CN111523050B - Content recommendation method, server and storage medium - Google Patents
Content recommendation method, server and storage medium Download PDFInfo
- Publication number
- CN111523050B CN111523050B CN202010298828.5A CN202010298828A CN111523050B CN 111523050 B CN111523050 B CN 111523050B CN 202010298828 A CN202010298828 A CN 202010298828A CN 111523050 B CN111523050 B CN 111523050B
- Authority
- CN
- China
- Prior art keywords
- recommendation
- frequency
- recommended
- target user
- adjusted
- 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
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000006399 behavior Effects 0.000 claims description 52
- 230000003993 interaction Effects 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 5
- 230000009471 action Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 239000002537 cosmetic Substances 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 102100029469 WD repeat and HMG-box DNA-binding protein 1 Human genes 0.000 description 2
- 101710097421 WD repeat and HMG-box DNA-binding protein 1 Proteins 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Primary Health Care (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the application relates to the technical field of computers and discloses a content recommendation method, a server and a storage medium. In the application, social information of a target user is obtained; the social information comprises social friends of the target user, affinity between the target user and the social friends, and recommendation directions corresponding to the social friends; selecting a heuristic recommendation direction from the recommendation directions according to the intimacy; and determining a recommendation frequency according to the intimacy, and recommending the content corresponding to the heuristic recommendation direction according to the recommendation frequency. By the method, the method and the device for actively recommending the content corresponding to the possible interest directions to the user is achieved.
Description
Technical Field
The embodiment of the application relates to the technical field of computers, in particular to a content recommendation method, a server and a storage medium.
Background
With the development of internet technology, listening to songs, watching videos, playing games and the like through a mobile terminal becomes a main form of daily leisure and relaxation of people, and in order to bring better personalized experience to users, the conventional APP can recommend interesting contents to the users according to habits of the users, so that the use effect of thousands of people and thousands of faces is achieved. At present, APP mainly calculates the interest and hobby direction of a user through a collaborative filtering algorithm and a tag similarity algorithm, and recommends content corresponding to the interest and hobby direction to the user.
The inventor finds that at least the following problems exist in the prior art: the collaborative filtering algorithm and the label similarity algorithm can only recommend contents corresponding to the interest directions to the user in the interest directions determined by the user, and cannot actively recommend contents corresponding to the possible interest directions to the user.
Disclosure of Invention
The embodiment of the application aims to provide a content recommendation method, a server and a storage medium, so that content corresponding to a possible interest direction is actively recommended to a user.
In order to solve the above technical problems, an embodiment of the present application provides a content recommendation method, including the following steps: acquiring social information of a target user; the social information comprises social friends of the target user, affinity between the target user and the social friends, and recommendation directions corresponding to the social friends; selecting a heuristic recommendation direction from the recommendation directions according to the intimacy; and determining a recommendation frequency according to the intimacy, and recommending the content corresponding to the heuristic recommendation direction according to the recommendation frequency.
The embodiment of the application also provides a server, which comprises: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the content recommendation method described above.
Embodiments of the present application also provide a computer-readable storage medium storing a computer program which, when executed by a processor, implements the content recommendation method described above.
Compared with the prior art, the social information comprises the social friends of the target user, the affinity between the target user and the social friends and the recommendation direction corresponding to the social friends, wherein the recommendation direction is the interest direction.
In addition, the intimacy is obtained by: and determining the affinity according to the interaction behavior and the social relationship between the target user and the social friends. The affinity determined according to the interaction behavior and the social relationship is more accurate, so that the method can further and more accurately recommend the content corresponding to the possible interest direction for the user.
In addition, after the recommending the content corresponding to the heuristic recommending direction, the method further comprises: collecting feedback behaviors of the target user on the content corresponding to the heuristic recommendation direction; adjusting the recommended frequency according to the feedback behavior to obtain an adjusted recommended frequency; and executing corresponding operation according to the adjusted recommended frequency. The feedback behavior of the target user can directly reflect the preference degree of the user for the content, so that the recommendation frequency is adjusted according to the feedback behavior, corresponding operation is executed according to the adjusted recommendation frequency, and the content corresponding to the possible preference direction can be further and more accurately recommended for the user.
In addition, the adjusting the recommended frequency according to the feedback behavior to obtain an adjusted recommended frequency includes: matching a score value for representing the favorites of the target user for the content corresponding to the heuristic recommendation direction for the feedback behavior; calculating the numerical value of the sum of the score values according to the feedback behavior; and adjusting the recommended frequency according to the numerical value to obtain the adjusted recommended frequency. The specific implementation mode for obtaining the adjusted recommended frequency is provided, so that the adjusted recommended frequency can be accurately obtained, and corresponding operation can be accurately executed.
In addition, the adjusting the recommended frequency according to the feedback behavior to obtain an adjusted recommended frequency includes: matching a tag for representing the preference degree of the target user for the content corresponding to the heuristic recommendation direction with the feedback behavior, wherein the tag comprises positive feedback representing the higher preference degree and negative feedback representing the lower preference degree; calculating the numerical value of the positive feedback proportion according to the feedback behavior; and adjusting the recommended frequency according to the numerical value to obtain the adjusted recommended frequency. The specific implementation mode for obtaining the adjusted recommended frequency is provided, so that the adjusted recommended frequency can be accurately obtained, and corresponding operation can be accurately executed.
In addition, the adjusting the recommended frequency according to the numerical value to obtain an adjusted recommended frequency includes: if the numerical value is larger than a first preset threshold value, the recommended frequency is adjusted to be the recommended frequency of the recommended direction of the target user, and the adjusted recommended frequency is obtained to be the recommended frequency of the recommended direction of the target user; if the numerical value is smaller than a second preset threshold value, the recommended frequency is adjusted to be zero, and the adjusted recommended frequency is obtained to be zero; and if the numerical value is not smaller than the second preset threshold value and not larger than the first preset threshold value, updating the recommended frequency, and obtaining the adjusted recommended frequency as the updated recommended frequency. Different values correspond to different adjustment modes, and the recommendation frequency can be accurately adjusted according to specific values, so that the adjusted recommendation frequency can be accurately obtained.
In addition, the updating the recommended frequency to obtain the adjusted recommended frequency as the updated recommended frequency includes: if the numerical value is not smaller than a third preset threshold value and smaller than the first preset threshold value, the recommended frequency is increased according to a preset rewarding value, and the adjusted recommended frequency is obtained as the increased recommended frequency; and if the numerical value is not smaller than the second preset threshold value and smaller than the third preset threshold value, reducing the recommended frequency according to a preset punishment value, and obtaining the adjusted recommended frequency as the reduced recommended frequency. And determining that the specific type of the update is to increase the recommended frequency or decrease the recommended frequency according to the magnitude between the numerical value and a third preset threshold value, thereby realizing accurate acquisition of the updated recommended frequency, namely the adjusted recommended frequency.
In addition, recommending the content corresponding to the heuristic recommendation direction according to the recommendation frequency includes: and according to the recommendation frequency, inserting and recommending the content corresponding to the heuristic recommendation direction in the content corresponding to the recommendation direction of the target user. By means of the recommendation insertion mode, the influence on the content corresponding to the recommendation direction of the recommendation target user can be reduced.
Drawings
One or more embodiments are illustrated by way of example and not limitation in the figures of the accompanying drawings, in which like references indicate similar elements, and in which the figures of the drawings are not to be taken in a limiting sense, unless otherwise indicated.
Fig. 1 is a flowchart of a content recommendation method in a first embodiment according to the present application;
fig. 2 is a flowchart of a content recommendation method in a second embodiment according to the present application;
FIG. 3 is a flow chart of one particular implementation of step 205 in a second embodiment of the present application;
FIG. 4 is a flow chart of another implementation of step 205 in a second embodiment of the present application;
fig. 5 is a schematic diagram of a structure of a server according to a third embodiment of the present application.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the embodiments of the present application will be described in detail below with reference to the accompanying drawings. However, it will be understood by those of ordinary skill in the art that in various embodiments of the present application, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, the claimed technical solution of the present application can be realized without these technical details and various changes and modifications based on the following embodiments. The following embodiments are divided for convenience of description, and should not be construed as limiting the specific implementation of the present application, and the embodiments can be mutually combined and referred to without contradiction.
The first embodiment of the present application relates to a content recommendation method, which is applied to a music APP, a video APP, a news APP, an electronic book APP, and the like, and the present embodiment and the following embodiments take the video APP as an example, but are not limited thereto.
The specific flow is shown in fig. 1, and includes:
step 101, obtaining social information of a target user; the social information comprises social friends of the target user, affinity between the target user and the social friends, and recommendation directions corresponding to the social friends.
Specifically, the target user is a user corresponding to the current login account of the video APP, and the video APP can obtain social information of the target user by reading related information of the current login account; the video APP can also obtain the social information of the target user by simultaneously obtaining the current login account of the video APP and the related information of the account of the third party APP installed on the same terminal.
In one example, a video APP obtains social friends of a target user by reading a friend list of a current login account and the like, and obtains recommendation directions corresponding to the social friends by reading interest tags of the social friends; the affinity is obtained by: and determining the affinity according to the interaction behavior and the social relationship between the target user and the social friends.
Specifically, the interaction behavior between the target user and the social friend includes, but is not limited to, the number of times of sending the message, the number of times of sharing, the number of times of praise and the number of comments, the video APP may determine the score corresponding to the number of times of different interaction behaviors according to the corresponding relationship between the interval where the number of times of interaction behaviors is located and the score, the number of times in the interval may be the same or different, and the score corresponding to each interval may also be set according to the actual requirement, which is not specifically limited in this embodiment, for example: number of times [0,40) time division number of 0.1, number of times [40,100) time division number of 0.2, number of times [100,120) time division number of 0.3, number of times [120,140) time division number of 0.4. The video APP may also bring the number of interactions and the score value into a preset function according to a functional relationship between the number of interactions and the score value, to determine score values corresponding to the number of different interactions, for example: the function is y=0.001x, x is the number of interactions and y is the score value.
The video APP may obtain social relationships between the target user and the social friends through a grouping of the acquisition buddy list, for example: the grouping of the friend list is the grouping of family, friends, colleagues, etc.; social relationships between the target user and the social friends can also be obtained by reading keywords in the sent message, for example: a message contains a keyword such as "sister" and "go", and can be considered as family. According to different preset score values of different social relations, determining the score value corresponding to the social relation, for example: the score value of family was 0.6, friends was 0.8, colleagues was 0.3. And determining the affinity between the target user and the social friends according to the product or the sum of the score value corresponding to the interaction behavior between the target user and the social friends and the score value corresponding to the social relationship. The affinity determined according to the interaction behavior and the social relationship is more accurate, so that the method can further and more accurately recommend the content corresponding to the possible interest direction for the user.
In the following description, for example, the social friends of the target user a include friends a, b, c, d and e, if the number of interactions between the target user a and the friends a, b, c, d, e is 100,120, 140, 160, 180, respectively, the score values of the interactions between the target user a and the friends a, b, c, d, e are determined to be 0.3, 0.4, 0.5, 0.6, 0.7 according to the preset correspondence; friends a and e are family members, friends b and c are coworkers, friend d is a friend, score values of social relationships between the target user A and the friends a, b, c, d, e are determined to be 0.6, 0.3, 0.8 and 0.6 according to preset corresponding relationships, and affinity is calculated according to products of the score values, so that the affinities between the target user and the friends a, b, c, d, e are respectively 0.18, 0.12, 0.15, 0.48 and 0.42.
In one example, the video APP may be based on a preset correspondence between a section and a hierarchy where the number of interactions is located, for example: a first level when the number of times is [0,40), a second level when the number of times is [40, 100), a third level when the number of times is [100, 120), a fourth level when the number of times is [120, 140), and determining the corresponding levels of the number of different interaction behaviors, for example: the hierarchy includes high, medium, low, etc.; or the first level, the second level, the third level, the fourth level, the fifth level, etc., and determining the corresponding levels of different social relationships according to the preset social relationship and the corresponding relationship of the levels, for example: . According to preset rules, the rules can be set according to actual requirements, the embodiment is not limited in detail, and the level of intimacy is determined according to the level corresponding to the interaction behavior between the target user and the social friends and the level corresponding to the social relationship. For example: family members are in a second level, friends are in a third level, colleagues are in a first level, the level of interaction between the target user A and the friends a, b, c, d, e is respectively a third level, a fourth level, a fifth level, a sixth level and a seventh level, the level of social relationship between the target user A and the friends a, b, c, d, e is respectively a second level, a first level, a third level and a second level, the preset rule is to accumulate the level numbers, and the intimacy between the target user A and the friends a, b, c, d, e is respectively a fifth level, a sixth level, a ninth level and a ninth level.
In one example, affinity is determined based on interaction behavior between the target user and the social friends. The video APP can determine the score value corresponding to the times of different interaction behaviors according to the corresponding relation between the preset times of the interaction behaviors and the score value, and the score value is used as the affinity; or determining the hierarchy corresponding to the times of different interaction behaviors according to the preset correspondence between the times of the interaction behaviors and the hierarchy, and taking the hierarchy as the affinity. As in the example above, the score values of the interaction between the target user a and the buddy a, b, c, d are 0.3, 0.4, 0.5, 0.6, 0.7, respectively, and the affinity between the target user and the buddy a, b, c, d, e is 0.3, 0.4, 0.5, 0.6, 0.7, respectively. Therefore, when the friend list of the current login user does not have a group, the social relationship cannot be judged according to the group, and the social relationship cannot be judged well according to the chat message, the intimacy between the target user and the social friends can be obtained.
In one example, the affinity is determined based on social relationships between the target user and the social friends. The score value of the social relationship is taken as the affinity, or the hierarchy of the social relationship is taken as the affinity. As in the example described above, the score values of the social relationship between the target user a and the friends a, b, c, d, e are 0.6, 0.3, 0.8, 0.6, and the affinity between the target user and the friends a, b, c, d, e are 0.6, 0.3, 0.8, 0.6, respectively.
And 102, selecting a heuristic recommendation direction from the recommendation directions according to the affinity.
Specifically, the affinity may reflect, to a certain extent, the interest and hobby direction of the target user, and the higher the affinity between the target user and the social friend is, the more interested the target user may be in the content corresponding to the recommendation direction corresponding to the social friend. In an example, the recommended directions of the friends corresponding to the highest N affinities may be used as the heuristic recommended directions according to the affinity ranking, and the value of N may be set according to the actual situation, which is not limited in this embodiment. As described above, the affinities between the target user and the friends a, b, c, d, e are 0.18, 0.12, 0.15, 0.48, and 0.42, the recommendation directions corresponding to the friends a, b, c, d, e are sports, games, cartoons, movies, and make-up, and if the recommendation directions of the friends corresponding to the highest 3 affinities are used as heuristic recommendation directions, the heuristic recommendation directions are movies, make-up, and sports.
In one example, a preset value may be set, and the recommended direction of the friend corresponding to the affinity exceeding the preset value is used as the heuristic recommended direction, if the affinity is a score value, the level is a score value, and if the affinity is at which level, the level is a level value. As an example, the affinities between the target user and the friends a, b, c, d, e are 0.18, 0.12, 0.15, 0.48 and 0.42, the recommended directions corresponding to the friends a, b, c, d, e are sports, games, cartoons, movies and make-up, and if the preset score value is 0.3, the tentative recommended directions are movies and make-up. For another example, the affinities between the target user a and the friends a, b, c, d, e are respectively a fifth level, a sixth level, a ninth level and a ninth level, the recommended directions corresponding to the friends a, b, c, d, e are sports, games, cartoons, movies and make-up, and if the preset level value is the sixth level, the tentative recommended directions are movies and make-up.
And 103, determining a recommendation frequency according to the intimacy, and recommending the content corresponding to the heuristic recommendation direction according to the recommendation frequency.
Specifically, the video APP stores a correspondence between the affinity and the recommended frequency, or stores a functional relationship between the affinity and the recommended frequency, and the higher the affinity is, the higher the recommended frequency is, the recommended frequency is the recommended frequency in a certain time, which may be the recommended frequency in one hour or one day, and may be set according to the actual requirement, and this embodiment is not limited specifically.
In one example, according to the recommendation frequency, recommending the content corresponding to the heuristic recommendation direction includes: and according to the recommendation frequency, inserting the content corresponding to the recommendation direction of the recommendation target user into the content corresponding to the recommendation direction of the recommendation target user. As the example, the heuristic recommendation directions are movies, cosmetics and sports, the affinities are 0.48, 0.42 and 0.18, the recommendation frequencies are 3, 3 and 2 according to the preset corresponding relations, and the recommendation directions of the target users are delicates, so that the content corresponding to 5 delicates can be recommended, the content corresponding to 3 movies can be recommended, the content corresponding to 5 delicates can be recommended, the content corresponding to 3 cosmetics can be recommended, the content corresponding to 5 delicates can be recommended, the content corresponding to 3 individual sports can be recommended, and the content corresponding to delicates can be recommended. In one example, the recommendation may be performed in a sequential recommendation format, or the like, as long as the number of times of recommendation is completed within a certain period of time.
It should be noted that, because the affinity between the target user and the social friend may change, the affinity between the target user and the social friend may be obtained again every other preset period, the obtained affinity may be used as the adjusted affinity, or the obtained affinities may be accumulated or multiplied to obtain the adjusted affinity, which may be set according to the actual requirement, and the embodiment is not limited specifically. In this way, it is possible to further and more accurately implement recommending the content corresponding to the possible interest direction for the user.
In this embodiment, social information includes social friends of a target user, affinity between the target user and the social friends, and recommendation directions corresponding to the social friends, where the recommendation directions are interest directions, and since the affinity between the target user and the social friends can reflect the interest directions of the target user to a certain extent, a heuristic recommendation direction can be selected from the recommendation directions according to the affinity, then recommendation frequency is determined according to the affinity, and content corresponding to the heuristic recommendation directions is recommended according to the recommendation frequency, so that possible content corresponding to the interest directions is actively recommended to the user.
A second embodiment of the present application is directed to a content recommendation method, which is substantially the same as the first embodiment, and is mainly different in that: and the recommended frequency is adjusted according to the feedback behavior of the target user on the content, and corresponding operation is executed according to the adjusted recommended frequency. The specific flow is shown in fig. 2, and includes:
step 201, obtaining social information of a target user; the social information comprises social friends of the target user, affinity between the target user and the social friends, and recommendation directions corresponding to the social friends.
Step 202, selecting a heuristic recommendation direction from the recommendation directions according to the affinity.
Step 203, determining a recommendation frequency according to the intimacy, and recommending the content corresponding to the heuristic recommendation direction according to the recommendation frequency.
Steps 201-203 are similar to steps 101-103 and are not described in detail herein.
Step 204, collecting feedback behaviors of the target user on the content corresponding to the heuristic recommendation direction.
Specifically, the feedback actions of the target user on the content corresponding to the heuristic recommendation direction include, but are not limited to, the following actions: switching, fast forwarding, viewing full films, commenting, sharing, collecting, etc. In one example, the feedback behavior of the target user on each content may be collected in a preset collection period, where the preset collection period may be set according to the actual requirement, for example: 24 hours, the present embodiment is not particularly limited. In one example, the feedback behavior of the target user to the content may be collected randomly, the feedback behavior of the user to one content may be collected randomly, or the feedback behavior of the user to a plurality of contents may be collected randomly.
And step 205, adjusting the recommended frequency according to the feedback behavior to obtain the adjusted recommended frequency.
In one example, the recommended frequency is adjusted according to the feedback behavior, and a flowchart of obtaining the adjusted recommended frequency is shown in fig. 3, including:
step 2051, matching a score value for representing the preference degree of the target user to the content corresponding to the heuristic recommendation direction for the feedback behavior.
Specifically, according to the calculation rule, the score value of the feedback behavior is calculated, and the preference degree of the target user to the content can be reflected by the score value. For example: and switching, fast forwarding, watching the whole piece, commenting, sharing and collecting the corresponding score value to be-2, -1,3,1,2,2, and calculating the score value according to a calculation rule, wherein if the feedback action of a certain content is fast forwarding and watching the whole piece, the sum of the two can be calculated, the score value is 2, or the score value can be taken according to the highest value in the two, and the score value is 3. It should be noted that, if the time period of watching by the target user is not the whole film, the corresponding score value may be calculated according to the time period of watching, for example: if the viewing time is one third of the full-film time, the corresponding score value is 1.
Step 2052, calculating the value of the fractional value sum according to the feedback behavior.
Specifically, the value of the score value sum of each feedback behavior is calculated, and when a plurality of heuristic recommendation directions exist, the value of the score value sum is calculated separately according to the heuristic recommendation directions for the content corresponding to different heuristic recommendation directions. For example: the heuristic recommendation direction is a film and makeup, the score values of the heuristic recommendation direction film are respectively 3, -1 and-1, the total value of the score values is 1, the score values of the heuristic recommendation direction makeup are respectively 3, 2 and-1, and the total value of the score values is 4.
Step 2053, adjusting the recommended frequency according to the numerical value to obtain an adjusted recommended frequency.
In one example, if the value is greater than a first preset threshold, adjusting the recommendation frequency to be the recommendation frequency of the recommendation direction of the target user, and obtaining the adjusted recommendation frequency to be the recommendation frequency of the recommendation direction of the target user; if the value is smaller than a second preset threshold value, the recommended frequency is adjusted to be zero, and the adjusted recommended frequency is obtained to be zero; if the value is not smaller than the second preset threshold value and not larger than the first preset threshold value, updating the recommended frequency, and obtaining the adjusted recommended frequency as the updated recommended frequency. Specifically, the larger the value is, the more liked the user is, the first preset threshold and the second preset threshold are set according to actual needs, and the embodiment is not specifically limited. Adding the tentative recommendation direction into the recommendation direction of the target user to form an interest tag of the user, wherein the adjusted recommendation frequency of the tentative recommendation direction is the recommendation frequency of the recommendation direction of the target user; after deleting the heuristic recommendation direction, the adjusted recommendation frequency is zero, namely, the content corresponding to the heuristic recommendation direction is not recommended any more; otherwise, updating the recommended frequency, wherein the adjusted recommended frequency is the updated recommended frequency.
In one example, updating the recommended frequency, the adjusted recommended frequency being the updated recommended frequency, includes: if the value is not smaller than the third preset threshold value and smaller than the first preset threshold value, the recommended frequency is increased according to the preset rewarding value, and the adjusted recommended frequency is the increased recommended frequency; and if the numerical value is not smaller than the second preset threshold value and smaller than the third preset threshold value, reducing the recommended frequency according to the preset punishment value, and obtaining the adjusted recommended frequency as the reduced recommended frequency. Specifically, the third preset threshold may be set according to actual needs, and the embodiment is not limited specifically. And if the numerical value is not less than the third preset threshold value and is less than the first preset threshold value, adding the preset rewarding value and the original recommended frequency to obtain a sum, wherein the sum is used as the adjusted recommended frequency, if the numerical value is not less than the second preset threshold value and is less than the third preset threshold value, subtracting the original recommended frequency from the preset punishment value to obtain a difference, and if the preset punishment value is negative, adding the original recommended frequency and the preset punishment value to obtain the sum, and taking the sum as the adjusted recommended frequency.
In one example, the recommended frequency is adjusted according to the feedback behavior, and a flowchart of obtaining the adjusted recommended frequency is shown in fig. 4, and includes:
step 2054, matching, for the feedback behavior, a tag for characterizing a preference degree of the target user for the content corresponding to the heuristic recommendation direction, where the tag includes positive feedback for characterizing a higher preference degree and negative feedback for characterizing a lower preference degree.
Specifically, the feedback behavior may indicate the preference degree of the target user for the content, and the preference degree judging standard may be set according to the needs, which is not limited in this embodiment. For example: switching indicates that the user is extremely annoying to the content, fast forward indicates that the user is disliked to the content, extremely annoying and disliked indicates that the user is less liked, and switching and fast forward belong to negative feedback; watching the whole piece indicates that the user likes the content, and commentary, sharing and collection indicates that the user likes the content particularly, and likes and particularly likes the content to indicate that the user has higher favorite degree, and then watching the whole piece, commentary, sharing and collection belongs to positive feedback.
Step 2055, calculating the value of the positive feedback ratio according to the feedback behavior.
Specifically, the value of the positive feedback ratio in each feedback action=the number of positive feedback times/the total number of feedback actions, and when there are a plurality of tentative recommendation directions, the positive feedback ratio is calculated separately according to the tentative recommendation directions for the contents corresponding to different tentative recommendation directions. As described above, if the heuristic recommendation direction is a movie or a cosmetic, the numerical value of the positive feedback ratio of the target user to the content corresponding to the movie or the content corresponding to the cosmetic is calculated.
Step 2056, adjusting the recommended frequency according to the value to obtain an adjusted recommended frequency.
Step 2056 is similar to step 2053 and will not be described again.
And 206, executing corresponding operation according to the adjusted recommended frequency.
Specifically, according to the adjusted recommendation frequency, the content corresponding to the heuristic recommendation direction may be recommended in a manner of interleaving recommendation, or may be recommended in a manner of sequential recommendation.
In this embodiment, since the feedback behavior of the target user may directly reflect the preference degree of the user for the content, the recommendation frequency is adjusted according to the feedback behavior, so that the adjusted recommendation frequency executes a corresponding operation, and the content corresponding to the possible preference direction of the user recommendation can be further and more accurately implemented.
A third embodiment of the application relates to a server, as shown in fig. 5, comprising at least one processor 302; and a memory 301 communicatively coupled to the at least one processor; the memory 301 stores instructions executable by the at least one processor 302, the instructions being executable by the at least one processor 302 to enable the at least one processor 302 to perform embodiments of the content recommendation method described above.
Where the memory 301 and the processor 302 are connected by a bus, the bus may comprise any number of interconnected buses and bridges, the buses connecting the various circuits of the one or more processors 302 and the memory 301 together. The bus may also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver may be one element or may be a plurality of elements, such as a plurality of receivers and transmitters, providing a means for communicating with various other apparatus over a transmission medium. The data processed by the processor 302 is transmitted over a wireless medium via an antenna, which further receives the data and transmits the data to the processor 302.
The processor 302 is responsible for managing the bus and general processing and may also provide various functions including timing, peripheral interfaces, voltage regulation, power management, and other control functions. And memory 301 may be used to store data used by processor 302 in performing operations.
A fourth embodiment of the present application relates to a computer-readable storage medium storing a computer program. The above-described content recommendation method embodiments are implemented when the computer program is executed by a processor.
That is, it will be understood by those skilled in the art that all or part of the steps in implementing the methods of the embodiments described above may be implemented by a program stored in a storage medium, where the program includes several instructions for causing a device (which may be a single-chip microcomputer, a chip or the like) or a processor (processor) to perform all or part of the steps in the methods of the embodiments of the application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
It will be understood by those of ordinary skill in the art that the foregoing embodiments are specific examples of carrying out the application and that various changes in form and details may be made therein without departing from the spirit and scope of the application.
Claims (9)
1. A content recommendation method, comprising:
acquiring social information of a target user; the social information comprises social friends of the target user, affinity between the target user and the social friends, and recommendation directions corresponding to the social friends;
selecting a heuristic recommendation direction from the recommendation directions according to the intimacy;
determining a recommendation frequency according to the intimacy, and recommending the content corresponding to the heuristic recommendation direction according to the recommendation frequency;
collecting feedback behaviors of the target user on the content corresponding to the heuristic recommendation direction;
adjusting the recommended frequency according to the feedback behavior to obtain an adjusted recommended frequency;
and executing corresponding operation according to the adjusted recommended frequency.
2. The content recommendation method according to claim 1, wherein the intimacy is obtained by: and determining the affinity according to the interaction behavior and the social relationship between the target user and the social friends.
3. The content recommendation method according to claim 1, wherein the adjusting the recommendation frequency according to the feedback behavior, to obtain the adjusted recommendation frequency, includes:
matching a score value for representing the favorites of the target user for the content corresponding to the heuristic recommendation direction for the feedback behavior;
calculating the numerical value of the sum of the score values according to the feedback behavior;
and adjusting the recommended frequency according to the numerical value to obtain the adjusted recommended frequency.
4. The content recommendation method according to claim 1, wherein the adjusting the recommendation frequency according to the feedback behavior, to obtain the adjusted recommendation frequency, includes:
matching a tag for representing the preference degree of the target user for the content corresponding to the heuristic recommendation direction with the feedback behavior, wherein the tag comprises positive feedback representing the higher preference degree and negative feedback representing the lower preference degree;
calculating the numerical value of the positive feedback proportion according to the feedback behavior;
and adjusting the recommended frequency according to the numerical value to obtain the adjusted recommended frequency.
5. The content recommendation method according to claim 3 or 4, wherein said adjusting the recommendation frequency according to the value, to obtain an adjusted recommendation frequency, comprises:
if the numerical value is larger than a first preset threshold value, the recommended frequency is adjusted to be the recommended frequency of the recommended direction of the target user, and the adjusted recommended frequency is obtained to be the recommended frequency of the recommended direction of the target user;
if the numerical value is smaller than a second preset threshold value, the recommended frequency is adjusted to be zero, and the adjusted recommended frequency is obtained to be zero;
and if the numerical value is not smaller than the second preset threshold value and not larger than the first preset threshold value, updating the recommended frequency, and obtaining the adjusted recommended frequency as the updated recommended frequency.
6. The content recommendation method according to claim 5, wherein the updating the recommendation frequency to obtain the adjusted recommendation frequency as an updated recommendation frequency comprises:
if the numerical value is not smaller than a third preset threshold value and smaller than the first preset threshold value, the recommended frequency is increased according to a preset rewarding value, and the adjusted recommended frequency is obtained as the increased recommended frequency;
and if the numerical value is not smaller than the second preset threshold value and smaller than the third preset threshold value, reducing the recommended frequency according to a preset punishment value, and obtaining the adjusted recommended frequency as the reduced recommended frequency.
7. The content recommendation method according to claim 1, wherein recommending the content corresponding to the tentative recommendation direction according to the recommendation frequency includes:
and according to the recommendation frequency, inserting and recommending the content corresponding to the heuristic recommendation direction in the content corresponding to the recommendation direction of the target user.
8. A server, comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein,,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the content recommendation method according to any one of claims 1 to 7.
9. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the content recommendation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010298828.5A CN111523050B (en) | 2020-04-16 | 2020-04-16 | Content recommendation method, server and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010298828.5A CN111523050B (en) | 2020-04-16 | 2020-04-16 | Content recommendation method, server and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111523050A CN111523050A (en) | 2020-08-11 |
| CN111523050B true CN111523050B (en) | 2023-09-19 |
Family
ID=71903852
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010298828.5A Active CN111523050B (en) | 2020-04-16 | 2020-04-16 | Content recommendation method, server and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111523050B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993627B (en) * | 2019-02-28 | 2021-09-07 | 华为技术有限公司 | Recommendation method, recommendation model training device and storage medium |
| CN112667897A (en) * | 2020-12-29 | 2021-04-16 | 上海掌门科技有限公司 | Information push method, equipment and storage medium |
| CN116150470A (en) * | 2021-11-19 | 2023-05-23 | 腾讯科技（深圳）有限公司 | Content recommendation method, device, apparatus, storage medium and program product |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130075630A (en) * | 2011-12-27 | 2013-07-05 | 엔에이치엔(주) | Social network service system and method for recommending friend of friend based on intimateness between users |
| CN103823888A (en) * | 2014-03-07 | 2014-05-28 | 安徽融数信息科技有限责任公司 | Node-closeness-based social network site friend recommendation method |
| CN104331459A (en) * | 2014-10-31 | 2015-02-04 | 百度在线网络技术（北京）有限公司 | Online learning-based network resource recommendation method and device |
| JP2017033083A (en) * | 2015-07-29 | 2017-02-09 | 富士フイルム株式会社 | Recommendation device, recommendation method, program and recording medium |
| CN107273538A (en) * | 2017-06-29 | 2017-10-20 | 广州优视网络科技有限公司 | Information recommends method, device and server |
| CN107545451A (en) * | 2016-06-27 | 2018-01-05 | 腾讯科技（深圳）有限公司 | A kind of advertisement sending method and device |
| CN109145221A (en) * | 2018-09-10 | 2019-01-04 | 北京点网聚科技有限公司 | Content recommendation method and device, electronic equipment, readable storage medium storing program for executing |
| CN109684507A (en) * | 2018-11-23 | 2019-04-26 | 北京达佳互联信息技术有限公司 | Video recommendation method, video recommendations device and computer readable storage medium |
| CN109769128A (en) * | 2018-12-25 | 2019-05-17 | 北京达佳互联信息技术有限公司 | Video recommendation method, video recommendations device and computer readable storage medium |
| CN110111152A (en) * | 2019-05-10 | 2019-08-09 | 腾讯科技（深圳）有限公司 | A kind of content recommendation method, device and server |
| CN110837598A (en) * | 2019-11-11 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9661364B2 (en) * | 2011-09-27 | 2017-05-23 | Verizon Patent And Licensing Inc. | Recommended media content based channel surfing methods and systems |
| US8805751B2 (en) * | 2011-10-13 | 2014-08-12 | Verizon Patent And Licensing Inc. | User class based media content recommendation methods and systems |
| US9311663B2 (en) * | 2012-06-22 | 2016-04-12 | Verizon Patent And Licensing Inc. | Media content location methods and systems |
-
2020
- 2020-04-16 CN CN202010298828.5A patent/CN111523050B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130075630A (en) * | 2011-12-27 | 2013-07-05 | 엔에이치엔(주) | Social network service system and method for recommending friend of friend based on intimateness between users |
| CN103823888A (en) * | 2014-03-07 | 2014-05-28 | 安徽融数信息科技有限责任公司 | Node-closeness-based social network site friend recommendation method |
| CN104331459A (en) * | 2014-10-31 | 2015-02-04 | 百度在线网络技术（北京）有限公司 | Online learning-based network resource recommendation method and device |
| JP2017033083A (en) * | 2015-07-29 | 2017-02-09 | 富士フイルム株式会社 | Recommendation device, recommendation method, program and recording medium |
| CN107545451A (en) * | 2016-06-27 | 2018-01-05 | 腾讯科技（深圳）有限公司 | A kind of advertisement sending method and device |
| CN107273538A (en) * | 2017-06-29 | 2017-10-20 | 广州优视网络科技有限公司 | Information recommends method, device and server |
| CN109145221A (en) * | 2018-09-10 | 2019-01-04 | 北京点网聚科技有限公司 | Content recommendation method and device, electronic equipment, readable storage medium storing program for executing |
| CN109684507A (en) * | 2018-11-23 | 2019-04-26 | 北京达佳互联信息技术有限公司 | Video recommendation method, video recommendations device and computer readable storage medium |
| CN109769128A (en) * | 2018-12-25 | 2019-05-17 | 北京达佳互联信息技术有限公司 | Video recommendation method, video recommendations device and computer readable storage medium |
| CN110111152A (en) * | 2019-05-10 | 2019-08-09 | 腾讯科技（深圳）有限公司 | A kind of content recommendation method, device and server |
| CN110837598A (en) * | 2019-11-11 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Li Chao 等.《A Social Network system oriented hybrid recommendation model》.《IEEE》.2012,全文. * |
| 闫婧.《社会化标签系统中基于本体的个性化推荐方法研究》.《无线互联科技》.2015,全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111523050A (en) | 2020-08-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11974019B2 (en) | Identifying related videos based on relatedness of elements tagged in the videos | |
| US11936610B2 (en) | Privacy aligned and personalized social media content sharing recommendations | |
| KR102405115B1 (en) | Using machine learning to recommend live-stream content | |
| CN111523050B (en) | Content recommendation method, server and storage medium | |
| Hawalah et al. | Dynamic user profiles for web personalisation | |
| KR102233805B1 (en) | Improved user experience for unrecognized and new users | |
| US11805165B2 (en) | Computerized system and method for an interactive online learning experience | |
| CN105574045B (en) | Video recommendation method and server | |
| CN110223186B (en) | User similarity determining method and information recommending method | |
| EP3709193A2 (en) | Media content discovery and character organization techniques | |
| US20170251263A1 (en) | Rating videos based on parental feedback | |
| US9864951B1 (en) | Randomized latent feature learning | |
| WO2008051338A2 (en) | Collaborative filtering based on attribute profiling | |
| US10206003B1 (en) | Identifying media channels that have a high likelihood of multiple consumptions by one or more users | |
| CN102165441A (en) | Method, system, and apparatus for ranking media sharing channels | |
| Barragáns-Martínez et al. | Developing a recommender system in a consumer electronic device | |
| US20140089238A1 (en) | Information processing device and information processing method | |
| Cao et al. | Identifying core users based on trust relationships and interest similarity in recommender system | |
| KR102661495B1 (en) | Dynamic integration of customized complementary media content | |
| US9015607B2 (en) | Virtual space providing apparatus and method | |
| JP2014222474A (en) | Information processor, method and program | |
| CN115481262A (en) | Cross-domain migration recommendation system based on user interest preference | |
| JP7387970B2 (en) | Calculation device, calculation method, calculation program | |
| Fernández et al. | Let´ s go to the cinema! A movie recommender system for ephemeral groups of users | |
| CN116910347A (en) | Media asset recommendation method and device and electronic equipment |
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 |