CN113239285A - Processing method, device and processing equipment for social network influence - Google Patents
Processing method, device and processing equipment for social network influence Download PDFInfo
- Publication number
- CN113239285A CN113239285A CN202110414260.3A CN202110414260A CN113239285A CN 113239285 A CN113239285 A CN 113239285A CN 202110414260 A CN202110414260 A CN 202110414260A CN 113239285 A CN113239285 A CN 113239285A
- Authority
- CN
- China
- Prior art keywords
- check
- user node
- place
- influence
- record
- 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.)
- Withdrawn
Links
- 238000012545 processing Methods 0.000 title claims abstract description 45
- 238000003672 processing method Methods 0.000 title abstract description 16
- 230000006399 behavior Effects 0.000 claims description 45
- 238000000034 method Methods 0.000 claims description 44
- 238000001914 filtration Methods 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 16
- 238000004590 computer program Methods 0.000 claims description 13
- 238000010276 construction Methods 0.000 claims description 4
- 230000003542 behavioural effect Effects 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 description 9
- 230000008030 elimination Effects 0.000 description 7
- 238000003379 elimination reaction Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000126 substance 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- 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)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- Mathematical Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Evolutionary Biology (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)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a processing method, a processing device and processing equipment for social network influence, which are used for separating the social network influence among users from complex confusion factors from the perspective of user positions to obtain more accurate social network influence among users.
Description
Technical Field
The application relates to the field of social networks, in particular to a method, a device and equipment for processing social network influence.
Background
With the continuous development of internet technology and various large social networks, various data in the social networks are increasingly huge, and the data can obviously serve as data support in the process of managing the social networks.
The influence among users in the social network plays an important role in the research of the social network, and the measurement of the influence of the social network has wide application requirements in social public opinion monitoring, personalized marketing, accurate marketing and other aspects, so the research on the measurement of the influence of the social network among users is highly valued.
In the existing research process of related technologies, the inventor finds that although a large number of related research results exist at home and abroad at present, in these researches, the influence strength of users based on behaviors is mainly determined by the occurrence of the same behaviors between the users, and in practical application, the determination method is unstable, or the existing determination method of social network influence between the users is not high in precision.
Disclosure of Invention
The application provides a processing method, a processing device and processing equipment for social network influence, which are used for separating the social network influence among users from complex confusion factors from the perspective of user positions to obtain more accurate social network influence among users.
In a first aspect, the present application provides a method for processing social network influence, where the method includes:
constructing initial social network data of a target social network, wherein the initial social network data is configured by an undirected graph G (V, E, C), the V is a user node contained in the target social network data, the E is an edge set in the undirected graph G (V, E, C), and the C is a place check-in record of the user node;
location check-in record C for traversing user node uuAnd a place check-in record C for user node vvDetermining that the user node u and the user node v have the same check-in placeAnd based on the place of check-inDetermining that the two have a connecting edge;
check-in record at location CuAnd place check-in record CvIn, filtering non-first configuration check-in placesThe location check-in record;
calculating sign-in locationLocation popularity ofAnd according to location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminated
Under influence of forceOn the basis, an e exponential time decay model is introduced, and the influence of the user node u on the user node v is redistributed by combining a softmax function to obtain the influence
Sign-in record according to location CuAnd place check-in record CvDetermining the similarity S of user behaviors between the user node u and the user node vu,v;
Will influence the forceSimilarity to user behavior Su,vIs output as the influence of the user node u on the user node v.
With reference to the first aspect of the present application, in a first possible implementation manner of the first aspect of the present application, the check-in location is based onDetermining that the two have the connecting edge, including:
if the user node u is at the check-in placeTime of attendanceLater than or equal to the place of check-in of user node vTime of attendanceDetermining that no connecting edge exists between the two;
if the user node u is at the check-in placeTime of attendanceEarlier than user node v is at check-in placeTime of attendanceIt is determined that there is a continuous edge between the two.
With reference to the first aspect of the present application, in the second possible aspect of the present applicationIn an implementation, the location popularity of check-in place c is calculatedThe method comprises the following steps:
check-in place based on filtering non-first configurationPost-location check-in record CuAnd filtering non-first-configuration check-in placesPost-location check-in record CvDetermining a check-in locationThe ratio of the number of sign-in times of all users in the current time period to the total number of sign-in times is used as the position popularity
Delta is the time period threshold value and is,representing a check-in place for user node vTime of attendanceCheck-in time of user node uEarly, | | represents the size of the set.
With reference to the first aspect of the present application, in a third possible implementation manner of the first aspect of the present application,based on location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminatedThe method comprises the following steps:
the preset adjusting coefficient f is compared with the position popularityAs the user node v is at the check-in placeInfluence received from friend user node after social influence factor is eliminated
With reference to the first aspect of the present application, in a fourth possible implementation manner of the first aspect of the present application, the check-in record C is recorded according to a placeuAnd place check-in record CvDetermining the similarity S of user behaviors between the user node u and the user node vu,vThe method comprises the following steps:
sign-in record according to location CuAnd place check-in record CvDetermining behavior similarity S by combining improved Jacard similarity coefficientu,v,
Check-in place for user node u in arrivalPrevious location check-in record, |v(cv) Check-in place for user node v in arrivalPrevious location check-in records.
With reference to the first aspect of the present application, in a fifth possible implementation manner of the first aspect of the present application, an expression of an e-exponential time decay model is:
p (v | u) is the current influence of the user node u on the user node v, σ is the attenuation coefficient,between 0 and 1, P' (v | u) represents the influence of user node u on user node v before the time decay.
With reference to the first aspect of the present application, in a sixth possible implementation manner of the first aspect of the present application, the influence isThe expression of (a) is:
in a second aspect, the present application provides a processing apparatus for social network influence, the apparatus including:
the system comprises a construction unit, a log-in unit and a log-in unit, wherein the construction unit is used for constructing initial social network data of a target social network, the initial social network data is configured by an undirected graph G (V, E, C), the V is a user node contained in the target social network data, the E is an edge set in the undirected graph G (V, E, C), and the C is a place sign-in record of the user node;
a determination unit for traversing the user node uLocation check-in record CuAnd a place check-in record C for user node vvDetermining that the user node u and the user node v have the same check-in placeAnd based on the place of check-inDetermining that the two have a connecting edge;
a filtering unit for checking in records C at the placeuAnd place check-in record CvIn, filtering non-first configuration check-in placesThe location check-in record;
a determination unit for calculating the check-in placeLocation popularity ofAnd according to location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminated
A redistribution unit for influencing forceOn the basis, an e exponential time decay model is introduced, and the influence of the user node u on the user node v is redistributed by combining a softmax function to obtain the influence
A determination unit for checking in record C according to the locationuAnd place check-in record CvDetermining the similarity S of user behaviors between the user node u and the user node vu,v;
An output unit for outputting the influenceSimilarity to user behavior Su,vIs output as the influence of the user node u on the user node v.
With reference to the second aspect of the present application, in a first possible implementation manner of the second aspect of the present application, the determining unit is specifically configured to:
if the user node u is at the check-in placeTime of attendanceLater than or equal to the place of check-in of user node vTime of attendanceDetermining that no connecting edge exists between the two;
if the user node u is at the check-in placeTime of attendanceEarlier than user node v is at check-in placeTime of attendanceIt is determined that there is a continuous edge between the two.
With reference to the second aspect of the present application, in a second possible implementation manner of the second aspect of the present application, the determining unit is specifically configured to:
check-in place based on filtering non-first configurationPost-location check-in record CuAnd filtering non-first-configuration check-in placesPost-location check-in record CvDetermining a check-in locationThe ratio of the number of sign-in times of all users in the current time period to the total number of sign-in times is used as the position popularity
Delta is the time period threshold value and is,representing a check-in place for user node vTime of attendanceCheck-in time of user node uEarly, | | represents the size of the set.
With reference to the second aspect of the present application, in a third possible implementation manner of the second aspect of the present application, the determining unit is specifically configured to:
the preset adjusting coefficient f is compared with the position popularityAs the user node v is at the check-in placeInfluence received from friend user node after social influence factor is eliminated
With reference to the second aspect of the present application, in a fourth possible implementation manner of the second aspect of the present application, the determining unit is specifically configured to:
sign-in record according to location CuAnd place check-in record CvDetermining behavior similarity S by combining improved Jacard similarity coefficientu,v,
Check-in place for user node u in arrivalPrevious location check-in record, |v(cv) Check-in place for user node v in arrivalPrevious location check-in records.
With reference to the second aspect of the present application, in a fifth possible implementation manner of the second aspect of the present application, an expression of the e-exponential time decay model is:
p (v | u) is the current influence of the user node u on the user node v, σ is the attenuation coefficient,between 0 and 1, P' (v | u) represents the influence of user node u on user node v before the time decay.
With reference to the second aspect of the present application, in a sixth possible implementation manner of the second aspect of the present application, the influence isThe expression of (a) is:
in a third aspect, the present application provides a processing device, including a processor and a memory, where the memory stores a computer program, and the processor executes the method provided in the first aspect of the present application or any one of the possible implementation manners of the first aspect of the present application when calling the computer program in the memory.
In a fourth aspect, the present application provides a computer-readable storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the method provided in the first aspect of the present application or any one of the possible implementations of the first aspect of the present application.
From the above, the present application has the following advantageous effects:
aiming at capturing the influence of the social network among users, the method firstly constructs initial social network data of a target social network in an undirected graph mode, configures a social network mode based on position access in the undirected graph, and determines that a user node u and a user node v have the same check-in place based on a place check-in recordAnd based on the check-in placeDetermining that the two have the connecting edge, namely after determining that the social network influence exists, filtering the non-first-configuration check-in place in the place check-in recordTo exclude interference factors of the user's personal habits, and then calculate the check-in locationLocation popularity ofAnd according to location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminatedInterference factors of social masses are eliminated, an e-exponential time decay model is continuously introduced, influence of the user node u on the user node v is redistributed by combining a soffmax function, and influence is obtainedSo as to eliminate the interference factor of the user' S own social circle and simultaneously determine the user behavior similarity S between the user node u and the user node v according to the place sign-in recordu,vAt this time, the force will be influencedAnd the userDegree of behavioral similarity Su,vThe ratio is used as the influence of the user node u on the user node v to be output, so that from the perspective of the user position, the influence of the social network among the users is separated from the complicated confusion factors through a multi-layer interference factor elimination mechanism, the more accurate influence of the social network among the users is obtained, and powerful data support is provided for the management and control of the target social network.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart illustrating a method for processing social network influence according to the present application;
FIG. 2 is a schematic diagram of a social network influencing power processing apparatus according to the present application;
FIG. 3 is a schematic diagram of a processing apparatus according to 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.
The terms "first," "second," and the like in the description and in the claims of the present application and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Moreover, the terms "comprises," "comprising," and any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not expressly listed or inherent to such process, method, article, or apparatus. The naming or numbering of the steps appearing in the present application does not mean that the steps in the method flow have to be executed in the chronological/logical order indicated by the naming or numbering, and the named or numbered process steps may be executed in a modified order depending on the technical purpose to be achieved, as long as the same or similar technical effects are achieved.
The division of the modules presented in this application is a logical division, and in practical applications, there may be another division, for example, multiple modules may be combined or integrated into another system, or some features may be omitted, or not executed, and in addition, the shown or discussed coupling or direct coupling or communication connection between each other may be through some interfaces, and the indirect coupling or communication connection between the modules may be in an electrical or other similar form, which is not limited in this application. The modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed in a plurality of circuit modules, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present disclosure.
Before introducing the social network influence processing method provided by the present application, the background related to the present application is first introduced.
The processing method and device for the social network influence and the computer-readable storage medium can be applied to processing equipment and are used for separating the social network influence among the users from complex confusion factors from the perspective of the positions of the users to obtain more accurate social network influence among the users.
In the processing method for social network influence, an execution subject may be a social network influence device, or a server, a physical host, or a User Equipment (UE) type processing device that integrates the social network influence device. The processing device of the social network influence may be implemented in a hardware or software manner, the UE may specifically be a terminal device such as a smart phone, a tablet computer, a notebook computer, a desktop computer, or a Personal Digital Assistant (PDA), and the processing device may be set in a device cluster manner.
Next, a method for processing social network influence provided by the present application is described.
First, referring to fig. 1, fig. 1 shows a schematic flow chart of a processing method of social network influence in the present application, and the processing method of social network influence in the present application may specifically include the following steps:
step S101, constructing initial social network data of a target social network, wherein the initial social network data is configured by an undirected graph G (V, E, C), the V is a user node contained in the target social network data, the E is an edge set in the undirected graph G (V, E, C), and the C is a place check-in record of the user node;
step S102, traversing the place check-in record C of the user node uuAnd a place check-in record C for user node vvDetermining that the user node u and the user node v have the same check-in placeAnd based on the place of check-inDetermining that the two have a connecting edge;
step S103, registering the record C in the placeuAnd place check-in record CvIn, filtering non-first configuration check-in placesThe location check-in record;
step S104, calculating the check-in placeLocation popularity ofAnd according to location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminated
Step S105, in the influenceOn the basis, an e exponential time decay model is introduced, and the influence of the user node u on the user node v is redistributed by combining a softmax function to obtain the influence
Step S106, according to the place check-in record CuAnd place check-in record CvDetermining the similarity S of user behaviors between the user node u and the user node vu,v;
Step S107, influence forceSimilarity to user behavior Su,vIs output as the influence of the user node u on the user node v.
As can be seen from the embodiment shown in fig. 1, for capturing the social network influence among users, the application first constructs initial social network data of a target social network in the form of an undirected graph in which a social network form based on location access is configured,determining that user node u and user node v have the same check-in place based on place check-in recordsAnd based on the check-in placeDetermining that the two have the connecting edge, namely after determining that the social network influence exists, filtering the non-first-configuration check-in place in the place check-in recordTo exclude interference factors of the user's personal habits, and then calculate the check-in locationLocation popularity ofAnd according to location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminatedInterference factors of social masses are eliminated, an e-exponential time decay model is continuously introduced, influence of the user node u on the user node v is redistributed by combining a softmax function, and influence is obtainedSo as to eliminate the interference factor of the user' S own social circle and simultaneously determine the user behavior similarity S between the user node u and the user node v according to the place sign-in recordu,vAt this time, the force will be influencedSimilarity to user behavior Su,vThe ratio is used as the influence of the user node u on the user node v to be output, so that from the perspective of the user position, the influence of the social network among the users is separated from the complicated confusion factors through a multi-layer interference factor elimination mechanism, the more accurate influence of the social network among the users is obtained, and powerful data support is provided for the management and control of the target social network.
The steps of the embodiment shown in fig. 1 and the possible implementation manner thereof in practical applications are described in detail below.
In the application, the processing device applying the processing method of the social network influence can be specifically a device in a target social network, so that for a manager of the target social network, more accurate social network influence among users can be captured from inside, and as a data support, management work such as more accurate data research, public opinion management and control on the target social network is realized.
Of course, the processing device may also be specifically a device outside the target social network, and capture the influence of the social network among the users outside the network, so as to provide data support for the management work of the target social network.
It is understood that capturing the social network influence among these users is, in practical applications, performed after meeting the privacy requirements of the users or obtaining the confirmation of the users.
The target social network may be any form of social network, such as a social network of a microblog, a post bar, and the like, and may be adjusted according to a defined social network form, where the target social network may be worth a social network constructed by a single social network product, and may also be a social network constructed by different social network products, which is not limited herein.
In the existing mechanism for judging the influence of the social network between users, the influence between two users is mainly judged by the occurrence of the same action between the users, and in many cases, it is found that the influence does not necessarily exist between the users acting at different times and in the same place, and even the situation that the majority of users are not related can occur.
Therefore, the method and the device start from the user position, a multi-layer interference factor elimination mechanism is constructed, so that the social network influence among the users is separated from the complex confusion factors, and the precision of the influence is obviously improved.
Firstly, for a target social network, constructing an undirected graph G (V, E, C) as initial social network data, wherein in the undirected graph G (V, E, C), V is a user node contained in the target social network data, E is an edge set in the undirected graph G (V, E, C), and C is a place check-in record of the user node.
Take user node u as an example, CuThe place check-in record of the user node u may include check-in records of the user node u at different places and different times, and specifically, may be represented asRepresents a check-in record for any user u, where i represents the ith check-in record for the user node,representing the ith check-in place for user node u,indicating that user node u is at a check-in placeThe check-in time of.
The place check-in record can be understood as a reporting record of the location of the place, and can be directly obtained through real-time location data or indirectly obtained through landmarks such as restaurants and the like.
It can be understood that, at this time, the undirected graph G (V, E, C) is in the initial state, and in expectation, in the undirected graph G (V, E, C) in the completed state, if there is social network influence between the user nodes, there is a situation of connecting edges between the user nodes.
Thus, the place check-in record C of the user node u is traversed by the applicationuAnd a place check-in record C for user node vvTo confirm that user node u and user node v have the same check-in placeAnd based on the check-in placeAnd determining that the two have connecting edges, and then capturing subsequent influence.
It can be understood that if the same check-in place exists in the check-in records of the two user nodesThe social network influence between the two can be directly identified.
In further studies, it is also believed that this determination may be optimized when the same check-in location existsIn case of (2), the check-in time is continued to be combinedAnd the accuracy of the judgment result of the social network influence is improved.
For example, if user node u is at a check-in locationTime of attendanceLater than or equal to the place of check-in of user node vTime of attendanceDetermining that no connecting edge exists between the two;
if the user node u is at the check-in placeTime of attendanceEarlier than user node v is at check-in placeTime of attendanceIt is determined that there is a continuous edge between the two.
When the influence of the user node u on the social network of the user node v is judged, the condition that the user node u needs to arrive at the check-in place earlier than the user node v can be limitedIn this case, user node u may be at the check-in locationGenerating social network influence on the user node v, namely:
if the two user nodes are signed in at the same place and the sign-in time of the user node u is earlier than that of the user node v, a connecting edge exists, and the user node u has an influence on the user node v based on a certain behavior.
And if the two user nodes check in at the same place, but the check-in time of the user node u is later than the check-in time of the user node v, or the two user nodes do not check in at the same place, no connecting edge exists, and the influence is 0.
After the existence of the connecting edge and the influence are determined, the specific measurement of the influence can be carried out by combining the multilayer interference factor elimination mechanism constructed by the application.
Specifically, the interference factor eliminating mechanism comprises three layers, namely, the interference factor of personal habits of the user, the interference factor of social masses and the interference factor of social circles of the user are respectively eliminated.
In the interference factor elimination mechanism of the personal habits of the user, the application considers that the behavior habits of the user can also cause the place check-in behavior of the user, for example, a certain user is in work in a unit, and the user can frequently perform the place check-in behavior at the place due to the repeated characteristic.
The filtering continues to take the user node u as an example, and the check-in record is recorded according to the locationTime of attendanceExtracting that the user node u is in the same check-in placeThe earliest check-in record is recorded, and then the check-in place of the user node u is deletedAnd the time of sign-in isAnd then, establishing a check-in record set of the initial place check-in behavior of the user node u according to the check-in records, wherein the check-in record set can be expressed as follows:
where n represents the total number of check-in records for user node u.
In the interference factor elimination mechanism of the social public, the social public is different from the influence of the social circle (all first-order friends) of the user, the social public is indirect friends of the user, for example, a new web restaurant at a certain place is popular with the public, is widely commented and shared on the homepage of a website such as 'popular comment' and the like, and is recommended to the user A by the website, and if the user A also checks the card at the restaurant, the user A is indicated to be influenced by the social public.
Interference factors of social masses, and popularity of specific available positionsThe position popularity H refers to the degree that the position is pursued by the public in the current time period, and the influence brought by the position popularity H is in an inverse relation with the influence factor of the public.
Exemplary, location popularityBased on the ratio of the number of check-in times of all users in the current time period to the total number of check-in times, which can be for check-in place c, the specific expression can be:
Delta is the time period threshold value and is,representing a check-in place for user node vTime of attendanceCheck-in time of user node uEarly, | | represents the size of the set.
The adopted place check-in record can be specifically the above-mentioned filtering non-first configuration check-in placePost-location check-in record CuAnd filtering non-first-configuration check-in placesThe post place check-in record Cv。
While determining location popularityLater, the popularity of the position can be determinedDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminated
As already mentioned above, location popularityThe brought influence and the public influence factor present an inverse relationship, and in the concrete quantization process, an adjusting coefficient can be introduced to more suitably quantize the brought influence,
the preset adjusting coefficient f is compared with the position popularityAs the user node v is at the check-in placeInfluence received from friend user node after social influence factor is eliminatedCan be expressed as:
wherein, f is an adjusting parameter,indicating that the same check-in place occurred at user node vAnd checking in the friend user which is in the active state before.
In the interference factor elimination mechanism of the user's own social circle, the influence received from the friend user node is specifically the influence received after the social influence factor is eliminated from the determined user node v at the check-in place cOn the basis, an e exponential time decay model is continuously introduced, influence of the user node u on the user node v is redistributed by combining a softmax function, and the obtained influence
According to the application, when a location card punching behavior starts to be popular in a friend circle (direct friends) of A and the behavior also occurs in the last A, the A is considered to be influenced by a social circle which is composed of a plurality of first-order friends of a target user node v, firstly, according to the theory that the influence of the behavior among users is attenuated along with the extension of time intervals, an influence time attenuation model is established, and can be expressed as:
p (v | u) is the current influence of the user node u on the user node v, σ is the attenuation coefficient,between 0 and 1, P' (v | u) represents the influence of user node u on user node v before the time decay.
Under an influence time attenuation model, for different influence values of different user nodes in the social circle on the target user node due to different activation time intervals, redistributing the influence of u on v according to the softmax function, and obtaining the influence of the user node u on the user node v after eliminating the interference factors of the social massesCan be expressed as:
besides, in addition to considering the three interference factors mentioned above, the present application also considers the influence of the similarity of the user on the phase, and the present application also considers that the influence of the behavior similarity is in an inverse relationship with itself.
The method and the device determine the behavior similarity S by adopting an improved Jacard similarity coefficient on the basis of the place check-in record in consideration of the fact that the behavior similarity changes along with time, the position behavior record of the user is less in the initial stage, the position behavior record of the user is more in the later stage, the calculation results of the behavior similarity before and after the calculation results are different, and the behavior similarity changes along with timeu,vSpecifically, it can be expressed as:
Wherein the content of the first and second substances,a check-in record for user node u before reaching check-in location c,lv(cv) A check-in record for a place before user node v arrives at check-in place c,
by combining the consideration of the above multi-layer interference factors, the interference factors of the personal habits of the users, the interference factors of the social masses, the interference factors of the social circles of the users and the interference factors of the behavior similarity are eliminated, and the influence can be realizedSimilarity to user behavior Su,vAs the influence of the user node u on the user node v, the ratio of (d) to (d) can be expressed as:
the above is the introduction of the processing method for social network influence provided by the present application, and in order to better implement the processing method for social network influence provided by the present application, the present application also provides a processing device for social network influence.
Referring to fig. 2, fig. 2 is a schematic structural diagram of a processing device for social network influence according to the present application, in which the processing device 200 for social network influence specifically includes the following structure:
the constructing unit 201 is configured to construct initial social network data of a target social network, where the initial social network data is configured by an undirected graph G (V, E, C), V is a user node included in the target social network data, E is an edge set in the undirected graph G (V, E, C), and C is a place check-in record of the user node;
a determining unit 202 for traversing the location check-in record C of the user node uuAnd a place check-in record C for user node vvDetermining that the user node u and the user node v have the same check-in placeAnd based on the place of check-inDetermining that the two have a connecting edge;
a filtering unit 203 for checking in records C at a placeuAnd place check-in record CvIn, filtering non-first configuration check-in placesThe location check-in record;
a determination unit 202 for calculating a check-in placeLocation popularity ofAnd according to location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminated
A redistribution unit 204 for influencing forceOn the basis, an e exponential time decay model is introduced, and a softmax function is combined to carry out user node u to user nodeRedistributing the influence of the point v to obtain the influence
A determination unit 202 for checking in the record C according to the locationuAnd place check-in record CvDetermining the similarity S of user behaviors between the user node u and the user node vu,v;
An output unit 205 for outputting the influenceSimilarity to user behavior Su,vIs output as the influence of the user node u on the user node v.
In another exemplary implementation manner, the determining unit 202 is specifically configured to:
if the user node u is at the check-in placeTime of attendanceLater than or equal to the place of check-in of user node vTime of attendanceDetermining that no connecting edge exists between the two;
if the user node u is at the check-in placeTime of attendanceEarlier than user node v is at check-in placeIs signed inTimeIt is determined that there is a continuous edge between the two.
In another exemplary implementation manner, the determining unit 202 is specifically configured to:
check-in place based on filtering non-first configurationPost-location check-in record CuAnd filtering non-first-configuration check-in placesPost-location check-in record CvDetermining a check-in locationThe ratio of the number of sign-in times of all users in the current time period to the total number of sign-in times is used as the position popularity
Delta is the time period threshold value and is,representing a check-in place for user node vTime of attendanceCheck-in time of user node uEarly, | | represents the size of the set.
In another exemplary implementation manner, the determining unit 202 is specifically configured to:
the preset adjusting coefficient f is compared with the position popularityAs the user node v is at the check-in placeInfluence received from friend user node after social influence factor is eliminated
In another exemplary implementation manner, the determining unit 202 is specifically configured to:
sign-in record according to location CuAnd place check-in record CvDetermining behavior similarity S by combining improved Jacard similarity coefficientu,v,
Check-in place for user node u in arrivalPrevious location check-in record, |v(cv) Check-in place for user node v in arrivalPrevious location check-in records.
In yet another exemplary implementation, the expression of the e-exponential time decay model is:
p (v | u) is the current influence of the user node u on the user node v, σ is the attenuation coefficient,between 0 and 1, P' (v | u) represents the influence of user node u on user node v before the time decay.
the present application further provides a processing device, and referring to fig. 3, fig. 3 shows a schematic structural diagram of the processing device of the present application, specifically, the processing device of the present application may include a processor 301, a memory 302, and an input/output device 303, where the processor 301 is configured to implement, when executing a computer program stored in the memory 302, the steps of the processing method for social network influence in the corresponding embodiment of fig. 1; alternatively, the processor 301 is configured to implement the functions of the units in the corresponding embodiment of fig. 2 when executing the computer program stored in the memory 302, and the memory 302 is configured to store the computer program required by the processor 301 to execute the processing method for social network influence in the corresponding embodiment of fig. 1.
Illustratively, the computer program may be divided into one or more modules/units, which are stored in the memory 302 and executed by the processor 301 to accomplish the present application. One or more modules/units may be a series of computer program instruction segments capable of performing certain functions, the instruction segments being used to describe the execution of a computer program in a computer device.
The processing devices may include, but are not limited to, a processor 301, a memory 302, and an input-output device 303. It will be appreciated by those skilled in the art that the illustration is merely an example of a processing device and does not constitute a limitation of a processing device and may include more or less components than those illustrated, or combine certain components, or different components, for example, the processing device may also include a network access device, bus, etc. through which the processor 301, memory 302, input output device 303, and network access device, etc. are connected.
The Processor 301 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, the processor being the control center for the processing device and the various interfaces and lines connecting the various parts of the overall device.
The memory 302 may be used to store computer programs and/or modules, and the processor 301 implements various functions of the computer device by running or executing the computer programs and/or modules stored in the memory 302 and invoking data stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required for at least one function, and the like; the storage data area may store data created according to the use of the processing apparatus, and the like. In addition, the memory may include high speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), at least one magnetic disk storage device, a Flash memory device, or other volatile solid state storage device.
The processor 301, when executing the computer program stored in the memory 302, may specifically implement the following functions:
location check-in record C for traversing user node uuAnd a place check-in record C for user node vvDetermining that the user node u and the user node v have the same check-in placeAnd based on the place of check-inDetermining that the two have a connecting edge;
check-in record at location CuAnd place check-in record CvIn, filtering non-first configuration check-in placesThe location check-in record;
calculating sign-in locationLocation popularity ofAnd according to location popularityDetermining that user node v is at a check-in locationInfluence received from friend user node after social influence factor is eliminated
Under influence of forceOn the basis, an e exponential time decay model is introduced, and the influence of the user node u on the user node v is redistributed by combining a softmax function to obtain the influence
Sign-in record according to location CuAnd place check-in record CvDetermining the similarity S of user behaviors between the user node u and the user node vu,v;
Will influence the forceSimilarity to user behavior Su,vIs output as the influence of the user node u on the user node v.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the processing apparatus and the processing device for social network influence and the corresponding units thereof described above may refer to the description of the processing method for social network influence in the corresponding embodiment shown in fig. 1, and are not described herein again in detail.
It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor.
For this reason, the present application provides a computer-readable storage medium, where a plurality of instructions are stored, where the instructions can be loaded by a processor to execute steps in the method for processing social network influence in the embodiment corresponding to fig. 1 in the present application, and specific operations may refer to the description of the method for processing social network influence in the embodiment corresponding to fig. 1, which is not repeated herein.
Wherein the computer-readable storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.
Because the instructions stored in the computer-readable storage medium may execute the steps in the method for processing social network influence according to the embodiment of the present application corresponding to fig. 1, the beneficial effects that can be achieved by the method for processing social network influence according to the embodiment of the present application corresponding to fig. 1 can be achieved, for details, see the foregoing description, and are not repeated herein.
The social network influence processing method, the social network influence processing device, the social network influence processing apparatus, and the computer-readable storage medium provided by the present application are introduced in detail above, and a specific example is applied in the present application to explain the principles and embodiments of the present application, and the description of the above embodiment is only used to help understanding the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (10)
1. A method for processing social network influence, the method comprising:
constructing initial social network data of a target social network, wherein the initial social network data is configured by an undirected graph G (V, E, C), V is a user node contained in the target social network data, E is an edge set in the undirected graph G (V, E, C), and C is a place check-in record of the user node;
location check-in record C for traversing user node uuAnd a place check-in record C for user node vvDetermining that the user node u and the user node v have the same check-in placeAnd based on the check-in placeDetermining that the two have a connecting edge;
check-in record C at the placeuAnd the place check-in record CvIn filtering the check-in placeThe location check-in record;
calculating the check-in locationLocation popularity ofAnd according to the position popularityDetermining that the user node v is at the check-in placeInfluence received from friend user node after social influence factor is eliminated
Under the influence ofOn the basis, an e exponential time decay model is introduced, and the influence of the user node u on the user node v is redistributed by combining a softmax function to obtain the influence
According to the place check-in record CuAnd the place check-in record CvDetermining the similarity S of the user behaviors between the user node u and the user node vu,v;
2. The method of claim 1, wherein the checking-in based on the check-in locationDetermining that the two have the connecting edge, including:
if the user node u is at the check-in placeTime of attendanceLater than or equal to the check-in place of the user node vTime of attendanceDetermining that the two have no connecting edge;
3. The method of claim 1, wherein calculating the location popularity of the check-in place cThe method comprises the following steps:
sign-in place based on filtering non-first configurationThe post place check-in record CuAnd filtering the non-first configuration check-in placesThe post place check-in record CvDetermining the check-in locationThe ratio of the number of check-in times of all users in the current time period to the total number of check-in times is used as the position popularity
4. The method of claim 1, wherein the popularity is based on the locationDetermining that the user node v is at the check-in placeInfluence received from friend user node after social influence factor is eliminatedThe method comprises the following steps:
5. The method of claim 1, wherein the check-in record C is recorded according to the locationuAnd the place check-in record CvDetermining the similarity S of the user behaviors between the user node u and the user node vu,vThe method comprises the following steps:
according to the place check-in record CuAnd the place check-in record CvDetermining the behavioral similarity S in combination with the improved Jacard similarity factoru,v,
6. The method of claim 1, wherein the e-exponential time decay model is expressed by:
8. an apparatus for processing social network influence, the apparatus comprising:
the system comprises a construction unit, a configuration unit and a processing unit, wherein the construction unit is used for constructing initial social network data of a target social network, the initial social network data is configured by an undirected graph G (V, E, C), V is a user node contained in the target social network data, E is an edge set in the undirected graph G (V, E, C), and C is a place check-in record of the user node;
a determination unit for traversing the location check-in record C of the user node uuAnd a place check-in record C for user node vvDetermining that the user node u and the user node v have the same check-in placeAnd based on the check-in placeDetermining that the two have a connecting edge;
a filtering unit for checking in records C at the placeuAnd the place check-in record CvIn filtering the check-in placeThe location check-in record;
the determining unit is further used for calculating the check-in placeLocation popularity ofAnd according to the position popularityDetermining that the user node v is at the check-in placeRemoving social influence factorsInfluence received by friend user node
A redistribution unit for redistributing the influenceOn the basis, an e exponential time decay model is introduced, and the influence of the user node u on the user node v is redistributed by combining a softmax function to obtain the influence
The determining unit is also used for checking in the record C according to the placeuAnd the place check-in record CvDetermining the similarity S of the user behaviors between the user node u and the user node vu,v;
9. A processing device comprising a processor and a memory, a computer program being stored in the memory, the processor performing the method according to any of claims 1 to 7 when calling the computer program in the memory.
10. A computer-readable storage medium storing a plurality of instructions adapted to be loaded by a processor to perform the method of any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110414260.3A CN113239285A (en) | 2021-04-16 | 2021-04-16 | Processing method, device and processing equipment for social network influence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110414260.3A CN113239285A (en) | 2021-04-16 | 2021-04-16 | Processing method, device and processing equipment for social network influence |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113239285A true CN113239285A (en) | 2021-08-10 |
Family
ID=77128477
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110414260.3A Withdrawn CN113239285A (en) | 2021-04-16 | 2021-04-16 | Processing method, device and processing equipment for social network influence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113239285A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014195958A1 (en) * | 2013-06-03 | 2014-12-11 | Hewlett-Packard Development Company, L.P. | Influence score of a social media domain |
US20150213022A1 (en) * | 2014-01-30 | 2015-07-30 | Linkedin Corporation | System and method for identifying trending topics in a social network |
CN107123056A (en) * | 2017-03-03 | 2017-09-01 | 华南理工大学 | A kind of location-based social big data information maximization method |
CN109213926A (en) * | 2018-07-12 | 2019-01-15 | 南京邮电大学 | A kind of location recommendation method divided based on community with Multi-source Information Fusion |
CN111680228A (en) * | 2020-06-11 | 2020-09-18 | 浙江工商大学 | Matrix decomposition interest point recommendation method based on geographic position and fusion of social influence and category popularity |
CN113127696A (en) * | 2021-03-21 | 2021-07-16 | 武汉大学深圳研究院 | Method for improving accuracy of influence measurement based on behaviors |
-
2021
- 2021-04-16 CN CN202110414260.3A patent/CN113239285A/en not_active Withdrawn
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014195958A1 (en) * | 2013-06-03 | 2014-12-11 | Hewlett-Packard Development Company, L.P. | Influence score of a social media domain |
US20150213022A1 (en) * | 2014-01-30 | 2015-07-30 | Linkedin Corporation | System and method for identifying trending topics in a social network |
CN107123056A (en) * | 2017-03-03 | 2017-09-01 | 华南理工大学 | A kind of location-based social big data information maximization method |
CN109213926A (en) * | 2018-07-12 | 2019-01-15 | 南京邮电大学 | A kind of location recommendation method divided based on community with Multi-source Information Fusion |
CN111680228A (en) * | 2020-06-11 | 2020-09-18 | 浙江工商大学 | Matrix decomposition interest point recommendation method based on geographic position and fusion of social influence and category popularity |
CN113127696A (en) * | 2021-03-21 | 2021-07-16 | 武汉大学深圳研究院 | Method for improving accuracy of influence measurement based on behaviors |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103368917B (en) | A kind of risk control method and system of network virtual user | |
CN107404408B (en) | Virtual identity association identification method and device | |
CN108985954B (en) | Method for establishing association relation of each identifier and related equipment | |
JP2016535369A (en) | Provision of application programs and user recommendation information | |
TW201737072A (en) | Application program project evaluation method and system | |
CN109359263B (en) | User behavior feature extraction method and system | |
EP3345154A1 (en) | Method, apparatus and system for detecting fraudulent software promotion | |
CN108959319B (en) | Information pushing method and device | |
CN108366012B (en) | Social relationship establishing method and device and electronic equipment | |
CN109857943B (en) | Permission level determination method and device, computer equipment and readable storage medium | |
WO2017080454A1 (en) | Website access path aggregation method and device | |
CN108170731A (en) | Data processing method, device, computer storage media and server | |
CN109600344A (en) | Identify the method, apparatus and electronic equipment of risk group | |
CN108989101B (en) | Log output system and method and electronic equipment | |
WO2017101652A1 (en) | Method and apparatus for determining an access path between website pages | |
CN106557282A (en) | The method and apparatus of response write request | |
US20210152576A1 (en) | Cluster-based precision mitigation of network attacks | |
CN113505272B (en) | Control method and device based on behavior habit, electronic equipment and storage medium | |
WO2016165414A1 (en) | Method and device for push information | |
CN110674413B (en) | User relationship mining method, device, equipment and storage medium | |
CN110929141A (en) | Group mining method, device, equipment and storage medium | |
CN103810248A (en) | Method and device for searching for interpersonal relationship based on photos | |
CN113572721B (en) | Abnormal access detection method and device, electronic equipment and storage medium | |
CN107092650B (en) | Weblog analysis method and device | |
CN113239285A (en) | Processing method, device and processing equipment for social network influence |
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 | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210810 |
|
WW01 | Invention patent application withdrawn after publication |