CN111079027A - User recommendation method and device, readable storage medium and electronic equipment - Google Patents
User recommendation method and device, readable storage medium and electronic equipment Download PDFInfo
- Publication number
- CN111079027A CN111079027A CN201911203811.0A CN201911203811A CN111079027A CN 111079027 A CN111079027 A CN 111079027A CN 201911203811 A CN201911203811 A CN 201911203811A CN 111079027 A CN111079027 A CN 111079027A
- Authority
- CN
- China
- Prior art keywords
- user
- node
- user node
- distance
- organization
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 230000008520 organization Effects 0.000 claims description 128
- 230000007246 mechanism Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 12
- 238000010586 diagram Methods 0.000 description 12
- 239000008186 active pharmaceutical agent Substances 0.000 description 9
- 238000004891 communication Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- G06F16/9536—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)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure relates to a user recommendation method, a user recommendation device, a readable storage medium and an electronic device. The method comprises the following steps: determining a target user node corresponding to a user account in a friend relationship subnet according to a currently logged user account, and determining a first distance between the target user node and at least one other user node which is not connected with the target user node in the friend relationship subnet; determining a second distance between the target user node and the at least one other user node according to the organizational relationship subnet; determining a recommended user according to at least the first distance and the second distance; and sending the user information of the recommended user to the terminal so that the terminal displays the user information through an enterprise social application program. Therefore, the time required by adding friends among users can be shortened to a certain extent, the user operation is reduced, and the user experience is improved.
Description
Technical Field
The present disclosure relates to the field of computer software, and in particular, to a user recommendation method, apparatus, readable storage medium, and electronic device.
Background
With the rapid development of mobile internet technology, the use of enterprise social software is more and more common. For enterprises, enterprise social software facilitates enterprise management, and for employees, enterprise social software facilitates communication among the enterprises.
In the current enterprise social software, if a user wants to request other users as friends, the user needs to accurately input information of the other users, such as names, job numbers or mobile phone numbers, and then click to find the other users, so that the operation is very complicated and the efficiency is low. Moreover, when there are employees with a heavy name in the enterprise, or when the user does not know the unique information such as the job number, the mobile phone number, etc. of the other user, there may be a case where the request is wrong.
Disclosure of Invention
The invention aims to provide a user recommendation method, a user recommendation device, a readable storage medium and electronic equipment, so that time required by adding friends among users is shortened to a certain extent, user operation is reduced, and user experience is improved.
In order to achieve the above object, in a first aspect, the present disclosure provides a user recommendation method, including:
determining a target user node corresponding to a user account in a friend relation subnet according to the user account currently logged in by an enterprise social application program running on a terminal, and determining a first distance between the target user node and at least one other user node which is not connected with the target user node in the friend relation subnet, wherein each user node in the friend relation subnet corresponds to a user account respectively, the connection between the user nodes represents that two user accounts corresponding to the two user nodes have friend relation, and the first distance is used for representing the friend association degree between users represented by the user nodes;
determining a second distance between a target user node and the at least one other user node according to an organization relation subnet, wherein each organization node in the organization relation subnet corresponds to an organization mechanism, the connection between the organization nodes represents that the two organization mechanisms corresponding to the two organization nodes have an upper-level and lower-level relation, the connection between the user node and the organization node represents that a user represented by the user node belongs to the organization mechanism corresponding to the organization node, and the second distance is used for representing the organization association degree between the users represented by the user node;
determining a recommended user according to at least the first distance and the second distance;
and sending the user information of the recommended user to a terminal so that the terminal displays the user information through the enterprise social application program.
Optionally, determining a first distance between the target user node and at least one other user node comprises:
and for a first other user node in at least one other user node, determining a first distance between the target user node and the first other user node according to the shortest path length between the first other user node and the target user node, the number of the shortest paths and the number of the user nodes connected with the first other user node in the friend relationship subnet, wherein the first other user node is a user node which is not zero in number of the user nodes connected with the first other user node in the friend relationship subnet.
Optionally, determining a first distance between the target user node and at least one other user node further includes:
and aiming at a second other user node in at least one other user node, determining a first distance between the target user node and the second other user node according to the determined maximum first distance between the target user node and the first other user node, wherein the first distance between the target user node and the second other user node is greater than the maximum first distance between the target user node and the first other user node.
Optionally, the first distance between the target user node and the first other user node is determined by the following formula:
wherein SF (i, j) represents a first distance between user node i and user node j; d (i, j) represents the shortest path length between the user node i and the user node j; n (i, j) represents the number of the shortest paths; and d (j) represents the number of user nodes connected to user node j.
Optionally, determining a second distance between the target user node and the at least one other user node comprises:
and for each other user node, determining a second distance between the target user node and the other user node according to the organization layer level difference between the organization node to which the other user node belongs and the organization node to which the target user node belongs and the shortest path length between the organization node to which the other user node belongs and the organization node to which the target user node belongs in the organization relation subnet.
Optionally, the second distance between the target user node and the other user node is determined by the following formula:
SO(i,j)=o(i,j)+do(i,j)
wherein SO (i, j) represents a second distance between user node i and user node j; o (i, j) represents the organizational layer level difference between the organizational node to which the user node j belongs and the organizational node to which the user node i belongs; and do (i, j) represents the shortest path length between the organization node to which the user node j belongs and the organization node to which the user node i belongs.
Optionally, the method further comprises:
acquiring personal characteristic information of a target user represented by a target user node according to a user account currently logged in by an enterprise social application program;
acquiring personal characteristic information of other users represented by each other user node according to the user account corresponding to each other user node;
for each other user, determining a similarity parameter between the other user and the target user according to the personal characteristic information of the other user and the personal characteristic information of the target user;
determining a recommended user according to at least the first distance and the second distance, including:
and determining the recommended user according to the first distance, the second distance and the similarity parameter.
Optionally, the personal characteristic information includes multi-dimensional characteristic information;
determining a similarity parameter between the other user and the target user according to the personal characteristic information of the other user and the personal characteristic information of the target user, including:
determining similarity parameters between the other users and the target user in the dimension according to the feature information of the other users and the target user in the same dimension;
and determining the similarity parameter between the other user and the target user according to the similarity parameter between the other user and the target user in each dimension.
Optionally, the personal characteristic information includes: gender, age, scholarly calendar and position;
determining a similarity parameter G (i, j) between the other user and the target user in the dimension of gender through the following formula:
wherein, giRepresenting the gender of the target user i; gjRepresents the gender of other users j;
determining a similarity parameter A (i, j) between the other user and the target user in the dimension of age by the following formula:
wherein, aiRepresents the age of the target user i; a isjRepresents the age of other users j;
determining a similarity parameter P (i, j) in the job dimension between the other user and the target user by the following formula:
wherein D (p)i,pj) Post p representing target user iiJob p with other users jjThe corresponding inter-node distances in the job tree;
determining a similarity parameter E (i, j) between the other user and the target user in the learning dimension by the following formula:
wherein D (e)i,ej) Calendar e representing target user iiCalendar e with other users jjThe corresponding inter-node distances in the learned tree.
In a second aspect, the present disclosure provides a user recommendation device, including:
the system comprises a first determining unit, a second determining unit and a third determining unit, wherein the first determining unit is used for determining a target user node corresponding to a user account in a friend relation subnet according to the user account currently logged in by an enterprise social application program running on a terminal, and determining a first distance between the target user node and at least one other user node which is not connected with the target user node in the friend relation subnet, each user node in the friend relation subnet corresponds to one user account respectively, the connection between the user nodes represents that two user accounts corresponding to the two user nodes have friend relation, and the first distance is used for representing the friend association degree between users represented by the user nodes;
a second determining unit, configured to determine, according to an organization relation subnet, a second distance between a target user node and the at least one other user node, where each organization node in the organization relation subnet corresponds to an organization respectively, a connection between the organization nodes represents that two organizations corresponding to the two organization nodes have a hierarchical relationship, the connection between the user node and the organization node represents that a user represented by the user node belongs to the organization corresponding to the organization node, and the second distance is used to represent an organization association degree between users represented by the user node;
a third determining unit, configured to determine a recommended user according to at least the first distance and the second distance;
and the sending unit is used for sending the user information of the recommended user to the terminal so that the terminal can display the user information through the enterprise social application program.
In a third aspect, the present disclosure provides a computer-readable storage medium, on which a computer program is stored, which when executed by a processor, implements the steps of the user recommendation method of the first aspect described above.
In a fourth aspect, the present disclosure provides an electronic device comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to implement the steps of the user recommendation method of the first aspect.
In the technical scheme, two information networks, namely a friend relation subnet and an organization relation subnet, are utilized. The friend relationship existing between the users is recorded in the friend relationship subnet, and the friend association degree between the two users, namely the probability of the two users contacting through the friend relationship subnet, can be determined through the friend relationship subnet. In addition, the organizational structure to which the existing user belongs and the hierarchical relationship between the organizational structures are recorded in the organizational relationship subnet, and the organizational association degree between two users, that is, the probability of the two users contacting through the organizational relationship subnet can be determined through the organizational relationship subnet. Therefore, the association degrees of the users in two dimensions of friend relationship and organization relationship can be integrated, and the recommended users suitable for being recommended to the target users can be determined. Because the relevancy of the user organization relationship is considered, the method is more suitable for user recommendation in enterprise social software, and the recommendation precision is higher. By feeding back the user information of the recommended user to the target user, the time required by the target user to add the friend can be shortened to a certain extent, user operation is reduced, the probability of wrong requests is reduced, and user experience is improved.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure without limiting the disclosure. In the drawings:
FIG. 1 is a schematic diagram illustrating one implementation environment according to an exemplary embodiment of the present disclosure.
Fig. 2 is a flowchart illustrating a user recommendation method according to an exemplary embodiment of the present disclosure.
Fig. 3 shows a schematic diagram of a user relationship network.
Fig. 4 is a flowchart illustrating another user recommendation method according to an exemplary embodiment of the present disclosure.
Fig. 5 shows a schematic diagram of a calendar tree.
Fig. 6 shows a schematic diagram of a job tree.
Fig. 7 is a block diagram illustrating a user recommendation device according to an exemplary embodiment of the present disclosure.
FIG. 8 is a block diagram illustrating an electronic device in accordance with an example embodiment.
Detailed Description
In the related art, friend recommendation is generally made according to the degree of association between personal feature information of users. For example, if two non-friend users have the same preference, one user is recommended to the other user. Or, recommending the users according to the friend association degree among the users. For example, if two non-friend users have a common friend, one user is recommended to the other user. The two recommendation methods can achieve relatively high recommendation precision in daily social application, but the recommendation precision is not high when the two recommendation methods are applied to enterprise social application.
The inventors have found this problem and have found through research that the reason for the lack of precision is that none of the above methods takes into account the organizational relevance between users. In fact, users in the enterprise social application generally belong to one enterprise, and the communication demand among the users is mostly derived from daily work. If only the relevance between the personal characteristic information of the two users or the relevance of the friends is considered, the finally recommended friends may not be actually connected in the enterprise social application. Particularly, when a new employee joins in, the new employee does not have a friend relationship with any other employee, so the recommendation method based on friend association degree cannot perform friend recommendation for the new employee.
In conclusion, the existing user recommendation method is directly applied to the enterprise social software, and the problem of low recommendation precision exists. In addition, too many invalid user recommendations also result in occupying network bandwidth, resulting in waste of network resources.
In order to solve the above problems, the present disclosure provides a user recommendation method and apparatus. The following detailed description of specific embodiments of the present disclosure is provided in connection with the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present disclosure, are given by way of illustration and explanation only, not limitation.
FIG. 1 is a schematic diagram illustrating one implementation environment according to an exemplary embodiment of the present disclosure. As shown in FIG. 1, the implementation environment may include: a plurality of terminals 101~10nAnd a server 20. Each terminal is installed with an enterprise social application program, and users of a plurality of terminals can communicate with each other in an instant manner through the enterprise social application program and the server 20. The terminal and the server 20 may communicate via various wired or wireless networks, such as any of bluetooth, WiFi, 2G, 3G, 4G, 5G, NB-IOT, etc.
When a user of the terminal logs in the enterprise social application program, or in response to a recommendation request manually input by the terminal user in the enterprise social application program, the server 20 may perform user recommendation on the terminal user, so as to facilitate the operation of adding friends by the user, shorten the time required for adding friends between users to a certain extent, reduce user operation, and improve user experience.
Fig. 2 is a flowchart illustrating a user recommendation method according to an exemplary embodiment of the disclosure, which may be applied to a server, for example, a social application server. As shown in fig. 2, the method may include four steps S201, S202, S203, and S204.
In S201, according to a user account currently logged in by an enterprise social application running on a terminal, a target user node corresponding to the user account in a friend relationship subnet is determined, and a first distance between the target user node and at least one other user node in the friend relationship subnet that is not connected to the target user node is determined.
In the present disclosure, the friend relationship subnet is used to record friend relationships between users. Fig. 3 shows a schematic diagram of a user relationship network. As shown in fig. 3, the user relationship network may include a friend relationship sub-network and an organization relationship sub-network, wherein the friend relationship sub-network is formed by user nodes represented by rectangular boxes shown in fig. 3 and connection relationships therebetween, and the organization relationship sub-network is formed by organization nodes represented by circular boxes shown in fig. 3 and connection relationships therebetween.
Each user node in the friend relation subnet corresponds to a user account respectively, and the connection between the user nodes represents that two user accounts corresponding to the two user nodes have friend relation. For example, as shown in fig. 3, the user accounts of the user node 1 and the user node 5 have a friend relationship, that is, the users represented by the two user nodes are friends; the user accounts corresponding to the user node 1 and the user node 4 do not have a friend relationship, that is, the users represented by the two user nodes are not friends.
And determining a corresponding target user node in the friend relation subnet according to the currently logged user account. At least one other user node mentioned in S201 is a user node that is not connected to the target user node in the buddy relationship subnet. If the target user node is user node 1, the at least one other user node comprises: user node 2, user node 3, user node 4, user node 6, user node 7, user node 8, user node 9, user node 10 and user node 11.
The first distance mentioned in S201 is used to characterize the friend association degree between the users represented by the user node, that is, the probability that the two users contact through the friend-relationship subnet.
In S202, a second distance between the target user node and the at least one other user node is determined according to an organization relation subnet, where each organization node in the organization relation subnet corresponds to an organization respectively, a connection between the organization nodes represents that two organizations corresponding to the two organization nodes have a hierarchical relationship, and a connection between a user node and an organization node represents that a user represented by the user node belongs to the organization corresponding to the organization node.
For example, as shown in FIG. 3, the architecture formed by the organization nodes 1-7 embodies an organizational architecture. The lower mechanisms of the organization mechanism corresponding to the organization node 1 are organization mechanisms corresponding to the organization nodes 2 and 3, the lower mechanisms of the organization mechanism corresponding to the organization node 2 are organization mechanisms corresponding to the organization nodes 4 and 5, and the lower mechanisms of the organization mechanism corresponding to the organization node 3 are organization mechanisms corresponding to the organization nodes 6 and 7.
Further, in the user relationship network shown in fig. 3, the organization relationship subnet and the friend relationship subnet are merged together, and the two are linked together based on which organization the user belongs to. For example, since the user represented by the user node 1 belongs to the organization corresponding to the organization node 4, the organization node 4 is connected to the user node 1.
The second distance mentioned in S202 is used to characterize the organizational relevance between the users represented by the user node, i.e., the probability that the two users are contacted through the organizational relationship subnet.
In S203, a recommended user is determined according to at least the first distance and the second distance.
That is, a recommended user to be recommended to the target user represented by the target user node is determined from the users represented by the at least one other user node based on at least two parameters, i.e., the first distance determined in S201 and the second distance determined in S202.
In S204, the user information of the recommended user is sent to the terminal, so that the terminal presents the user information through the enterprise social application program.
By the technical scheme, the association degrees of the users in two dimensions of the friend relationship and the organization relationship can be integrated, and the recommended users suitable for being recommended to the target users can be determined. Because the relevancy of the user organization relationship is considered, the method is more suitable for user recommendation in enterprise social software, and the recommendation precision is higher. By feeding back the user information of the recommended user to the target user, the time required by the target user to add the friend can be shortened to a certain extent, user operation is reduced, the probability of wrong requests is reduced, and user experience is improved.
In one embodiment, the first distance in S201 may be determined as follows:
and for a first other user node in the at least one other user node, determining a first distance between the target user node and the first other user node according to the shortest path length between the first other user node and the target user node, the number of the shortest paths and the number of the user nodes connected with the first other user node in the friend relationship subnet, wherein the first other user node is a user node of which the number of the user nodes connected with the first other user node in the friend relationship subnet is not zero.
Continuing with the above example, assuming that the target user node is user node 1, the at least one other user node includes user node 2, user node 3, user node 4, user node 6, user node 7, user node 8, user node 9, user node 10, and user node 11. Since the user node 8 has no connection to any other user node, it does not belong to the first other user node. In this example, the first further user nodes comprise user node 2, user node 3, user node 4, user node 6, user node 7, user node 9, user node 10 and user node 11. Next, for each first other user node, determining, in the buddy relationship subnet, a shortest path length between the first other user node and the target user node, the number of the shortest paths, and the number of user nodes connected to the first other user node, as shown in table 1:
|
2 | 3 | 4 | 6 | 7 | 9 | 10 | 11 |
D(i,j) | 2 | 4 | 2 | 2 | 3 | 2 | 3 | 3 |
d(j) | 1 | 1 | 1 | 2 | 2 | 4 | 3 | 2 |
n(i,j) | 1 | 1 | 1 | 1 | 2 | 1 | 1 | 1 |
SF(i,j) | 2 | 4 | 2 | 1 | 0.75 | 0.5 | 1 | 1.5 |
TABLE 1
Wherein D (i, j) represents the shortest path length between the user node i and the user node j; n (i, j) represents the number of the shortest paths; d (j) represents the number of user nodes connected to user node j; SF (i, j) represents a first distance between user node i and user node j.
For example, a first distance between user node i and user node j belonging to the first other user node may be determined by the following formula:
in another embodiment, the first distance in S201 may be determined as follows:
and aiming at a second other user node in the at least one other user node, determining a first distance between the target user node and the second other user node according to the determined maximum first distance between the target user node and the first other user node, wherein the first distance between the target user node and the second other user node is greater than the maximum first distance between the target user node and the first other user node.
Continuing with the above example, user node 8 belongs to a second other user node since it has no connection with any other user node. In this way, the first distance between the target user node 1 and the second further user node 8 may be determined according to the largest first distance, i.e. 4, as shown in table 1. For example, a first distance between user node i and user node j belonging to a second other user node may be determined according to the following formula:
SF(i,j)=Max(SF')+1 (2)
where Max (SF') represents the largest first distance between the target user node and the first other user node.
As such, the first distance between the target user node 1 and the second further user node 8 is 5.
It should be appreciated that although the present disclosure presents an embodiment for calculating the first distance between the target user node and the second other user node as shown in equation (2), the present disclosure is not so limited as long as it suffices that the determined first distance between the target user node and the second other user node is greater than the maximum first distance between the target user node and the first other user node.
In one embodiment, the second distance in S202 may be determined by:
and for each other user node in the at least one other user node, determining a second distance between the target user node and the other user node according to the organization level difference between the organization node to which the other user node belongs and the organization node to which the target user node belongs and the shortest path length between the organization node to which the other user node belongs and the organization node to which the target user node belongs in the organization relation subnet.
Continuing with the above example, assuming that the target user node is user node 1, the at least one other user node includes user node 2, user node 3, user node 4, user node 6, user node 7, user node 8, user node 9, user node 10, and user node 11. Next, for each other user node, determining an organization level difference between the organization node to which the other user node belongs and the organization node to which the target user node belongs, and a shortest path length between the organization node to which the other user node belongs and the organization node to which the target user node belongs in the organization relation subnet, as shown in table 2:
|
2 | 3 | 4 | 6 | 7 | 8 | 9 | 10 | 11 |
o(i,j) | 1 | 0 | 0 | 2 | 0 | 1 | 0 | 0 | 0 |
do(i,j) | 1 | 4 | 0 | 2 | 4 | 3 | 2 | 4 | 4 |
SO(i,j) | 2 | 4 | 0 | 4 | 4 | 4 | 2 | 4 | 4 |
TABLE 2
For example, the second distance between the target user node and each other user node may be determined by the following formula:
SO(i,j)=o(i,j)+do(i,j) (3)
wherein SO (i, j) represents a second distance between user node i and user node j; o (i, j) represents the organizational layer level difference between the organizational node to which the user node j belongs and the organizational node to which the user node i belongs; and do (i, j) represents the shortest path length between the organization node to which the user node j belongs and the organization node to which the user node i belongs.
After determining the first distance and the second distance, one embodiment of S203 may be:
and for each other user node, normalizing the first distance between the target user node and the other user node, normalizing the second distance between the target user node and the other user node, adding, and taking the sum as a similarity parameter between the target user represented by the target user node and the other user represented by the other user node. For the sake of distinction from other similarity parameters described below, this similarity parameter is referred to herein as a first similarity parameter.
For example, the first similarity parameter DS (i, j) is obtained by the following formula (4):
where min (sf) represents a minimum first distance value between the target user node and the other user nodes, max (sf) represents a maximum first distance value between the target user node and the other user nodes, min (so) represents a minimum second distance value between the target user node and the other user nodes, max (so) represents a maximum second distance value between the target user node and the other user nodes, and DS (i, j) represents a first similarity parameter between the target user represented by the target user node and the other users represented by the other user nodes. Wherein, the smaller the DS (i, j), the more similar the characterization.
Continuing with the above example, the results shown in Table 3 are obtained according to equation (4):
|
2 | 3 | 4 | 6 | 7 | 8 | 9 | 10 | 11 |
DS(i,j) | 5/6 | 16/9 | 1/3 | 10/9 | 19/18 | 2 | 1/2 | 10/9 | 11/9 |
TABLE 3
According to the sequence from small to large of DS (i, j), the sequence of other user nodes 2-11 can be obtained as follows:
user node 4-user node 9-user node 2-user node 7-user node 6-user node 10-user node 11-user node 3-user node 8.
Thereafter, a recommended user may be determined accordingly. For example, the other users represented by the other user nodes ranked at the top N are determined as the recommended users, where N is less than the total number of the at least one other user node. In this way, in S204, the determined user information of the recommended user may be sent to the terminal for display, so as to complete the friend recommendation operation. The user information may include, for example, name, department, job title, contact address, and the like.
One embodiment of S203 is described above, in which the friend association and the organization association between users are considered. In another embodiment of the present disclosure, in order to improve recommendation accuracy, on the basis of the friend relevance and the organization relevance, similarity between personal feature information of users is also considered, as described below in conjunction with fig. 4.
Fig. 4 is a flowchart illustrating another user recommendation method according to an exemplary embodiment of the present disclosure. As shown in fig. 4, the method may further include the following steps in addition to the above S201 and S202:
in S205, the personal feature information of the target user represented by the target user node is obtained according to the user account currently logged in by the enterprise social application.
In S206, the personal feature information of the other users represented by each other user node is obtained according to the user account corresponding to each other user node.
In S207, for each other user, a similarity parameter between the other user and the target user is determined according to the personal characteristic information of the other user and the personal characteristic information of the target user. For the purpose of distinguishing from the first similarity parameter described above, the similarity parameter determined by this S207 will be referred to as a second similarity parameter herein.
Thus, S203 is: and determining the recommended user according to the first distance determined in the S201, the second distance determined in the S202 and the second similarity parameter determined in the S207.
In this embodiment, the user account and the personal characteristic information of the user may be stored in association in the server. Therefore, the server can obtain the corresponding personal characteristic information through query operation according to the user account. The personal characteristic information of the user may include characteristic information of multiple dimensions. Illustratively, the personal characteristic information may include one or more of: gender, age, school calendar and position. In order to achieve more accurate similarity matching, the personal characteristic information includes four of gender, age, academic calendar and position. Therefore, the personal characteristic information is selected by considering that in the enterprise social application program, people with different genders generally have different hobbies and people with different ages contact different things; people in different positions have different work contents, and people in the same study can have similar topics.
For example, the Jaccard similarity coefficient may be employed to characterize a second similarity parameter between the other user and the target user. The conventional Jaccard similarity coefficient is defined as: the proportion of the number of intersection elements of the two sets A and B in the A, B union is as follows:
for example, taking six users a, b, c, d, e, and f as an example, the personal feature information of the six users is shown in table 4:
sex | Age (age) | Position of employment | Study calendar | |
a | For male | 25 | Special person for tongue and groove | This section |
b | For male | 27 | Special advertisement member | Master's soldier |
c | For male | 25 | Creative specialist | This section |
d | Woman | 29 | Tongued and grooved main pipe | This section |
e | Woman | 28 | Advertising supervisor | This section |
f | Woman | 26 | Special clerk of the official document | This section |
TABLE 4
For user a and user b, the only difference between them is the same, so according to equation (5), the similarity between user a and user b can be found as follows:
similarly, the similarity between user a and the other four users is:
thus, the order of recommending friends for user a is c, followed by b/d/e/f. The recommendations of the other four users except the user c are parallel, the granularity is coarse, and more accurate recommendation cannot be realized. In this regard, the present disclosure improves upon this problem.
That is, in the case where the personal feature information of the user includes multi-dimensional feature information, S207 may further include:
determining similarity parameters between the other users and the target user in the dimension according to the feature information of the other users and the target user in the same dimension; and determining the similarity parameter between the other user and the target user according to the similarity parameter between the other user and the target user in each dimension.
Illustratively, the similarity parameter G (i, j) in the dimension of gender between the other user and the target user is determined by the following formula:
wherein, giRepresenting the gender of the target user i; gjIndicating the gender of the other user j.
Illustratively, the similarity parameter a (i, j) in the age dimension between the other user and the target user is determined by the following formula:
wherein, aiRepresents the age of the target user i; a isjIndicating the age of the other user j.
Illustratively, the similarity parameter E (i, j) in the academic dimension between the other user and the target user is determined by the following formula:
wherein D (e)i,ej) Calendar e representing target user iiCalendar e with other users jjThe corresponding inter-node distances in the learned tree.
Fig. 5 shows a schematic diagram of a calendar tree. As shown in fig. 5, each academic name corresponds to a node in the tree, and the hierarchical relationship between the nodes reflects the low-to-high distribution of the academic. The distance between two nodes in the learned tree can be understood as the length of the shortest path between the nodes.
Illustratively, the similarity parameter P (i, j) in the job dimension between the other user and the target user is determined by the following formula:
wherein D (p)i,pj) Post p representing target user iiJob p with other users jjThe corresponding inter-node distances in the job tree.
Fig. 6 shows a schematic diagram of a job tree. As shown in fig. 6, each job title corresponds to a node in the tree, and the hierarchical relationship between the nodes reflects the distribution of jobs from high to low. The distance between two nodes in the job tree can be understood as the length of the shortest path between the nodes.
For example, assuming that the gender of the user i is the same as that of the user j, the age of the user i is 25, the age of the user j is 27, the position of the user i is a planning specialist, the position of the user j is an advertising specialist, the academic story of the user i is a subject, and the academic story of the user j is a master, then, according to the above equations (6) to (9), it is possible to obtain: g (i, j) is 1, a (i, j) is 1/3, E (i, j) is 1/2, and P (i, j) is 1/5.
After the similarity parameter of each dimension between the other user and the target user is determined, the similarity parameter of each dimension between the other user and the target user is determined according to the similarity parameter of each dimension between the other user and the target user.
Illustratively, the improved Jaccard similarity formula is obtained by combining four dimensions:
wherein J '(i, J) represents a second similarity parameter between the user i and the user J, i ∪ J represents the number of elements of the union of the personal characteristic information between the user i and the user J, and the larger J' (i, J) is, the more similar the representation is.
From equation (10), the second similarity parameter J' (a, b) for users a and b listed in table 4 is:
obtaining the following by the same method: j '(a, c) is 0.640, J' (a, d) is 0.243, J '(a, e) is 0.214, and J' (a, f) is 0.262.
Therefore, the friend recommendation sequence for the user a is obtained as follows: c, b, f, d, e. The improved implementation presented in this disclosure may provide finer grained recommendation results than the prior art.
Then, in S203, a target similarity parameter between the target user and each of the other users is determined according to the first distance determined in S201, the second distance determined in S202, and the second similarity parameter determined in S207, and a recommended user is determined according to the target similarity parameter.
Illustratively, this is determined using the above equation (4)Determining a first similarity parameter DS (i, j), inverting it to obtain M, and usingPerforming normalization processing, wherein min (M) represents the minimum value in M, and max (M) represents the maximum value in M, so as to obtain a third similarity parameter DS' (i, j):
the trend of the third similarity parameter DS '(i, J) obtained after the above-mentioned processing is the same as that of the second similarity parameter J' (a, b), that is, the larger the value, the more similar the users are.
And then, according to the third similarity parameter and the second similarity parameter, obtaining a target similarity parameter by the following formula:
R(i,j)=J'(i,j)+DS'(i,j) (12)
wherein R (i, j) represents a target similarity parameter between user i and user j.
Thereafter, a recommended user may be determined accordingly. For example, according to the descending order of the target similarity parameter R (i, j), determining that the other users represented by the other user nodes ranked at the top N are recommended users, where N is less than the total number of the at least one other user node. In this way, in S204, the determined user information of the recommended user may be sent to the terminal for display, so as to complete the friend recommendation operation.
Based on the same inventive concept, the disclosure also provides a user recommendation device. Fig. 7 is a block diagram illustrating a user recommendation device 700 according to an exemplary embodiment of the present disclosure. As shown in fig. 7, the apparatus 700 may include:
a first determining unit 701, configured to determine, according to a user account currently logged in by an enterprise social application running on a terminal, a target user node corresponding to the user account in a friend relationship subnet, and determine a first distance between the target user node and at least one other user node in the friend relationship subnet that is not connected to the target user node, where each user node in the friend relationship subnet corresponds to a user account, a connection between user nodes represents that two user accounts corresponding to two user nodes have a friend relationship, and the first distance is used to represent a friend association degree between users represented by the user node;
a second determining unit 702, configured to determine, according to an organization relationship subnet, a second distance between the target user node and the at least one other user node, where each organization node in the organization relationship subnet corresponds to an organization respectively, a connection between the organization nodes represents that two organizations corresponding to two organization nodes have a hierarchical relationship, the connection between the user node and the organization node represents that a user represented by the user node belongs to the organization corresponding to the organization node, and the second distance is used to represent an organization association degree between users represented by the user node;
a third determining unit 703, configured to determine, according to at least the first distance and the second distance, a recommended user;
a sending unit 704, configured to send the user information of the recommended user to the terminal, so that the terminal displays the user information through the enterprise social application.
In the technical scheme, two information networks, namely a friend relation subnet and an organization relation subnet, are utilized. The friend relationship existing between the users is recorded in the friend relationship subnet, and the friend association degree between the two users, namely the probability of the two users contacting through the friend relationship subnet, can be determined through the friend relationship subnet. In addition, the organizational structure to which the existing user belongs and the hierarchical relationship between the organizational structures are recorded in the organizational relationship subnet, and the organizational association degree between two users, that is, the probability of the two users contacting through the organizational relationship subnet can be determined through the organizational relationship subnet. Therefore, the association degrees of the users in two dimensions of friend relationship and organization relationship can be integrated, and the recommended users suitable for being recommended to the target users can be determined. Because the relevancy of the user organization relationship is considered, the method is more suitable for user recommendation in enterprise social software, and the recommendation precision is higher. By feeding back the user information of the recommended user to the target user, the time required by the target user to add the friend can be shortened to a certain extent, user operation is reduced, the probability of wrong requests is reduced, and user experience is improved.
Optionally, the first determining unit 701 is further configured to: and for a first other user node in the at least one other user node, determining a first distance between the target user node and the first other user node according to the shortest path length between the first other user node and the target user node, the number of the shortest paths, and the number of user nodes connected with the first other user node in the friend relationship subnet, wherein the first other user node is a user node in the friend relationship subnet, the number of the user nodes connected with the first other user node is not zero.
Optionally, the first determining unit 701 is further configured to: and for a second other user node in the at least one other user node, determining a first distance between the target user node and the second other user node according to the determined maximum first distance between the target user node and the first other user node, wherein the first distance between the target user node and the second other user node is greater than the maximum first distance between the target user node and the first other user node.
Optionally, the second determining unit 702 is further configured to: and for each other user node, determining a second distance between the target user node and the other user node according to the organization layer level difference between the organization node to which the other user node belongs and the organization node to which the target user node belongs and the shortest path length between the organization node to which the other user node belongs and the organization node to which the target user node belongs in the organization relation subnet.
Optionally, the apparatus 700 may further include:
a first obtaining unit 705, configured to obtain, according to a user account currently logged in by the enterprise social application program, personal feature information of a target user represented by the target user node;
a second obtaining unit 706, configured to obtain, according to the user account corresponding to each of the other user nodes, personal feature information of the other user represented by each of the other user nodes;
a fourth determining unit 707, configured to determine, for each of the other users, a similarity parameter between the other user and the target user according to the personal characteristic information of the other user and the personal characteristic information of the target user;
the third determining unit 703 is further configured to: and determining a recommended user according to the first distance, the second distance and the similarity parameter.
Optionally, the personal characteristic information includes multi-dimensional characteristic information; the fourth determining unit 707 is further configured to: determining similarity parameters between the other users and the target user in the dimension according to the feature information of the other users and the target user in the same dimension; and determining the similarity parameter between the other user and the target user according to the similarity parameter between the other user and the target user in each dimension.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 8 is a block diagram illustrating an electronic device 800 in accordance with an example embodiment. For example, the electronic device 800 may be provided as a server. Referring to fig. 8, an electronic device 800 includes a processor 822, which may be one or more in number, and a memory 832 for storing computer programs executable by the processor 822. The computer programs stored in memory 832 may include one or more modules that each correspond to a set of instructions. Further, the processor 822 may be configured to execute the computer program to perform the user recommendation method described above.
Additionally, the electronic device 800 may also include a power component 826 and a communication component 850, the power component 826 may be configured to perform power management of the electronic device 800, and the communication component 850 may be configured to enable communication, e.g., wired or wireless communication, of the electronic device 800. The electronic device 800 may also include input/output (I/O) interfaces 858. The electronic device 800 may operate based on an operating system stored in the memory 832, such as Windows Server, Mac OSXTM, UnixTM, LinuxTM, and the like.
In another exemplary embodiment, there is also provided a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the user recommendation method described above. For example, the computer readable storage medium may be the memory 832 including program instructions executable by the processor 822 of the electronic device 800 to perform the user recommendation method described above.
In another exemplary embodiment, a computer program product is also provided, which comprises a computer program executable by a programmable apparatus, the computer program having code portions for performing the above-mentioned user recommendation method when executed by the programmable apparatus.
The preferred embodiments of the present disclosure are described in detail with reference to the accompanying drawings, however, the present disclosure is not limited to the specific details of the above embodiments, and various simple modifications may be made to the technical solution of the present disclosure within the technical idea of the present disclosure, and these simple modifications all belong to the protection scope of the present disclosure.
It should be noted that the various features described in the above embodiments may be combined in any suitable manner without departing from the scope of the invention. In order to avoid unnecessary repetition, various possible combinations will not be separately described in this disclosure.
In addition, any combination of various embodiments of the present disclosure may be made, and the same should be considered as the disclosure of the present disclosure, as long as it does not depart from the spirit of the present disclosure.
Claims (10)
1. A user recommendation method, comprising:
determining a target user node corresponding to a user account in a friend relation subnet according to the user account currently logged in by an enterprise social application program running on a terminal, and determining a first distance between the target user node and at least one other user node which is not connected with the target user node in the friend relation subnet, wherein each user node in the friend relation subnet corresponds to a user account respectively, the connection between the user nodes represents that two user accounts corresponding to the two user nodes have friend relation, and the first distance is used for representing the friend association degree between users represented by the user nodes;
determining a second distance between the target user node and the at least one other user node according to an organization relation subnet, wherein each organization node in the organization relation subnet corresponds to an organization mechanism, the connection between the organization nodes represents that the two organization mechanisms corresponding to the two organization nodes have a superior-inferior relation, the connection between the user node and the organization nodes represents that the user represented by the user node belongs to the organization mechanism corresponding to the organization node, and the second distance is used for representing the organization association degree between the users represented by the user node;
determining a recommended user according to at least the first distance and the second distance;
and sending the user information of the recommended user to the terminal so that the terminal displays the user information through the enterprise social application program.
2. The method of claim 1, wherein determining a first distance between the target user node and the at least one other user node comprises:
and for a first other user node in the at least one other user node, determining a first distance between the target user node and the first other user node according to the shortest path length between the first other user node and the target user node, the number of the shortest paths, and the number of user nodes connected with the first other user node in the friend relationship subnet, wherein the first other user node is a user node in the friend relationship subnet, the number of the user nodes connected with the first other user node is not zero.
3. The method of claim 2, wherein determining a first distance between the target user node and the at least one other user node further comprises:
and for a second other user node in the at least one other user node, determining a first distance between the target user node and the second other user node according to the determined maximum first distance between the target user node and the first other user node, wherein the first distance between the target user node and the second other user node is greater than the maximum first distance between the target user node and the first other user node.
4. The method of claim 2, wherein the first distance between the target user node and the first other user node is determined by the following formula:
wherein SF (i, j) represents a first distance between user node i and user node j; d (i, j) represents the shortest path length between the user node i and the user node j; n (i, j) represents the number of the shortest paths; and d (j) represents the number of user nodes connected to user node j.
5. The method of claim 1, wherein the determining a second distance between the target user node and the at least one other user node comprises:
and for each other user node, determining a second distance between the target user node and the other user node according to the organization layer level difference between the organization node to which the other user node belongs and the organization node to which the target user node belongs and the shortest path length between the organization node to which the other user node belongs and the organization node to which the target user node belongs in the organization relation subnet.
6. The method of claim 5, wherein the second distance between the target user node and the other user node is determined by the following formula:
SO(i,j)=o(i,j)+do(i,j)
wherein SO (i, j) represents a second distance between user node i and user node j; o (i, j) represents the organizational layer level difference between the organizational node to which the user node j belongs and the organizational node to which the user node i belongs; and do (i, j) represents the shortest path length between the organization node to which the user node j belongs and the organization node to which the user node i belongs.
7. The method according to any one of claims 1-6, further comprising:
acquiring personal characteristic information of a target user represented by the target user node according to a user account currently logged in by the enterprise social application program;
acquiring personal characteristic information of other users represented by each other user node according to the user account corresponding to each other user node;
for each other user, determining a similarity parameter between the other user and the target user according to the personal characteristic information of the other user and the personal characteristic information of the target user;
the determining a recommended user according to at least the first distance and the second distance includes:
and determining a recommended user according to the first distance, the second distance and the similarity parameter.
8. A user recommendation device, comprising:
the system comprises a first determining unit, a second determining unit and a third determining unit, wherein the first determining unit is used for determining a target user node corresponding to a user account in a friend relation subnet according to the user account currently logged in by an enterprise social application program running on a terminal, and determining a first distance between the target user node and at least one other user node which is not connected with the target user node in the friend relation subnet, each user node in the friend relation subnet corresponds to one user account respectively, the connection between the user nodes represents that two user accounts corresponding to the two user nodes have friend relation, and the first distance is used for representing the friend association degree between users represented by the user nodes;
a second determining unit, configured to determine, according to an organization relationship subnet, a second distance between the target user node and the at least one other user node, where each organization node in the organization relationship subnet corresponds to an organization respectively, a connection between the organization nodes represents that two organizations corresponding to two organization nodes have a hierarchical relationship, the connection between the user node and the organization node represents that a user represented by the user node belongs to the organization corresponding to the organization node, and the second distance is used to represent an organization association degree between users represented by the user node;
a third determining unit, configured to determine, according to at least the first distance and the second distance, a recommended user;
and the sending unit is used for sending the user information of the recommended user to the terminal so that the terminal can display the user information through the enterprise social application program.
9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.
10. An electronic device, comprising:
a memory having a computer program stored thereon;
a processor for executing the computer program in the memory to carry out the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911203811.0A CN111079027B (en) | 2019-11-29 | 2019-11-29 | User recommendation method and device, readable storage medium and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911203811.0A CN111079027B (en) | 2019-11-29 | 2019-11-29 | User recommendation method and device, readable storage medium and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111079027A true CN111079027A (en) | 2020-04-28 |
CN111079027B CN111079027B (en) | 2023-03-31 |
Family
ID=70312312
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911203811.0A Active CN111079027B (en) | 2019-11-29 | 2019-11-29 | User recommendation method and device, readable storage medium and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111079027B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113742603A (en) * | 2021-04-19 | 2021-12-03 | 重庆邮电大学 | Object recommendation method, device and system and electronic equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104518948A (en) * | 2013-09-27 | 2015-04-15 | 北京新媒传信科技有限公司 | Method and device for friend recommendation |
CN109544396A (en) * | 2019-01-10 | 2019-03-29 | 腾讯科技(深圳)有限公司 | Account recommended method, device, server, terminal and storage medium |
CN110457573A (en) * | 2019-07-04 | 2019-11-15 | 平安科技(深圳)有限公司 | Products Show method, apparatus, computer equipment and storage medium |
-
2019
- 2019-11-29 CN CN201911203811.0A patent/CN111079027B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104518948A (en) * | 2013-09-27 | 2015-04-15 | 北京新媒传信科技有限公司 | Method and device for friend recommendation |
CN109544396A (en) * | 2019-01-10 | 2019-03-29 | 腾讯科技(深圳)有限公司 | Account recommended method, device, server, terminal and storage medium |
CN110457573A (en) * | 2019-07-04 | 2019-11-15 | 平安科技(深圳)有限公司 | Products Show method, apparatus, computer equipment and storage medium |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113742603A (en) * | 2021-04-19 | 2021-12-03 | 重庆邮电大学 | Object recommendation method, device and system and electronic equipment |
CN113742603B (en) * | 2021-04-19 | 2023-09-05 | 重庆邮电大学 | Object recommendation method, device and system and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN111079027B (en) | 2023-03-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11968105B2 (en) | Systems and methods for social graph data analytics to determine connectivity within a community | |
US8775570B2 (en) | Geographic recommendation online search system | |
US10277692B2 (en) | Method and apparatus for identifying common interest between social network users | |
WO2019024496A1 (en) | Enterprise recommendation method and application server | |
US9396275B2 (en) | Geographically partitioned online search system | |
CN108280115B (en) | Method and device for identifying user relationship | |
US9299059B1 (en) | Generating a summary of social media content | |
CN107690637B (en) | Connecting semantically related data using large-table corpus | |
US20130072233A1 (en) | Geographically partitioned online content services | |
JP2008539528A (en) | Scoring user compatibility in social networks | |
CN104572889A (en) | Method, device and system for recommending search terms | |
US8977642B2 (en) | Recommending relevant and personalized content accessing keywords for users of a text messaging service based global textsite platform | |
US8972278B2 (en) | Recommending print locations | |
US8620946B2 (en) | Storage and searching of temporal entity information | |
US20140289140A1 (en) | Systems and methods of opportunistically identifying networking prospects | |
Xiong et al. | Reward-based spatial crowdsourcing with differential privacy preservation | |
US10296509B2 (en) | Method, system and apparatus for managing contact data | |
US9158790B2 (en) | Server, dictionary creation method, dictionary creation program, and computer-readable recording medium recording the program | |
CN111913960A (en) | Form processing method and related equipment | |
CN105453081A (en) | Answering people-related questions | |
CN103678624A (en) | Searching method, searching server, and searching request executing method and terminal | |
US9965812B2 (en) | Generating a supplemental description of an entity | |
CN111079027B (en) | User recommendation method and device, readable storage medium and electronic equipment | |
US20170004531A1 (en) | Advertisement selection using information retrieval systems | |
US10298704B2 (en) | Determining geolocation of IP addresses using user transitions over time |
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 |