighted summation on the period evaluation data corresponding to all the time periods according to the attenuation coefficient to obtain historical statistical evaluation data.
Illustratively, the correspondence between the preset attenuation coefficient and time is expressed as formula (2):
N(t)＝N0e-t/T (2)
wherein N is0The initial attenuation coefficient represents the starting moment of the preset statistical period, T represents the time interval between the time information corresponding to each time period and the starting moment of the preset statistical period, N (T) represents the attenuation coefficient corresponding to each time period, e represents an index, and T represents the ending moment of the preset statistical period.
S204: and sequencing the whole quantity of history selected songs according to the history statistical evaluation data to obtain a sequencing result of the whole quantity of history selected songs.
And the information recommending device performs ascending sorting or descending sorting on the total quantity of history selected songs according to the historical statistical evaluation data of each user and the size of the historical statistical evaluation data to obtain a sorting result.
S205: and recommending the target song according to the sequencing result and the preset full-scale song list.
The information recommending device selects n reference songs from the full-volume history for each user based on the sequencing result, wherein n is an integer larger than 0, the historical statistical evaluation data of the n reference songs are larger than the historical statistical evaluation data of other songs in the full-volume history selection songs, the n reference songs are the n reference songs which are the favorite of each user, and the target song is determined according to the n reference songs and the preset full-volume song list of the music application program.
It should be noted that, as shown in fig. 3, the step S205 is implemented as steps S2051 to S2053, and includes:
s2051: according to the sorting result, determining a reference song from the songs selected from the full-scale history;
the information recommendation device selects n reference songs from the full amount of history selection songs.
S2052: determining a reference song list from a preset total song list according to the reference song;
the information recommending device determines all the song sheets containing each reference song from the preset total song sheets for each reference song, and takes all the song sheets as sub-reference song sheets of each reference song so as to obtain the reference song sheets consisting of all the sub-reference song sheets.
S2053: and recommending the target song according to the reference song list.
The information recommendation device determines a target song from the songs of the reference menu and recommends the target song to the user by transmitting the target song to the terminal.
In some embodiments, the reference menu includes a sub-reference menu corresponding to each of the reference songs; and the information recommending device determines the target song from the reference song list according to the repeated times of each song in the sub-reference song list and recommends the target song.
Illustratively, one reference song A of n reference songs that the user likes mostLFor example, L is an integer greater than 0 and not greater than n, and the information recommendation device may determine that the reference song A is included from the preset total volume menuLThe sub-reference song list is expressed as { song list L1, song list L2, … and song list LN }, N is an integer larger than 2, the repetition times of each song in the sub-reference song list are counted, the repetition times of all songs in the sub-reference song list are sorted, and the order sequence number of the times of each song in the sub-reference song list is obtained; determining the songs with the order sequence numbers belonging to the top q bits as reference songs ALCorresponding song T to be recommendedLQ is a positive integer greater than 0, and then at least one song to be recommended consisting of the songs to be recommended corresponding to the n reference songs is obtained, wherein the song to be recommended TLAt least one song may be included.
In some embodiments, the information recommendation device ranks the historical statistical evaluation data of each reference song to obtain a data ranking sequence number of each reference song; sequencing the repeated times of each song in the sub-reference song list to obtain a sequencing number of the times of each song in the sub-reference song list; determining the songs to be recommended corresponding to each reference song from the sub-reference song list according to the time sequencing sequence number to obtain at least one song to be recommended; determining the favorite score of the song to be recommended according to the data sorting sequence number and the frequency sorting sequence number of the song to be recommended; and determining a target song from at least one song to be recommended according to the preference score.
Illustratively, the information recommendation device performs descending sorting on the historical statistical evaluation data of n reference songs to obtain a data sorting sequence number g of each reference song; determining the frequency sequencing sequence number h of the song to be recommended corresponding to each reference song from the frequency sequencing sequence number of each song in the sub-reference song list; calculating the preference Score of the song to be recommended according to formula (3):
and according to the preference score of the songs to be recommended, determining a target song with a higher preference score from at least one song to be recommended, and recommending the target song to the user.
It should be noted that, after step S2052, the information recommendation device quantifies the songs of each of the reference menus according to the preset song attributes to obtain the reference attribute characteristics of each of the reference menus; obtaining scoring data containing scoring results of each song in the reference song list according to a preset song list scoring model and the reference attribute characteristics; presetting a corresponding relation between the characteristic attribute characteristics of the singing sheet scoring model and scoring data; and when the first scoring data in the scoring data is smaller than a preset scoring threshold, deleting the singing sheet corresponding to the first scoring data in the reference singing sheet from the reference singing sheet to obtain an updated reference singing sheet.
Illustratively, the information recommendation device quantifies each song in each song list according to preset song attributes to avoid the influence of the inferior song list so as to obtain a multi-dimensional attribute vector of each song, wherein each song attribute in the preset song attributes corresponds to one-dimensional data; generating a multi-dimensional attribute matrix by using the attribute vectors of all songs in each song list, wherein the multi-dimensional attribute matrix is the reference attribute characteristic of each song list; calculating a scoring result of each song bill by using a preset song bill scoring model and a multi-dimensional attribute matrix of each song bill; and finally, screening the reference singing lists according to the scoring result of each singing list.
Illustratively, the preset song attributes include the year of release of the song, the age of the singer, the sex information of the singer, the rhythm information of the song, the tone information of the song, etc., the sex information of the singer includes that the sex information of the singer of a male singer is equal to 0 and the sex information of the singer of a female singer is equal to 1, the rhythm information of the song is the beats per minute of the song, the tone information of the song includes that the CDEFGAB tones of the song respectively correspond to 1-7, the male singer who is born in 1979 issues the song in 2007, the song is released in A tone and 60 beats per minute, and the multidimensional attribute vector of the released song is expressed as {2007, 1979, 0, 60, 6 }; the position of each song attribute in the preset song attributes in the multidimensional attribute vector can be changed, and the song attribute sequence of the multidimensional attribute vector is only required to be consistent with the song attribute sequence corresponding to the preset song menu scoring model.
In some embodiments, the information recommendation device obtains sample operation data of a target user for a full amount of sample songs in a preset training period; determining a positive sample song list and a negative sample song list from a preset total amount of song lists according to the sample operation data; acquiring positive scoring data corresponding to the positive sample song list, wherein the positive sample song list and the positive scoring data form a positive sample; acquiring negative scoring data corresponding to the negative sample song list, wherein the negative sample song list and the negative scoring data form a negative sample; training a preset initial singing bill scoring model by using a positive sample and a negative sample until the accuracy of the trained initial singing bill scoring model is not less than a preset accuracy threshold; and taking the trained initial singing bill scoring model as a preset singing bill scoring model.
Illustratively, the information recommendation device acquires sample operation data of a target user for a full amount of sample songs in a preset training period; obtaining sample statistical evaluation data of each sample song according to the sample operation data; sequencing the full amount of sample songs according to the sample statistical evaluation data to obtain a sample sequencing result of the full amount of sample songs; determining a positive sample song list and a negative sample song list from a preset total amount of song lists according to a sample sequencing result, wherein the positive sample song list is the song list containing the favorite songs of the user, and the negative sample song list is the song list containing the favorite songs of the user; quantifying the songs of each song list in the positive sample song list according to the preset song attributes to obtain the positive attribute characteristics of each song list in the positive sample song list, and quantifying the songs of each song list in the negative sample song list according to the preset song attributes to obtain the negative attribute characteristics of each song list in the negative sample song list; the positive attribute features and the positive scoring data form positive examples, and the negative attribute features and the negative scoring data form negative examples.
Illustratively, when the positive score data is 1 and the negative score data is-1, the information recommending means deletes the singing sheet whose score result is less than 0 from the reference singing sheet according to the score result of each singing sheet.
It should be noted that, the full sample songs correspond to the full history selection songs, and the full sample songs and the full history selection songs are only different in acquisition time; therefore, the process of obtaining the sample statistical evaluation data and the sample ranking result based on the sample operation data is consistent with the process of obtaining the historical statistical evaluation data and the ranking result based on the historical operation data.
Illustratively, an Adboost algorithm may be adopted to obtain the preset singing bill scoring model, specifically including: setting an initial song list scoring model, wherein the initial song list scoring model comprises weak classifiers corresponding to every two adjacent dimensional data; training a weak classifier by using every two adjacent dimensional data corresponding to the positive sample and the negative sample according to the initial data weight of every dimensional data in the multi-dimensional data, calculating the classifier weight of every weak classifier, and generating a strong classifier by using every weak classifier and the classifier weight; calculating the accuracy of the strong classifier, and when the accuracy is smaller than a preset accuracy threshold, adjusting the initial data weight until the accuracy of the strong classifier is not smaller than the preset accuracy threshold; and taking the strong classifier as a preset singing bill grading model.
Illustratively, the preset song attributes include 5 song attributes, which are respectively the song release year, the artist age, the artist gender information, the song rhythm information, and the song tone information, and the strong classifier is generated by K ═ 5 weak classifiers, for example, when every two adjacent dimensional data are expressed as { first song attribute x, second song attribute y }, the sample input is derived as { (x) from the positive sample and the negative sample1，y1),(x2,y2),...，(xm,ym) Sample output is { -1, +1}, m is an integer greater than 0; training a preset initial song list scoring model by using sample input and sample output according to the initial data weight to obtain a weak classifier G corresponding to the first song attribute x1(x) Further obtaining a weak classifier corresponding to each song attribute; wherein, firstThe starting data weight D (1) is shown in equation (4):
D(1)＝(w11,w12,…,w1m) (4)
wherein the content of the first and second substances,
further, for weak classifiers Gk(x) Training is carried out, and specifically the method comprises the following steps: using weak classifiers Gk(x) Corresponding data weight D (k) for training the initial singing sheet scoring model to obtain weak classifier Gk(x) Where K is 1, 2.. K, and the data weight d (K) is as shown in equation (5):
D(k)＝(wk1,wk2,…,wkm) (5)
wherein the content of the first and second substances,
weak classifier G is calculated using equation (6)k(x) E classification error rate ofk：
Wherein, the weak classifier Gk(xi) Input (x)i，yi) Then outputting a scoring result which is not equal to yiAnd (4) correspondingly presetting a threshold value I.
Calculating the weak classifier G using equation (7)k(x) By weight of classifier ak：
Calculating the data weight D (k +1) corresponding to the weak classifier k +1 by using the formula (8):
wherein Z iskIs a normalization factor, as shown in equation (9):
the strong classifier is generated using equation (10):
it should be noted that the larger the classification error rate of the weak classifier is, the smaller the classifier weight of the weak classifier is.
The information recommending device can perform time attenuation on the historical evaluation data in the preset statistical period to obtain the historical statistical evaluation data after the time attenuation, and further obtain the sequencing result of the whole quantity of historical selected songs, and recommend the target song according to the sequencing result.
Example two
The further description will be made based on the same inventive concept of the first embodiment.
An embodiment of the present invention provides an information recommendation apparatus 4, as shown in fig. 4, where the apparatus 4 includes: an acquisition unit 40, a calculation unit 41, an attenuation processing unit 42, and a recommendation unit 43; wherein the content of the first and second substances,
the obtaining unit 40 is configured to obtain historical operation data of a target user for selecting songs for a full amount of history in a preset statistical period;
the calculating unit 41 is configured to obtain historical evaluation data of each history selected song in the total number of history selected songs based on a preset corresponding relationship between operation data and operation scores and the history operation data; the historical evaluation data is used for representing the likeness degree of the target user to the song;
the attenuation processing unit 42 is configured to perform time attenuation processing on the historical evaluation data to obtain historical statistical evaluation data of each historically selected song;
the recommending unit 43 is configured to sort the full-scale historically-selected songs according to the historical statistical evaluation data to obtain a sorting result of the full-scale historically-selected songs; and recommending the target song according to the sequencing result and the preset full-scale song list.
In some embodiments, the preset statistical period comprises at least one preset time period; the historical evaluation data comprises time interval evaluation data corresponding to each time interval in the at least one preset time interval;
the attenuation processing unit 42 is specifically configured to obtain an attenuation coefficient corresponding to each time period according to a corresponding relationship between a preset attenuation coefficient and time and the time information represented by each time period; and according to the attenuation coefficient, summing the time interval evaluation data corresponding to the at least one preset time interval to obtain the historical statistical evaluation data.
In some embodiments, the recommending unit 43 is specifically configured to determine a reference song from the full-volume history selection songs according to the sorting result; determining a reference song list from the preset full-volume song list according to the reference song; and recommending the target song according to the reference song list.
In some embodiments, the reference menu includes a sub-reference menu corresponding to each of the reference songs;
the recommending unit 43 is specifically configured to determine the target song from the reference list according to the repetition number of each song in the sub-reference list, and recommend the target song.
In some embodiments, the recommending unit 43 is specifically configured to sort the historical statistical evaluation data of each reference song to obtain a data sorting order number of each reference song; sequencing the repeated times of each song in the sub-reference song list to obtain a sequencing sequence number of the times of each song in the sub-reference song list; determining the songs to be recommended corresponding to each reference song from the sub-reference song list according to the time sequencing sequence number to obtain at least one song to be recommended; determining the favorite score of the song to be recommended according to the data sorting serial number and the frequency sorting serial number of the song to be recommended; and determining the target song from the at least one song to be recommended according to the preference score.
In some embodiments, the recommending unit 43 is further configured to, after the reference menu is determined from the preset total volume menu according to the reference song, quantify the songs of each menu in the reference menu according to a preset song attribute, so as to obtain a reference attribute feature of each menu in the reference menu; obtaining scoring data containing scoring results of each song in the reference song list according to a preset song list scoring model and the reference attribute characteristics; the preset singing bill grading model represents the corresponding relation between attribute characteristics and grading data; and when first scoring data in the scoring data is smaller than a preset scoring threshold value, deleting the singing sheet corresponding to the first scoring data in the reference singing sheet from the reference singing sheet to obtain an updated reference singing sheet.
In some embodiments, the model training unit is configured to, before obtaining score data including a score result of each of the reference singals according to a preset singal sheet scoring model and the reference attribute features, obtain sample operation data of a target user for a full number of sample songs in a preset training period; determining a positive sample song list and a negative sample song list from the preset full-volume song list according to the sample operation data; acquiring positive scoring data corresponding to the positive sample song list, wherein the positive sample song list and the positive scoring data form a positive sample; acquiring negative scoring data corresponding to the negative sample song list, wherein the negative sample song list and the negative scoring data form a negative sample; training a preset initial singing bill scoring model by using the positive sample and the negative sample until the accuracy of the trained initial singing bill scoring model is not less than a preset accuracy threshold; and taking the trained initial singing bill scoring model as the preset singing bill scoring model.
In practical applications, the obtaining Unit 40, the calculating Unit 41, the attenuation Processing Unit 42, and the recommending Unit 43 may be implemented by a processor 44 located on the information recommending apparatus 4, specifically, implemented by a CPU (Central Processing Unit), an MPU (micro processor Unit), a DSP (Digital Signal Processing) or a Field Programmable Gate Array (FPGA).
An embodiment of the present invention further provides an information recommendation apparatus 4, as shown in fig. 5, where the apparatus 4 includes: a processor 44, a memory 45 and a communication bus 46, the memory 45 communicating with the processor 44 via the communication bus 46, the memory 45 storing one or more programs executable by the processor 44, the one or more programs, when executed, causing the processor 44 to perform any one of the information recommendation methods as described in the previous embodiments.
In practical applications, the Memory 45 may be a volatile first Memory (volatile Memory), such as a Random-Access Memory (RAM); or a non-volatile first Memory (non-volatile Memory), such as a Read-Only first Memory (ROM), a flash Memory (flash Memory), a Hard Disk Drive (HDD) or a Solid-State Drive (SSD); or a combination of first memories of the above kind and provides programs and data to the processor 44.
The embodiment of the present invention provides a computer-readable storage medium, which stores one or more programs that are executable by one or more processors, and when the programs are executed by the processors 44, the programs implement any one of the information recommendation methods according to the foregoing embodiments.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention.
Claims (14)
1. An information recommendation method, characterized in that the method comprises:
acquiring historical operation data of a target user for selecting songs aiming at the full amount of history in a preset statistical period; the preset statistical period comprises at least one preset time period;
obtaining historical evaluation data of each historical selected song in the total historical selected songs based on the corresponding relation between preset operation data and operation scores and the historical operation data; the historical evaluation data is used for representing the likeness degree of the target user to the song;
determining an attenuation coefficient corresponding to each preset time period according to the time interval between the time information corresponding to each preset time period and the initial time corresponding to the preset statistical period;
performing time attenuation processing on the historical evaluation data according to the attenuation coefficient of each preset time period to obtain historical statistical evaluation data of each historical selected song;
sorting the full-scale historical selected songs according to the historical statistical evaluation data to obtain a sorting result of the full-scale historical selected songs;
according to the sorting result, determining a reference song from the songs selected from the full-scale history;
determining a reference song list from the preset total song list according to the reference song;
obtaining an updated reference song list according to the multidimensional attribute matrix of each song list in the reference song list;
and recommending the target song according to the updated reference song list.
2. The method of claim 1, wherein the historical rating data comprises time period rating data corresponding to each of the at least one preset time period;
the time attenuation processing is performed on the historical evaluation data according to the attenuation coefficient of each preset time period to obtain historical statistical evaluation data of each historical selected song, and the method comprises the following steps:
and summing the time interval evaluation data corresponding to the at least one preset time interval according to the attenuation coefficient to obtain the historical statistical evaluation data.
3. The method of claim 1, wherein the updated reference menu comprises a sub-reference menu corresponding to each of the reference songs;
the recommending the target song according to the updated reference song list comprises:
and determining the target song from the updated reference list according to the repeated times of each song in the sub-reference list, and recommending the target song.
4. The method of claim 3, wherein said determining the target song from the reference menu based on the number of repetitions of each song in the sub-reference menu comprises:
sorting the historical statistical evaluation data of each reference song to obtain a data sorting sequence number of each reference song;
sequencing the repeated times of each song in the sub-reference song list to obtain a sequencing sequence number of the times of each song in the sub-reference song list;
determining the songs to be recommended corresponding to each reference song from the sub-reference song list according to the time sequencing sequence number to obtain at least one song to be recommended;
determining the favorite score of the song to be recommended according to the data sorting sequence number and the frequency sorting sequence number of the song to be recommended;
and determining the target song from the at least one song to be recommended according to the preference score.
5. The method of claim 1, wherein said deriving an updated reference song list based on a multi-dimensional attribute matrix for each of said reference song lists comprises:
according to preset song attributes, quantifying the songs of each of the reference singals to obtain a multi-dimensional attribute matrix of each of the reference singals;
obtaining scoring data containing scoring results of each song in the reference song list according to a preset song list scoring model and the multi-dimensional attribute matrix; the preset singing menu scoring model represents the corresponding relation between the multi-dimensional attribute matrix and the scoring data;
and when first scoring data in the scoring data is smaller than a preset scoring threshold value, deleting the singing sheet corresponding to the first scoring data in the reference singing sheet from the reference singing sheet to obtain an updated reference singing sheet.
6. The method as claimed in claim 5, wherein before the obtaining of the scoring data including the scoring result of each song in the reference song list according to the preset song list scoring model and the multi-dimensional attribute matrix, the method further comprises:
in a preset training period, obtaining sample operation data of a target user for a full amount of sample songs;
determining a positive sample song list and a negative sample song list from the preset full-volume song list according to the sample operation data;
acquiring positive scoring data corresponding to the positive sample song list, wherein the positive sample song list and the positive scoring data form a positive sample;
acquiring negative scoring data corresponding to the negative sample song list, wherein the negative sample song list and the negative scoring data form a negative sample;
training a preset initial singing bill scoring model by using the positive sample and the negative sample until the accuracy of the trained initial singing bill scoring model is not less than a preset accuracy threshold;
and taking the trained initial singing bill scoring model as the preset singing bill scoring model.
7. An information recommendation apparatus, characterized in that the apparatus comprises: the device comprises an acquisition unit, a calculation unit, an attenuation processing unit and a recommendation unit; wherein the content of the first and second substances,
the acquisition unit is used for acquiring historical operation data of a target user for selecting songs aiming at the full amount of history in a preset statistical period; the preset statistical period comprises at least one preset time period;
the computing unit is used for obtaining historical evaluation data of each historical selected song in the total historical selected songs based on the corresponding relation between preset operation data and operation scores and the historical operation data; the historical evaluation data is used for representing the likeness degree of the target user to the song;
the attenuation processing unit is used for determining an attenuation coefficient corresponding to each preset time period according to the time information corresponding to each preset time period and the time interval of the initial time corresponding to the preset statistical period; performing time attenuation processing on the historical evaluation data according to the attenuation coefficient of each preset time period to obtain historical statistical evaluation data of each historical selected song;
the recommending unit is used for sequencing the full-amount historical selected songs according to the historical statistical evaluation data to obtain a sequencing result of the full-amount historical selected songs; according to the sorting result, determining a reference song from the songs selected from the full-scale history; determining a reference song list from the preset total song list according to the reference song; obtaining an updated reference song list according to the multidimensional attribute matrix of each song list in the reference song list; and recommending the target song according to the updated reference song list.
8. The apparatus according to claim 7, wherein the historical evaluation data comprises period evaluation data corresponding to each of the at least one preset time period;
the attenuation processing unit is specifically configured to sum the time period evaluation data corresponding to the at least one preset time period according to the attenuation coefficient of each preset time period to obtain the historical statistical evaluation data.
9. The apparatus of claim 7, wherein the updated reference menu comprises sub-reference menus corresponding to each of the reference songs;
and the recommending unit is specifically used for determining the target song from the updated reference list according to the repetition times of each song in the sub-reference list and recommending the target song.
10. The apparatus of claim 8,
the recommending unit is specifically configured to sort the historical statistical evaluation data of each reference song to obtain a data sorting sequence number of each reference song; sequencing the repeated times of each song in the sub-reference song list to obtain the sequencing sequence number of the times of each song in the sub-reference song list; determining the songs to be recommended corresponding to each reference song from the sub-reference song list according to the time sequencing sequence number to obtain at least one song to be recommended; determining the favorite score of the song to be recommended according to the data sorting serial number and the frequency sorting serial number of the song to be recommended; and determining the target song from the at least one song to be recommended according to the preference score.
11. The apparatus of claim 10,
the recommending unit is further used for quantifying the songs of each of the reference singals according to preset song attributes to obtain a multi-dimensional attribute matrix of each of the reference singals; obtaining scoring data containing scoring results of each song in the reference song list according to a preset song list scoring model and the multi-dimensional attribute matrix; the preset singing menu scoring model represents the corresponding relation between the multi-dimensional attribute matrix and the scoring data; and when first scoring data in the scoring data is smaller than a preset scoring threshold value, deleting the singing sheet corresponding to the first scoring data in the reference singing sheet from the reference singing sheet to obtain an updated reference singing sheet.
12. The apparatus of claim 11, further comprising:
the model training unit is used for acquiring sample operation data of a target user for a full amount of sample songs in a preset training period before obtaining scoring data containing a scoring result of each song in the reference song list according to a preset song list scoring model and reference attribute characteristics; determining a positive sample song list and a negative sample song list from the preset full-volume song list according to the sample operation data; acquiring positive scoring data corresponding to the positive sample song list, wherein the positive sample song list and the positive scoring data form a positive sample; acquiring negative scoring data corresponding to the negative sample song list, wherein the negative sample song list and the negative scoring data form a negative sample; training a preset initial singing bill scoring model by using the positive sample and the negative sample until the accuracy of the trained initial singing bill scoring model is not less than a preset accuracy threshold; and taking the trained initial singing bill scoring model as the preset singing bill scoring model.
13. An information recommendation apparatus, characterized in that the apparatus comprises: a processor, a memory and a communication bus, the memory in communication with the processor through the communication bus, the memory storing one or more programs executable by the processor, the one or more programs, when executed, causing the processor to perform the method of any of claims 1-6.
14. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a program which, when executed by at least one processor, causes the at least one processor to perform the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811642134.8A CN109408665B (en) | 2018-12-29 | 2018-12-29 | Information recommendation method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811642134.8A CN109408665B (en) | 2018-12-29 | 2018-12-29 | Information recommendation method and device and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109408665A CN109408665A (en) | 2019-03-01 |
| CN109408665B true CN109408665B (en) | 2021-11-23 |
Family
ID=65462618
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811642134.8A Active CN109408665B (en) | 2018-12-29 | 2018-12-29 | Information recommendation method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109408665B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109982155B (en) * | 2019-03-25 | 2021-10-12 | 北京奇艺世纪科技有限公司 | Playlist recommendation method and system |
| CN109992694A (en) * | 2019-04-17 | 2019-07-09 | 重庆誉存大数据科技有限公司 | A kind of music intelligent recommendation method and system |
| CN110363569A (en) * | 2019-06-17 | 2019-10-22 | 深圳壹账通智能科技有限公司 | Data product recommended method, device, computer equipment and storage medium |
| CN110599052B (en) * | 2019-09-19 | 2023-07-21 | 携程计算机技术（上海）有限公司 | OTA hotel evaluation method, system, electronic equipment and medium |
| CN110659425B (en) * | 2019-09-25 | 2022-05-17 | 秒针信息技术有限公司 | Resource allocation method and device, electronic equipment and readable storage medium |
| CN110996177B (en) * | 2019-11-27 | 2022-04-22 | 北京爱奇艺智慧娱乐科技有限公司 | Video recommendation method, device and equipment for video-on-demand cinema |
| CN111078931B (en) * | 2019-12-10 | 2023-08-01 | 腾讯科技（深圳）有限公司 | Song list pushing method, device, computer equipment and storage medium |
| CN110990621B (en) * | 2019-12-16 | 2023-10-13 | 腾讯科技（深圳）有限公司 | Song recommendation method and device |
| CN113127674B (en) * | 2019-12-31 | 2023-07-21 | 中移(成都)信息通信科技有限公司 | Song list recommendation method and device, electronic equipment and computer storage medium |
| CN112052379A (en) * | 2020-01-20 | 2020-12-08 | 河南紫联物联网技术有限公司 | Intelligent terminal, and recommendation method and system of intelligent household equipment |
| CN111383049A (en) * | 2020-02-21 | 2020-07-07 | 中国平安财产保险股份有限公司 | Product recommendation method, device and storage medium |
| CN111581452B (en) * | 2020-03-26 | 2023-10-17 | 浙江口碑网络技术有限公司 | Recommendation object data obtaining method and device and electronic equipment |
| CN111488484B (en) * | 2020-04-16 | 2023-04-28 | 北京雷石天地电子技术有限公司 | Song recommendation method, medium and device based on class index distribution and TF-IDF characteristics |
| CN111626615A (en) * | 2020-05-27 | 2020-09-04 | 腾讯音乐娱乐科技（深圳）有限公司 | Method, device and computer storage medium for determining singing order score |
| CN112231511A (en) * | 2020-10-20 | 2021-01-15 | 腾讯音乐娱乐科技（深圳）有限公司 | Neural network model training method and song mining method and device |
| CN113536027A (en) * | 2021-07-27 | 2021-10-22 | 咪咕音乐有限公司 | Music recommendation method, device, equipment and computer readable storage medium |
| CN113570257A (en) * | 2021-07-30 | 2021-10-29 | 北京房江湖科技有限公司 | Index data evaluation method and device based on scoring model, medium and equipment |
| CN113688314B (en) * | 2021-08-13 | 2024-03-19 | 今彩慧健康科技(苏州)有限公司 | Physical therapy store recommending method and device |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678388B (en) * | 2012-09-19 | 2016-09-14 | 腾讯科技（深圳）有限公司 | Online Music recommends method and device |
| CN104090912B (en) * | 2014-06-10 | 2017-03-15 | 腾讯科技（深圳）有限公司 | Information-pushing method and device |
| US20180285893A1 (en) * | 2017-03-31 | 2018-10-04 | International Business Machines Corporation | Determining music to influence customer behavior |
| CN108595550A (en) * | 2018-04-10 | 2018-09-28 | 南京邮电大学 | A kind of music commending system and recommendation method based on convolutional neural networks |
-
2018
- 2018-12-29 CN CN201811642134.8A patent/CN109408665B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109408665A (en) | 2019-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109408665B (en) | Information recommendation method and device and storage medium | |
| US11182422B2 (en) | Media unit retrieval and related processes | |
| CN107644089B (en) | Hot event extraction method based on network media | |
| CN106651542B (en) | Article recommendation method and device | |
| CN110543598B (en) | Information recommendation method and device and terminal | |
| JP4524709B2 (en) | Information processing apparatus and method, and program | |
| CN107885745B (en) | Song recommendation method and device | |
| CN109783671B (en) | Method for searching picture by picture, computer readable medium and server | |
| CN108460082B (en) | Recommendation method and device and electronic equipment | |
| WO2018157625A1 (en) | Reinforcement learning-based method for learning to rank and server | |
| CN106951527B (en) | Song recommendation method and device | |
| CN111191111B (en) | Content recommendation method, device and storage medium | |
| WO2011054245A1 (en) | Mobile search method, device and system | |
| CN109597858B (en) | Merchant classification method and device and merchant recommendation method and device | |
| CN108665148B (en) | Electronic resource quality evaluation method and device and storage medium | |
| CN112287094B (en) | Similar case text retrieval system | |
| US20130346385A1 (en) | System and method for a purposeful sharing environment | |
| CN108133058A (en) | A kind of video retrieval method | |
| JP5048852B2 (en) | Search device, search method, search program, and computer-readable recording medium storing the program | |
| CN106919588A (en) | A kind of application program search system and method | |
| CN113656699B (en) | User feature vector determining method, related equipment and medium | |
| CN111160699A (en) | Expert recommendation method and system | |
| CN112328775B (en) | Case text information retrieval system | |
| CN109462635B (en) | Information pushing method, computer readable storage medium and server | |
| CN107315807B (en) | Talent recommendation method 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 |