CN112836127A - Method and device for recommending social users, storage medium and electronic equipment - Google Patents

Method and device for recommending social users, storage medium and electronic equipment Download PDF

Info

Publication number
CN112836127A
CN112836127A CN202110174284.6A CN202110174284A CN112836127A CN 112836127 A CN112836127 A CN 112836127A CN 202110174284 A CN202110174284 A CN 202110174284A CN 112836127 A CN112836127 A CN 112836127A
Authority
CN
China
Prior art keywords
user
list
recommended
interaction
users
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110174284.6A
Other languages
Chinese (zh)
Other versions
CN112836127B (en
Inventor
郑礼雄
任彦
曹华平
薛晨
易立
陆希玉
王云荣
窦禹
王一宇
杨昕雨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National Computer Network and Information Security Management Center
Original Assignee
National Computer Network and Information Security Management Center
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by National Computer Network and Information Security Management Center filed Critical National Computer Network and Information Security Management Center
Priority to CN202110174284.6A priority Critical patent/CN112836127B/en
Publication of CN112836127A publication Critical patent/CN112836127A/en
Application granted granted Critical
Publication of CN112836127B publication Critical patent/CN112836127B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—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
    • G06F16/9535—Search customisation based on user profiles and personalisation
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • Health & Medical Sciences (AREA)
  • Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Marketing (AREA)
  • Primary Health Care (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the application discloses a method for recommending social users, which comprises the following steps: removing the hotspot users in the first attention list of the target user to obtain a second attention list; based on the second concern list, obtaining a list of users to be recommended in a user database of the social platform; calculating the recommendation weight of each user to be recommended in the user list to be recommended based on the user list to be recommended; and acquiring a recommended user from the at least one user to be recommended based on the recommendation weight and a preset rule, and recommending the recommended user to the target user. By adopting the embodiment of the application, personalized recommendation for the target user can be realized, and the recommendation effect of the social network user recommendation is obviously improved.

Description

Method and device for recommending social users, storage medium and electronic equipment
Technical Field
The application relates to the technical field of internet data mining, in particular to a method, a device, a storage medium and electronic equipment for recommending social users.
Background
With the advent of the "internet +" era, the internet ecosystem is gradually formed, and netizens, as the most active and important component part in the ecosystem, influence the ecosphere by one action. At present, the demand of netizens on the internet is not only for looking up data and browsing information, but also for more and more netizens to do shopping, social contact, entertainment and other activities through the internet. Online social contact becomes a part of internet life of net citizens, social platforms such as microblog and tremble are popular, relatively open information real-time sharing places are provided for the net citizens, meanwhile, a user attention mechanism of the social platforms provides a friend making way for the net citizens, and information sharing and social contact requirements of the net citizens are met. However, in the prior art, when a user wants to obtain more users with the same interests and hobbies as the user on an open social platform so as to perform friendly social contact, the user often only sees the user who is not interested because the recommendation algorithm of the social platform is behind or the recommendation algorithm is not suitable for the social platform.
Disclosure of Invention
The embodiment of the application provides a method, a device, a storage medium and electronic equipment for recommending social users, which can realize personalized recommendation for target users and remarkably improve the recommendation effect of network social user recommendation. The technical scheme is as follows:
in a first aspect, an embodiment of the present application provides a method for recommending social users, including:
removing the hotspot users in the first attention list of the target user to obtain a second attention list; the hotspot users are users with the number of interactions exceeding an interaction threshold value and/or the number of concerned users exceeding an attention threshold value in the social platform;
based on the second concern list, obtaining a list of users to be recommended in a user database of the social platform; the user list to be recommended comprises at least one user to be recommended, and the attention list of the user to be recommended comprises at least one user in the second attention list;
calculating the recommendation weight of each user to be recommended in the user list to be recommended based on the user list to be recommended;
and acquiring a recommended user from the at least one user to be recommended based on the recommendation weight and a preset rule, and recommending the recommended user to the target user.
In one or more possible embodiments, before removing the hot spot user in the first interest list of the target user and obtaining the second interest list, the method further includes:
acquiring a recessive concern list and an explicit concern list of a target user based on a preset algorithm;
the first concern list is a union of the explicit concern list and the implicit concern list, the explicit concern list is a concern list of the target user displayed on the social platform, and the implicit concern list is a user set which is concerned by the target user in an indirect concern manner on the social platform.
In one or more possible embodiments, the obtaining an implicit interest list and an explicit interest list of the target user based on a preset algorithm includes:
acquiring a published text of the target user within a preset time period;
extracting user pronouns in the published text; wherein the user pronouns include at least one of: user name, user nickname, topic;
and determining a hidden user corresponding to the user appointed word based on the user appointed word, and constructing the hidden concern list based on the hidden user.
In one or more possible embodiments, the obtaining an implicit interest list and an explicit interest list of the target user based on a preset algorithm includes:
judging the activity of the target user;
determining that the target user is an inactive user when the activity of the target user is less than an activity threshold;
constructing a social network relationship network of the target user based on the mutual attention list of the target user;
acquiring active users in the social network relationship network based on the social network relationship network;
acquiring similar users based on the active users; the similar users are users of which the similarity between the active users and the target users exceeds a similarity threshold;
and acquiring the implicit concern list of the similar user based on the similar user, and taking the implicit concern list of the similar user as the implicit concern list of the target user.
In one or more possible embodiments, the calculating, based on the list of users to be recommended, a recommendation weight of each user to be recommended in the list of users to be recommended includes:
acquiring an intersection list of the concern list of the user to be recommended and the second concern list of the target user;
calculating the interaction weight of the user to be recommended and each user in the intersection list within a preset time length;
and accumulating the interaction weights to obtain the recommendation weight of the user to be recommended.
In one or more possible embodiments, the calculating the interaction weight between the user to be recommended and each user in the intersection list within the preset time includes:
based on a preset time granularity, configuring an attenuation weight for the initial interaction weight in the preset time length to obtain an interaction attenuation function;
acquiring interaction data of the user to be recommended and each user in the intersection list, wherein the interaction data comprises interaction types, interaction frequencies and interaction timeliness;
and acquiring the interaction weight based on the interaction attenuation function and the interaction data.
In one or more possible embodiments, the calculating the interaction weight between the user to be recommended and each user in the intersection list within the preset time includes:
calculating an interaction aging factor TF (t) of the user to be recommended and each user in the intersection list, wherein:
Figure RE-GDA0003024291630000031
wherein T represents the time of the kth interaction after the preset time length is divided based on the first time granularity, and T is1Representing an interaction end time based on a second time granularity, the T0Representing an interaction start time based on a second time granularity;
acquiring an interaction type of the interaction between the user to be recommended and each user in the intersection list and an interaction type weight V (i, j) corresponding to the interaction type, wherein the value of V (i, j) is an integer between 1 and 5;
acquiring the kth interaction weight W of the user to be recommended and each user in the intersection list within the preset time length based on the interaction aging factor TF (t) and the interaction type weight V (i, j)k(i, j, t), wherein:
Wk(i，j，t)＝V(i，j，k)×TF(t)；
based on the interaction weight Wk(i, j, t) calculating the recommendation weight W (i, j), wherein:
Figure RE-GDA0003024291630000041
and K represents the total number of times of interaction between the recommended user and each user in the intersection list within the preset time length.
In a second aspect, an embodiment of the present application provides an apparatus for recommending social users, where the apparatus includes:
the hot spot removing module is used for removing the hot spot users in the first attention list of the target user to obtain a second attention list; the hotspot users are users with the number of interactions exceeding an interaction threshold value and/or the number of concerned users exceeding an attention threshold value in the social platform;
the list obtaining module is used for obtaining a list of users to be recommended from a user database of the social platform based on the second concern list; the user list to be recommended comprises at least one user to be recommended, and the attention list of the user to be recommended comprises at least one user in the second attention list;
the calculation weight module is used for calculating the recommendation weight of each user to be recommended in the user list to be recommended based on the user list to be recommended;
and the recommending user module is used for acquiring a recommending user from the at least one user to be recommended based on the recommending weight and a preset rule and recommending the recommending user to the target user.
In a third aspect, embodiments of the present application provide a computer storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the above-mentioned method steps.
In a fourth aspect, an embodiment of the present application provides an electronic device, which may include: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the above-mentioned method steps.
The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise: the method comprises the steps of obtaining a list of users to be recommended by removing a second concern list of hot users, calculating the recommendation weight of each user to be recommended in the list of users to be recommended, determining a recommended user based on the recommendation weight and a preset rule, and recommending the recommended user to a target user, so that personalized recommendation for the target user can be realized, the recommendation effect of network social user recommendation is remarkably improved, the probability of recommending interested recommended users to the target user is improved, and the social friend network of the target user is further mined.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a schematic flowchart illustrating a method for recommending social users according to an embodiment of the present disclosure;
FIG. 2 is an interface diagram of an explicit focus list of a target user according to an embodiment of the present disclosure;
fig. 3 is a schematic flowchart illustrating a process of establishing an implicit concern list of a target user according to an embodiment of the present application;
FIG. 4 is a schematic flowchart of calculating recommendation weights according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of an intersection of interest lists of a target user and a user to be recommended according to an embodiment of the present application;
fig. 6 is a schematic flowchart of computing interaction weights according to an embodiment of the present disclosure;
FIG. 7 is a schematic flow chart of another method for recommending social users according to an embodiment of the present disclosure;
FIG. 8 is a schematic structural diagram of a social networking network of a target user according to an embodiment of the present disclosure;
FIG. 9 is a schematic flow chart of another calculation of recommendation weight according to an embodiment of the present application;
fig. 10 is a schematic diagram of an apparatus for recommending social friends according to an embodiment of the present disclosure;
fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the description of the present application, it is to be understood that the terms "first," "second," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. In the description of the present application, it is noted that, unless explicitly stated or limited otherwise, "including" and "having" and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus. The specific meaning of the above terms in the present application can be understood in a specific case by those of ordinary skill in the art. Further, in the description of the present application, "a plurality" means two or more unless otherwise specified. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship.
The present application will be described in detail with reference to specific examples.
In one embodiment, as shown in fig. 1, a method of recommending social users is proposed, which may be implemented by means of a computer program, and which may be run on an image source tracking device based on von neumann architecture. The computer program may be integrated into the application or may run as a separate tool-like application.
Specifically, the method for recommending social users comprises the following steps:
s101, removing the hot spot users in the first attention list of the target user to obtain a second attention list.
The target user can be understood as any user account on the social platform, and the user account can be an identifier with unique directivity, such as a device number, a mobile phone number, a user name and the like, for example, an account of the last id of the microblog, called zhang san. The social platform may be understood as a service platform having social attribute functions such as a social function, a friend function, and a text publishing function in the prior art, for example, micro blog, trembling, or soul.
The hotspot user may be understood as an account number in which the number of interactions in the social platform exceeds an interaction threshold and/or the number of focused users exceeds an attention threshold, for example, the social platform is a microblog, the number of interactions is a sum of numbers including praise, forward, and comment, the interaction threshold is 10000 for a single microblog text, the number of focused users is a sum of the number of people focusing on the user, the attention threshold is 100 ten thousands for the user, and the hotspot user may be understood as a user account number satisfying one or more of the above conditions, for example, a user account number representing celebrity.
In one embodiment, removing the hot spot users in the first interest list of the target user and obtaining the second interest list further includes: acquiring a recessive concern list and an explicit concern list of a target user based on a preset algorithm; the first concern list is a union set of an explicit concern list and a implicit concern list, the explicit concern list is a concern list of a target user displayed on the social platform, and the implicit concern list is a user set which is concerned by the target user in an indirect concern mode on the social platform.
As shown in fig. 2, an interface schematic diagram of an explicit concern list of a target user provided by the embodiment of the present application includes zhang san 201 of the target user, where the explicit concern list is all user identifiers included in a concern list 202. In other words, when the terminal device detects a trigger condition on the [ focus ] control, the focus list 202 is displayed on the display interface, and the focus list 202 is an explicit focus list.
It should be noted that when a target user pays attention to another user, it may be understood that a node representing the target user and a node representing the another user establish a unidirectional link, and the link direction is from the node of the target user to the node of the another user, and the attention mode is defined as direct attention; when the target user mentions that the frequency of a certain user-indicated pronoun in the published text exceeds an indirect attention threshold, but the target user does not establish a unidirectional link or a bidirectional link with the user-indicated pronoun, the attention mode is defined as indirect attention. The user-designated pronouns include at least one of: user name, user nickname, topic, e.g., user name: xxx, user nickname: JayChou, topic: you are my you-le #, which are the user's pronouns for the people of the entity in week xx and the account JayChou in week xx on the social platform. In other words, indirect attention may be understood that the number of times that the target user zhang refers to the user's indicator representing the entity character week xx in the publication text exceeds the indirect attention threshold, but does not directly pay attention to the account number of week xx on the social platform, and the publication text includes text data left by all target users on the social platform, such as private letters of the social users on the social platform, the comment content published in the own account, and the comment content of the comment area under the comment content of other social users.
As shown in fig. 3, a schematic flow chart for acquiring an implicit attention list of a target user according to an embodiment of the present application includes:
s301, obtaining published texts of a target user in a preset time period;
for example, the preset time period is a publication text of the target user from No. 1/2020 to No. 12/31/2020, the publication text includes text data left by all target users on the social platform, such as private letters of the social users on the social platform, comment content published in the own account, comment content of a comment area under the comment content of other social users, and the like, for example, the comment content of the target user on a microblog is: you are my you happy # drink today with the same strawberry flavor in my week.
S302, extracting user pronouns in the published text;
the user-designated pronouns include at least one of: user name, user nickname, topic. For example, the content of the speech of the target user in the microblog is as follows: and # you are my week same strawberry flavor as my you do # drinking today, wherein the topic # you are user referees for my you do # and the user nickname "my week" both being the entity person week xx and the account number JayChou of the week xx on the social platform.
The extraction algorithm in the step can adopt a natural language processing method, firstly, the part of speech segmentation and labeling are carried out on the content of the issued text, and the corresponding entity characters are matched in a preset natural language word stock based on the labeling.
S303, determining a hidden user corresponding to the user appointed pronouns based on the user appointed pronouns, and constructing a hidden concern list based on the hidden user.
The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise: the hidden user list is constructed by extracting the hidden user from the published text of the user, and the first concern list of the target user is formed by merging the hidden user list and the explicit user list, so that the degree of mining the interest and hobbies of the target user is improved, the establishment of the concern list of the target user is perfected, and the deviation and omission caused by the follow-up matching of recommended users for the target user are avoided.
S102, acquiring a list of users to be recommended from a user database of the social platform based on the second concern list.
It should be noted that the to-be-recommended user list includes at least one to-be-recommended user, and the attention list of the to-be-recommended user includes at least one user in the second attention list.
For example, the second interest list of the third target user includes a user a, a user B, a user C, and a user D, the server of the social platform matches at least one to-be-recommended user in the user database through the second interest list, the interest list of the to-be-recommended user at least includes one or more of the user a, the user B, the user C, and the user D, for example, matches to a to-be-recommended user 1 and a to-be-recommended user 2, the interest list of the to-be-recommended user 1 includes the user a, the interest list of the to-be-recommended user 2 includes the user B and the user D, and the to-be-recommended user list is constructed based on the at least one to-be-.
It can be understood that the attention list of the user to be recommended is acquired based on the above-mentioned acquisition process of the second attention list of the target user, in other words, the attention list of the user to be recommended is also composed of a union of the explicit attention list and the implicit attention list of the user to be recommended.
S103, based on the list of the users to be recommended, calculating the recommendation weight of each user to be recommended in the list of the users to be recommended.
As shown in fig. 4, a schematic flow chart for calculating a recommendation weight provided in the embodiment of the present application includes the specific steps of:
s1031, acquiring an intersection list of the attention list of the user to be recommended and the second attention list of the target user;
as shown in fig. 6, an intersection schematic diagram of attention lists of a target user and a user to be recommended provided in an embodiment of the present application is shown, where a second attention list of a third target user includes a user a, a user B, a user C, and a user D, a server of a social platform matches a user 1 to be recommended and a user 2 to be recommended in a user database through the second attention list, the attention list of the user 1 to be recommended includes the user a, the attention list of the user 2 to be recommended includes the user B and the user D, and then the user list to be recommended includes the user 1 and the user 2; and acquiring an intersection list of the attention list of the user 1 to be recommended and the second attention list of the target user, wherein the intersection list comprises a user A, the intersection list of the attention list of the user 2 to be recommended and the second attention list of the target user, and the intersection list comprises a user B and a user D.
S1032, calculating the interaction weight of the user to be recommended and each user in the intersection list within a preset time length;
for example, as shown in fig. 5, the interaction weight between the user to be recommended and each user in the intersection list in the preset time period is calculated, that is, the interaction weight between the user 1 to be recommended and the user a, and the interaction weight between the user 2 to be recommended and the user B and the user D, respectively, in the preset time period are calculated.
As shown in fig. 6, a schematic flowchart of calculating an interaction weight provided in the embodiment of the present application includes the specific steps of:
s10321, configuring attenuation weights for the initial interaction weights within a preset time length based on a preset time granularity to obtain an interaction attenuation function;
the preset duration may be set based on a recommendation period, for example, if the social platform recommends a social user for the target user every other week, the preset duration may be one week, the recommendation weight of at least one to-be-recommended user in each week is calculated, a recommended user of the at least one to-be-recommended user is obtained based on the recommendation weight and a preset rule, and the recommended user is recommended to the target user.
Configuring attenuation weights for initial interaction weights within a preset duration based on a preset time granularity, wherein,
N(t)＝N0e-at；
n (t) is an interactive attenuation function in t unit times after being divided based on preset time granularity in preset time duration, N0Is the initial interaction weight, and a is the decay weight.
For example, if the time granularity is days and the predetermined duration is one week, then N (6) is the interaction decay function for the sixth day.
S10322, acquiring interaction data of the user to be recommended and each user in the intersection list;
the interaction data comprises an interaction type, an interaction frequency and an interaction time limit, for example, the interaction type comprises a message left by a user to be recommended to a user in an intersection list, the message left by the user to be recommended to the user in the intersection list is mutually complied with the message left by the user in the intersection list, and the like, the interaction frequency is the frequency of sending interaction between the user to be recommended and the user in the intersection list in unit time after the preset time length is divided based on time granularity, and the interaction time limit is the time from the beginning to the end of one interaction.
And S10323, acquiring interaction weight based on the interaction attenuation function and the interaction data.
For example, the time granularity is day, and the preset time duration is one week; on the first day, the server obtains an initial interaction weight 30 based on interaction data of a user to be recommended on the first day and the user in the intersection list, and obtains an interaction weight of 30 on the first day based on an interaction attenuation function; on the next day, the server obtains initial interaction weight 25 based on the interaction data of the user to be recommended on the next day and the user in the intersection list, and obtains the interaction weight of 18 on the next day based on an interaction attenuation function; on the third day, the server obtains an initial interaction weight 32 based on interaction data of a user to be recommended on the third day and a user in the intersection list, obtains an interaction weight 19 based on an interaction decay function on the third day, and so on, within a preset time length of one week, the interaction weights of the user to be recommended and the user in the intersection list are respectively 30, 18, 19, 15, 14, 17 and 25, and then the interaction weight of the user to be recommended and the user in the intersection list is finally 138.
In summary, based on the steps S10321, S10322, and S10323, the interaction weight between the user to be recommended and each user in the intersection list can be obtained. The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise: the initial interaction weight is obtained based on the interaction mode diversity included in the interaction data, different intimacy relationships between users displayed by different interaction modes on the social platform are effectively utilized, and the accuracy of recommending the social users to the target user is improved; and the attenuation weight is configured based on time, so that the calculation of the interaction weight and even the recommendation weight is more reasonable and reliable.
It is to be understood that the present application is not limited to the method for obtaining the interaction weight.
And S1033, accumulating the interaction weights to obtain the recommendation weight of the user to be recommended.
For example, as shown in fig. 5, if the interaction weight of the user 1 to be recommended and the user a in the intersection list is 138, the recommendation weight of the user 1 to be recommended is 138; the interaction weight between the user 2 to be recommended and the user B in the intersection list is 50, the interaction weight between the user D and the user 2 to be recommended is 120, and the recommendation weight of the user 2 to be recommended is 170.
In summary, the recommendation weight of the user to be recommended is calculated based on steps S1031, S1032, and S1033.
S104, acquiring a recommended user from at least one user to be recommended based on the recommendation weight and a preset rule, and recommending the recommended user to a target user.
The preset rule can be that according to the numerical value of the recommendation weight, if the numerical value is larger, the recommendation priority of the user to be recommended is higher, and the recommendation user of the user to be recommended is determined based on the recommendation denomination of the social platform. For example, if the recommendation weight of the user 1 to be recommended is 138, the recommendation weight of the user 2 to be recommended is 170, and the recommendation name of the social platform is 1, recommending the user 2 to be recommended as the recommending user to the target user.
In another embodiment, the preset rule is that when the recommendation weight of the user to be recommended is greater than a recommendation threshold, the user to be recommended is determined to be the recommending user. For example, if the recommendation threshold is 90, the recommendation weight of the user 1 to be recommended is 138, and the recommendation weight of the user 2 to be recommended is 170, the user 1 to be recommended and the user 2 to be recommended are recommended to the target user as recommendation users.
The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise: the method comprises the steps of obtaining a list of users to be recommended by removing a second concern list of hot users, calculating the recommendation weight of each user to be recommended in the list of users to be recommended, determining a recommended user based on the recommendation weight and a preset rule, and recommending the recommended user to a target user, so that personalized recommendation for the target user can be realized, the recommendation effect of network social user recommendation is remarkably improved, the probability of recommending interested recommended users to the target user is improved, and the social friend network of the target user is further mined.
As shown in fig. 7, a flowchart of another method for recommending social users provided in the embodiment of the present application includes the specific steps of:
and S701, judging the activity of the target user.
In one embodiment, the method for determining the liveness of the target user may be: and judging the frequency of the target user interacting with other social users on the social platform, such as 30 times per week, including giving other social users private letters, leaving messages or praise.
In another embodiment, the method for determining the liveness of the target user may be to calculate a ratio of the number of interest of the target user to the number of interest of the target user, where the number of interest of the target user generally reflects the participation of the target user in the social platform, and the number of interest of the target user generally reflects the authority degree and the favorite degree of the target user, when the number of interest/number of interest is greater than 1, the target user is preliminarily determined to be an active user, and when the number of interest/number of interest is less than 1, the target user is preliminarily determined to be a hotspot character, and it is necessary to further determine whether the target user is an active user. The above further judgment may adopt the judgment method in the first embodiment.
S702, under the condition that the activity of the target user is smaller than the activity threshold value, determining that the target user is an inactive user.
And judging whether the target user is an inactive user or not based on the method in the embodiment. For example, if the number of interactions with other social users per week is defined as 35 times as the activity threshold, and the interaction frequency of the target user is 30/week, the target user is determined to be an inactive user.
In another embodiment, when the target user is determined to be an active user, social users are recommended for the target user according to the method shown in fig. 1 to 6.
S703, constructing a social network relationship network of the target user based on the mutual attention list of the target user;
the mutual interest list can be understood as a set of users who establish bidirectional interest with the target user on the basis of the social platform. In order to describe the social relationship of the target user conveniently, the social network relationship of the target user is mined and calculated, and a social network relationship network of the target user is constructed.
Fig. 8 is a schematic structural diagram of a social network relationship network of a target user according to an embodiment of the present disclosure. When the target user focuses on another user, the target user can understand that a node representing the target user establishes a unidirectional link with a node representing the other user, and the link direction is pointed to the node of the other user by the node of the target user. When a target user is interested in another user, it can be understood that the node representing the target user establishes a bidirectional link with the node representing the other user.
S704, acquiring active users in the social network relation network based on the social network relation network.
The method for determining the active user is based on the method described in step S704, in other words, the active user of the social network relationship network of the target user is obtained.
S705, acquiring similar users based on the active users.
Similar users may be understood as users whose similarity between the active user and the target user exceeds a similarity threshold. The judgment algorithm of the similar users can be obtained by Friend of Friend algorithm (FOFA), and the algorithm mainly judges the similarity based on the number of common friends between the target user and the active user. The judgment algorithm of the similar users may also be an Adamic/Adar Algorithm (AA), which represents the similarity between nodes, i.e. the similarity between the target user and the active user, mainly based on calculating the weighted sum of common attributes between the nodes.
S706, acquiring the implicit attention list of the similar user based on the similar user, and taking the implicit attention list of the similar user as the implicit attention list of the target user.
The implicit concern list is a user set which aims at the social platform and is concerned by similar users in an indirect concern mode. When the similar users mention that the frequency of a certain user-indicated pronoun in the published text exceeds an indirect attention threshold, but the similar users do not establish a unidirectional link or a bidirectional link with the user-indicated pronoun, the attention mode is defined as indirect attention.
The method for obtaining the invisible attention list of similar users refers to the steps S301, S302 and S303 shown in fig. 3, and will not be described herein again.
The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise: when the target user is judged to be the inactive user, the implicit attention list of the similar user is used for replacing the implicit attention list of the target user, the situation that the implicit user cannot be extracted due to too few published texts of the target user and the implicit attention list of the target user cannot be built is avoided, and the success rate and the effect of recommending the social user to the target user are improved.
S707, acquiring an explicit concern list of the target user, and taking a union of the explicit concern list and the implicit concern list as a first concern list.
As shown in fig. 2, an interface schematic diagram of an explicit concern list of a target user provided by the embodiment of the present application includes zhang san 201 of the target user, where the explicit concern list is all user identifiers included in a concern list 202. In other words, when the terminal device detects a trigger condition on the [ focus ] control, the focus list 202 is displayed on the display interface, and the focus list 202 is an explicit focus list.
The method comprises the steps that a server of a social platform obtains an explicit concern list of a target user, obtains an implicit concern list through similar users, and takes the explicit concern list and the implicit concern list as a first concern list.
S708, removing the hotspot users in the first attention list of the target user to obtain a second attention list.
The hotspot users are users with the number of interactions exceeding an interaction threshold and/or the number of concerned users exceeding a concerned degree threshold in the social platform. The detailed content of step 708 refers to step S101 shown in fig. 1, and is not described herein again.
And S709, acquiring a list of users to be recommended from a user database of the social platform based on the second concern list.
The user list to be recommended comprises at least one user to be recommended, and the attention list of the user to be recommended comprises at least one user in the second attention list. The specific content of step 709 refers to step S102 shown in fig. 1, and is not described herein again.
S710, based on the to-be-recommended user list, calculating the recommendation weight of each to-be-recommended user in the to-be-recommended user list.
As shown in fig. 9, another schematic flow chart for obtaining recommendation weight provided in the embodiment of the present application includes the specific steps of:
s7101, calculating an interactive aging factor TF (t) of each user in the intersection list and the user to be recommended, wherein:
Figure RE-GDA0003024291630000131
wherein T represents the time of the kth interaction after the preset time duration is divided based on the first time granularity, and T1Indicating an interaction end time, T, based on a second time granularity0Indicating an interaction start time based on the second time granularity.
For example, the preset time duration is one week, the first time granularity is day, the time T of the kth interaction is day 3, the second time granularity is 24 hours, and the interaction end time T1At 8 am, the start time of the interaction T0At 7 am.
S7102, acquiring the interaction type of each user to be recommended and each user in the intersection list, and acquiring the interaction type weight V (i, j) corresponding to the interaction type.
V (i, j) is an integer with a value of 1 to 5, and a larger value of V (i, j) indicates that the social affinity between the target user and each user in the intersection list is higher, for example, the weight corresponding to the interaction type that the target user and the user in the intersection list comment each other is 5, and the weight corresponding to the interaction type that the target user approves the user in the intersection list is 5.
S7103, acquiring the kth interaction weight W of each user in the intersection list and the user to be recommended within a preset time length based on the interaction aging factor TF (t) and the interaction type weight V (i, j)k(i,j,t) Wherein:
Wk(i，j，t)＝V(i，j，k)×TF(t)
s7104 based on interaction weight Wk(i, j, t) calculating a recommendation weight W (i, j), wherein:
Figure RE-GDA0003024291630000141
wherein K represents the total number of interactions performed between the recommended user and each user in the intersection list within a preset time length, for example, within one week of the preset time length, the total number of interactions K is 40.
The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise: the algorithm provided by the embodiment is simpler, the computing resources of the server are saved, the weight corresponding to the interaction type is improved, namely, the intimacy between the user to be recommended and the user in the intersection list is concerned, the indirect intimacy degree between the target user and the recommended user is improved, and the direct intimacy between the target user and the recommended user is better predicted.
And S711, acquiring a recommended user from at least one user to be recommended based on the recommendation weight and a preset rule, and recommending the recommended user to the target user.
The preset rule can be that according to the numerical value of the recommendation weight, if the numerical value is larger, the recommendation priority of the user to be recommended is higher, and the recommendation user of the user to be recommended is determined based on the recommendation denomination of the social platform. For example, if the recommendation weight of the user 1 to be recommended is 138, the recommendation weight of the user 2 to be recommended is 170, and the recommendation name of the social platform is 1, recommending the user 2 to be recommended as the recommending user to the target user.
The details of step S711 refer to step S104 shown in fig. 1, and are not described herein again.
The beneficial effects brought by the technical scheme provided by some embodiments of the application at least comprise: when the target user is judged to be an inactive user, the implicit attention list of the similar user is used for replacing the implicit attention list of the target user, so that the situations that the implicit user cannot be extracted due to too few published texts of the target user and the implicit attention list of the target user cannot be constructed are avoided; the method comprises the steps of obtaining a list of users to be recommended by removing a second concern list of hot users, calculating the recommendation weight of each user to be recommended in the list of users to be recommended, determining a recommended user based on the recommendation weight and a preset rule, and recommending the recommended user to a target user, so that personalized recommendation for the target user can be realized, the recommendation effect of network social user recommendation is remarkably improved, the probability of recommending interested recommended users to the target user is improved, and the social friend network of the target user is further mined.
The following are embodiments of the apparatus of the present application that may be used to perform embodiments of the method of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method of the present application.
Referring to fig. 10, a schematic structural diagram of a recommended social user device according to an exemplary embodiment of the present application is shown. The recommended social user device may be implemented as all or part of a device in software, hardware, or a combination of both. The recommended social user device includes a remove hotspot module 1001, an obtain list module 1002, a calculate weight module 1003, and a recommend user module 1004.
A hot spot removing module 1001 configured to remove a hot spot user in the first attention list of the target user to obtain a second attention list; the hotspot users are users with the number of interactions exceeding an interaction threshold value and/or the number of concerned users exceeding an attention threshold value in the social platform;
the list obtaining module 1002 is configured to obtain a list of users to be recommended in a user database of the social platform based on the second interest list; the user list to be recommended comprises at least one user to be recommended, and the attention list of the user to be recommended comprises at least one user in the second attention list;
the weight calculation module 1003 calculates a recommendation weight of each user to be recommended in the user list to be recommended based on the user list to be recommended;
the recommending user module 1004 is used for acquiring a recommending user of the at least one user to be recommended based on the recommending weight and a preset rule, and recommending the recommending user to the target user.
Optionally, the apparatus for recommending social users further includes:
the acquisition module acquires a recessive concern list and an explicit concern list of a target user based on a preset algorithm;
the first concern list is a union of the explicit concern list and the implicit concern list, the explicit concern list is a concern list of the target user displayed on the social platform, and the implicit concern list is a user set which is concerned by the target user in an indirect concern manner on the social platform.
Optionally, the obtaining module includes:
the preset unit is used for acquiring published texts of the target user within a preset time period;
the extraction unit is used for extracting the user representative words in the published text; wherein the user pronouns include at least one of: user name, user nickname, topic;
the determining unit is used for determining the implicit user corresponding to the user representative word based on the user representative word and constructing the implicit concern list based on the implicit user.
Optionally, the obtaining module includes:
the judging unit is used for judging the activity of the target user;
an inactivity unit to determine that the target user is an inactive user if the target user's activity is less than an activity threshold;
the construction unit is used for constructing a social network relationship network of the target user based on the mutual attention list of the target user;
the active user obtaining unit obtains active users in the social network relationship network based on the social network relationship network;
acquiring a similar user unit, and acquiring a similar user based on the active user; the similar users are users of which the similarity between the active users and the target users exceeds a similarity threshold;
and the replacing unit is used for acquiring the implicit attention list of the similar user based on the similar user and taking the implicit attention list of the similar user as the implicit attention list of the target user.
Optionally, the weight calculating module 1003 includes:
an intersection unit is obtained, and an intersection list of the concern list of the user to be recommended and the second concern list of the target user is obtained;
the interaction weight unit is used for calculating the interaction weight between the user to be recommended and each user in the intersection list within a preset time length;
and the accumulation unit is used for accumulating the interaction weight to obtain the recommendation weight of the user to be recommended.
Optionally, the interaction weight unit is specifically configured to:
based on a preset time granularity, configuring an attenuation weight for the initial interaction weight in the preset time length to obtain an interaction attenuation function;
acquiring interaction data of the user to be recommended and each user in the intersection list, wherein the interaction data comprises interaction types, interaction frequencies and interaction timeliness;
and acquiring the interaction weight based on the interaction attenuation function and the interaction data.
Optionally, the weight calculating module 1003 includes:
and the time efficiency calculating unit is used for calculating the interactive time efficiency factor TF (t) of the user to be recommended and each user in the intersection list, wherein:
Figure RE-GDA0003024291630000171
wherein T represents the time of the kth interaction after the preset time length is divided based on the first time granularity, and T is1Represents a radicalAt the interaction end time of the second time granularity, the T0Representing an interaction start time based on a second time granularity;
an interaction type obtaining unit, which obtains an interaction type of the user to be recommended interacting with each user in the intersection list, and an interaction type weight V (i, j) corresponding to the interaction type, wherein the value of V (i, j) is an integer between 1 and 5;
an interaction weight obtaining unit, configured to obtain, based on the interaction aging factor tf (t) and the interaction type weight V (i, j), a kth interaction weight W of the user to be recommended and each user in the intersection list within the preset time periodk(i, j, t), wherein:
Wk(i，j，t)＝V(i，j，k)×TF(t)；
a unit for calculating recommendation weight based on the interaction weight Wk(i, j, t) calculating the recommendation weight W (i, j), wherein:
Figure RE-GDA0003024291630000172
and K represents the total number of times of interaction between the recommended user and each user in the intersection list within the preset time length.
It should be noted that, when the method for recommending social users is executed by the apparatus for recommending social users provided in the foregoing embodiment, only the division of the functional modules is used as an example, in practical applications, the functions may be distributed to different functional modules according to needs, that is, the internal structure of the apparatus is divided into different functional modules, so as to complete all or part of the functions described above. In addition, the social user recommending device and the social user recommending method provided by the above embodiments belong to the same concept, and the details of the implementation process are referred to in the method embodiments, and are not described herein again.
According to the method and the device, the user list to be recommended is obtained by removing the second concern list of the hotspot users, the recommendation weight of each user to be recommended in the user list to be recommended is calculated, the recommended users are determined based on the recommendation weight and the preset rule, and the recommended users are recommended to the target users, so that personalized recommendation for the target users can be realized, the recommendation effect of network social user recommendation is remarkably improved, the probability of recommending interested recommended users to the target users is improved, and the social friend network of the target users is further mined.
The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments.
An embodiment of the present application further provides a computer storage medium, where the computer storage medium may store a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the image source tracking method according to the embodiments shown in fig. 1 to 9, and a specific execution process may refer to specific descriptions of the embodiments shown in fig. 1 to 9, which is not described herein again.
The present application further provides a computer program product, where at least one instruction is stored, and the at least one instruction is loaded by the processor and executes the image source tracking method according to the embodiment shown in fig. 1 to 9, where a specific execution process may refer to specific descriptions of the embodiment shown in fig. 1 to 9, and is not described herein again.
Please refer to fig. 11, which is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure. As shown in fig. 11, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105, at least one communication bus 1102.
Wherein a communication bus 1102 is used to enable connective communication between these components.
The user interface 1103 may include a Display screen (Display) and a Camera (Camera), and the optional user interface 1103 may also include a standard wired interface and a wireless interface.
The network interface 1104 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface).
Processor 1101 may include one or more processing cores, among other things. The processor 1101 connects various portions throughout the server 1100 using various interfaces and lines to perform various functions of the server 1100 and process data by executing or performing instructions, programs, code sets, or instruction sets stored in the memory 1105 and invoking data stored in the memory 1105. Optionally, the processor 1101 may be implemented in at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The processor 1101 may integrate one or more of a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a modem, and the like. Wherein, the CPU mainly processes an operating system, a user interface, an application program and the like; the GPU is used for rendering and drawing the content required to be displayed by the display screen; the modem is used to handle wireless communications. It is understood that the modem may not be integrated into the processor 1101, but may be implemented by a single chip.
The Memory 1105 may include a Random Access Memory (RAM) or a Read-Only Memory (Read-Only Memory). Optionally, the memory 1105 includes non-transitory computer-readable storage media. The memory 1105 may be used to store instructions, programs, code, sets of codes, or sets of instructions. The memory 1105 may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playing function, an image playing function, etc.), instructions for implementing the various method embodiments described above, and the like; the storage data area may store data and the like referred to in the above respective method embodiments. The memory 1105 may alternatively be at least one storage device located remotely from the processor 1101. As shown in FIG. 11, memory 1105, which is a type of computer storage medium, may include an operating system, a network communication module, a user interface module, and a recommended social user application.
In the electronic device 1100 shown in fig. 11, the user interface 1103 is mainly used as an interface for providing input for a user, and acquiring data input by the user; and the processor 1101 may be configured to invoke the recommended social friend application stored in the memory 1105, and specifically perform the following operations:
removing the hotspot users in the first attention list of the target user to obtain a second attention list; the hotspot users are users with the number of interactions exceeding an interaction threshold value and/or the number of concerned users exceeding an attention threshold value in the social platform;
based on the second concern list, obtaining a list of users to be recommended in a user database of the social platform; the user list to be recommended comprises at least one user to be recommended, and the attention list of the user to be recommended comprises at least one user in the second attention list;
calculating the recommendation weight of each user to be recommended based on the list of users to be recommended;
and acquiring a recommended user from the at least one user to be recommended based on the recommendation weight and a preset rule, and recommending the recommended user to the target user.
In a possible embodiment, the processor 1101 performs the following operations before removing the hot spot user in the first interest list of the target user to obtain the second interest list:
acquiring a recessive concern list and an explicit concern list of a target user based on a preset algorithm;
the first concern list is a union of the explicit concern list and the implicit concern list, the explicit concern list is a concern list of the target user displayed on the social platform, and the implicit concern list is a user set which is concerned by the target user in an indirect concern manner on the social platform.
In a possible embodiment, the processor 1101 executes the preset algorithm to obtain the implicit interest list and the explicit interest list of the target user, specifically executing the following operations:
acquiring a published text of the target user within a preset time period;
extracting user pronouns in the published text; wherein the user pronouns include at least one of: user name, user nickname, topic;
and determining a hidden user corresponding to the user appointed word based on the user appointed word, and constructing the hidden concern list based on the hidden user.
In a possible embodiment, the processor 1101 executes the preset algorithm to obtain the implicit interest list and the explicit interest list of the target user, specifically executing the following operations:
judging the activity of the target user;
determining that the target user is an inactive user when the activity of the target user is less than an activity threshold;
constructing a social network relationship network of the target user based on the mutual attention list of the target user;
acquiring active users in the social network relationship network based on the social network relationship network;
acquiring similar users based on the active users; the similar users are users of which the similarity between the active users and the target users exceeds a similarity threshold;
and acquiring the implicit concern list of the similar user based on the similar user, and taking the implicit concern list of the similar user as the implicit concern list of the target user.
In a possible embodiment, the processor 1101 performs the following operation of calculating the recommendation weight of each user to be recommended based on the list of users to be recommended:
acquiring an intersection list of the concern list of the user to be recommended and the second concern list of the target user;
calculating the interaction weight of the user to be recommended and each user in the intersection list within a preset time length;
and accumulating the interaction weights to obtain the recommendation weight of the user to be recommended.
In a possible embodiment, the processor 1101 performs the following operation to calculate the interaction weight between the user to be recommended and each user in the intersection list within the preset time period:
based on a preset time granularity, configuring an attenuation weight for the initial interaction weight in the preset time length to obtain an interaction attenuation function;
acquiring interaction data of the user to be recommended and each user in the intersection list, wherein the interaction data comprises interaction types, interaction frequencies and interaction timeliness;
and acquiring the interaction weight based on the interaction attenuation function and the interaction data.
In a possible embodiment, the processor 1101 performs the following operation to calculate the interaction weight between the user to be recommended and each user in the intersection list within the preset time period:
calculating an interaction aging factor TF (t) of the user to be recommended and each user in the intersection list, wherein:
Figure RE-GDA0003024291630000211
wherein T represents the time of the kth interaction after the preset time length is divided based on the first time granularity, and T is1Representing an interaction end time based on a second time granularity, the T0Representing an interaction start time based on a second time granularity;
acquiring an interaction type of the interaction between the user to be recommended and each user in the intersection list and an interaction type weight V (i, j) corresponding to the interaction type, wherein the value of V (i, j) is an integer between 1 and 5;
acquiring the kth interaction weight W of the user to be recommended and each user in the intersection list within the preset time length based on the interaction aging factor TF (t) and the interaction type weight V (i, j)k(i, j, t), wherein:
Wk(i，j，t)＝V(i，j，k)×TF(t)；
based on the interaction weight Wk(i, j, t) calculating the recommendation weight W (i, j), wherein:
Figure RE-GDA0003024291630000212
and K represents the total number of times of interaction between the recommended user and each user in the intersection list within the preset time length.
According to the method and the device, the user list to be recommended is obtained by removing the second concern list of the hotspot users, the recommendation weight of each user to be recommended in the user list to be recommended is calculated, the recommended users are determined based on the recommendation weight and the preset rule, and the recommended users are recommended to the target users, so that personalized recommendation for the target users can be realized, the recommendation effect of network social user recommendation is remarkably improved, the probability of recommending interested recommended users to the target users is improved, and the social friend network of the target users is further mined.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a read-only memory or a random access memory.
The above disclosure is only for the purpose of illustrating the preferred embodiments of the present application and is not to be construed as limiting the scope of the present application, so that the present application is not limited thereto, and all equivalent variations and modifications can be made to the present application.

Claims (10)

1. A method of recommending social users, the method comprising:
removing the hotspot users in the first attention list of the target user to obtain a second attention list; the hotspot users are users with the number of interactions exceeding an interaction threshold value and/or the number of concerned users exceeding an attention threshold value in the social platform;
based on the second concern list, obtaining a list of users to be recommended in a user database of the social platform; the user list to be recommended comprises at least one user to be recommended, and the attention list of the user to be recommended comprises at least one user in the second attention list;
calculating the recommendation weight of each user to be recommended based on the list of users to be recommended;
and acquiring a recommended user from the at least one user to be recommended based on the recommendation weight and a preset rule, and recommending the recommended user to the target user.
2. The method of claim 1, further comprising, before said removing the hot users in the first interest list of the target user and obtaining a second interest list, the step of:
acquiring a recessive concern list and an explicit concern list of a target user based on a preset algorithm;
the first concern list is a union of the explicit concern list and the implicit concern list, the explicit concern list is a concern list of the target user displayed on the social platform, and the implicit concern list is a user set which is concerned by the target user in an indirect concern manner on the social platform.
3. The method of claim 2, wherein the obtaining of the implicit attention list and the explicit attention list of the target user based on the preset algorithm comprises:
acquiring a published text of the target user within a preset time period;
extracting user pronouns in the published text; wherein the user pronouns include at least one of: user name, user nickname, topic;
and determining a hidden user corresponding to the user appointed word based on the user appointed word, and constructing the hidden concern list based on the hidden user.
4. The method of claim 2, wherein the obtaining of the implicit attention list and the explicit attention list of the target user based on the preset algorithm comprises:
judging the activity of the target user;
determining that the target user is an inactive user when the activity of the target user is less than an activity threshold;
constructing a social network relationship network of the target user based on the mutual attention list of the target user;
acquiring active users in the social network relationship network based on the social network relationship network;
acquiring similar users based on the active users; the similar users are users of which the similarity between the active users and the target users exceeds a similarity threshold;
and acquiring the implicit concern list of the similar user based on the similar user, and taking the implicit concern list of the similar user as the implicit concern list of the target user.
5. The method according to claim 1, wherein the calculating the recommendation weight of each user to be recommended based on the list of users to be recommended comprises:
acquiring an intersection list of the concern list of the user to be recommended and the second concern list of the target user;
calculating the interaction weight of the user to be recommended and each user in the intersection list within a preset time length;
and accumulating the interaction weights to obtain the recommendation weight of the user to be recommended.
6. The method according to claim 5, wherein the calculating the interaction weight between the user to be recommended and each user in the intersection list within a preset time period includes:
based on a preset time granularity, configuring an attenuation weight for the initial interaction weight in the preset time length to obtain an interaction attenuation function;
acquiring interaction data of the user to be recommended and each user in the intersection list, wherein the interaction data comprises interaction types, interaction frequencies and interaction timeliness;
and acquiring the interaction weight based on the interaction attenuation function and the interaction data.
7. The method according to claim 5, wherein the calculating the interaction weight between the user to be recommended and each user in the intersection list within a preset time period includes:
calculating an interaction aging factor TF (t) of the user to be recommended and each user in the intersection list, wherein:
Figure FDA0002940092010000031
wherein T represents the time of the kth interaction after the preset time length is divided based on the first time granularity, and T is1Representing an interaction end time based on a second time granularity, the T0Representing an interaction start time based on a second time granularity;
acquiring an interaction type of the interaction between the user to be recommended and each user in the intersection list and an interaction type weight V (i, j) corresponding to the interaction type, wherein the value of V (i, j) is an integer between 1 and 5;
acquiring the kth interaction weight W of the user to be recommended and each user in the intersection list within the preset time length based on the interaction aging factor TF (t) and the interaction type weight V (i, j)k(i, j, t), wherein:
Wk(i，j，t)＝V(i，j，k)×TF(t)；
based on the interaction weight Wk(i, j, t) calculating the recommendation weight W (i, j), wherein:
Figure FDA0002940092010000032
and K represents the total number of times of interaction between the recommended user and each user in the intersection list within the preset time length.
8. An apparatus for recommending social users, the apparatus comprising:
the hot spot removing module is used for removing the hot spot users in the first attention list of the target user to obtain a second attention list; the hotspot users are users with the number of interactions exceeding an interaction threshold value and/or the number of concerned users exceeding an attention threshold value in the social platform;
the list obtaining module is used for obtaining a list of users to be recommended from a user database of the social platform based on the second concern list; the user list to be recommended comprises at least one user to be recommended, and the attention list of the user to be recommended comprises at least one user in the second attention list;
the calculation weight module is used for calculating the recommendation weight of each user to be recommended in the user list to be recommended based on the user list to be recommended;
and the recommending user module is used for acquiring a recommending user from the at least one user to be recommended based on the recommending weight and a preset rule and recommending the recommending user to the target user.
9. A computer storage medium, characterized in that it stores a plurality of instructions adapted to be loaded by a processor and to carry out the method steps according to any one of claims 1 to 7.
10. An electronic device, comprising: a processor and a memory; wherein the memory stores a computer program adapted to be loaded by the processor and to perform the method steps of any of claims 1 to 7.
CN202110174284.6A 2021-02-09 2021-02-09 Method and device for recommending social users, storage medium and electronic equipment Active CN112836127B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110174284.6A CN112836127B (en) 2021-02-09 2021-02-09 Method and device for recommending social users, storage medium and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110174284.6A CN112836127B (en) 2021-02-09 2021-02-09 Method and device for recommending social users, storage medium and electronic equipment

Publications (2)

Publication Number Publication Date
CN112836127A true CN112836127A (en) 2021-05-25
CN112836127B CN112836127B (en) 2023-06-02

Family

ID=75932888

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110174284.6A Active CN112836127B (en) 2021-02-09 2021-02-09 Method and device for recommending social users, storage medium and electronic equipment

Country Status (1)

Country Link
CN (1) CN112836127B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113934941A (en) * 2021-10-12 2022-01-14 北京朗玛数联科技有限公司 User recommendation system and method based on multi-dimensional information

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110246907A1 (en) * 2010-03-31 2011-10-06 Wang James H Promoting participation of low-activity users in social networking system
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
CN107679239A (en) * 2017-10-27 2018-02-09 天津理工大学 Recommend method in a kind of personalized community based on user behavior
CN111782963A (en) * 2020-06-15 2020-10-16 中国铁塔股份有限公司 Social network data mining method and system based on SNS and service equipment

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110246907A1 (en) * 2010-03-31 2011-10-06 Wang James H Promoting participation of low-activity users in social networking system
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
CN107679239A (en) * 2017-10-27 2018-02-09 天津理工大学 Recommend method in a kind of personalized community based on user behavior
CN111782963A (en) * 2020-06-15 2020-10-16 中国铁塔股份有限公司 Social network data mining method and system based on SNS and service equipment

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113934941A (en) * 2021-10-12 2022-01-14 北京朗玛数联科技有限公司 User recommendation system and method based on multi-dimensional information

Also Published As

Publication number Publication date
CN112836127B (en) 2023-06-02

Similar Documents

Publication Publication Date Title
CN106357517B (en) Directional label generation method and device
US20190362252A1 (en) Learning user preferences in a conversational system
US10628030B2 (en) Methods and systems for providing user feedback using an emotion scale
CN112868004B (en) Resource recommendation method and device, electronic equipment and storage medium
KR101733011B1 (en) Apparatus for providing recommendation based social network service and method using the same
US20140201125A1 (en) Conversation management systems
CN106487655B (en) Message interaction method and device and processing server
KR101873339B1 (en) System and method for providing interest contents
CN111078065A (en) Method, system and readable recording medium for collecting non-reply message
CN112532507A (en) Method and device for presenting expression image and method and device for sending expression image
JP5881647B2 (en) Determination device, determination method, and determination program
KR102676791B1 (en) Method and system for providing multiple profiles
US9058090B1 (en) Collaborative information sharing in a virtual world
KR20190003709A (en) Facilitate efficient retrieval using message exchange threads
CN111787042A (en) Method and device for pushing information
CN112836127B (en) Method and device for recommending social users, storage medium and electronic equipment
CN107562788B (en) Interaction method, interaction device and computer-readable storage medium
JP2020057221A (en) Information processing method, information processing device, and program
CN106776990A (en) A kind of information processing method, device and electronic equipment
KR20210052912A (en) Method and apparatus for recommending app function shortcuts through app usage pattern and conversation analysis
CN112348614B (en) Method and device for pushing information
CN114677168A (en) Resource recommendation method, device, equipment and medium
CN114969493A (en) Content recommendation method and related device
CN112861007A (en) Screen saver display method, device, equipment, medium and program product
CN104991973A (en) Method and apparatus for determining area of interest of users

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