[go: up one dir, main page]

CN113239285A - Processing method, device and processing equipment for social network influence - Google Patents

Processing method, device and processing equipment for social network influence Download PDF

Info

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
location
influence
place
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Withdrawn
Application number
CN202110414260.3A
Other languages
Chinese (zh)
Inventor
李登实
张宇
曾露
梁晓聪
赵兰馨
官端正
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jianghan University
Original Assignee
Jianghan University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jianghan University filed Critical Jianghan University
Priority to CN202110414260.3A priority Critical patent/CN113239285A/en
Publication of CN113239285A publication Critical patent/CN113239285A/en
Withdrawn legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; 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 OR CALCULATING; 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

一种社交网络影响力的处理方法、装置以及处理设备A method, device and processing device for processing social network influence

技术领域technical field

本申请涉及社交领域，具体涉及一种社交网络影响力的处理方法、装置以及处理设备。The present application relates to the field of social networking, and in particular, to a method, an apparatus, and a processing device for processing social network influence.

背景技术Background technique

随着互联网技术以及各大社交网络的不断发展，社交网络中的各种数据愈加庞大，而在对社交网络进行管理的过程中，这些数据显然可以作为数据支持。With the continuous development of Internet technology and major social networks, various data in social networks are becoming larger and larger, and in the process of managing social networks, these data can obviously be used as data support.

其中，在社交网络中度量用户之间的影响力，对于社交网络的研究具有非常重要的作用，并且社交网络影响力的度量在社会舆论监控、个性化营销、精准营销等诸多方面都有广泛的应用需求，因此，对于用户间的社交网络影响力度量的研究受到了很高的重视。Among them, measuring the influence between users in social networks plays a very important role in the research of social networks, and the measurement of social network influence has a wide range of aspects such as social public opinion monitoring, personalized marketing, and precision marketing. Therefore, research on the magnitude of social network influence among users has received high attention.

而在现有的相关技术的研究过程中，发明人发现，尽管目前国内外已有大量相关研究成果，然而在这些研究中基于行为的用户影响力度量主要是通过用户之间的发生了相同行为来判定两用户之间存在影响力，而在实际应用中，这种判断方式存在不稳定的情况，或者说，现有的用户间的社交网络影响力的判断方式，其精度不高。In the research process of the existing related technologies, the inventor found that although there are a large number of related research results at home and abroad, the influence of users based on behavior in these researches is mainly through the same behavior between users. to determine the influence between two users, but in practical applications, this judgment method is unstable, or the existing social network influence judgment method between users is not accurate.

发明内容SUMMARY OF THE INVENTION

本申请提供一种社交网络影响力的处理方法、装置以及处理设备，用于从用户位置角度出发，将用户间的社交网络影响力从复杂的混淆因素中分离出来，得到更为精确的用户间的社交网络影响力。The present application provides a method, device and processing device for processing social network influence, which are used to separate the social network influence among users from complex confounding factors from the perspective of user location, so as to obtain a more accurate information between users. social network influence.

第一方面，本申请提供了一种社交网络影响力的处理方法，方法包括：In a first aspect, the present application provides a method for processing social network influence, the method comprising:

构建目标社交网络的初始社交网络数据，其中，初始社交网络数据以无向图G(V，E，C)配置，V为目标社交网络数据中包含的用户节点，E为无向图G(V，E，C)中的边集，C为用户节点的地点签到记录；Construct the initial social network data of the target social network, where the initial social network data is configured with an undirected graph G(V, E, C), V is the user node contained in the target social network data, and E is the undirected graph G(V , E, C) in the edge set, C is the location check-in record of the user node;

遍历用户节点u的地点签到记录Cu以及用户节点v的地点签到记录Cv，确定用户节点u与用户节点v存在相同签到地点

Figure BDA00030251330800000229
并基于签到地点
Figure BDA00030251330800000228
确定两者存在连边；Traverse the location check-in record C u of the user node u and the location check-in record C v of the user node v, and determine that the user node u and the user node v have the same check-in location
Figure BDA00030251330800000229
and based on the check-in location
Figure BDA00030251330800000228
Make sure there is a connection between the two;

在地点签到记录Cu以及地点签到记录Cv中，过滤非首次配置签到地点

Figure BDA00030251330800000230
的地点签到记录；In the location check-in record C u and the location check-in record C v , filter the non-first-time configuration check-in location
Figure BDA00030251330800000230
location check-in record;

计算签到地点

Figure BDA0003025133080000021
的位置流行度
Figure BDA0003025133080000022
并根据位置流行度
Figure BDA0003025133080000023
确定用户节点v在签到地点
Figure BDA0003025133080000024
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000025
Calculate check-in locations
Figure BDA0003025133080000021
location popularity
Figure BDA0003025133080000022
and based on location popularity
Figure BDA0003025133080000023
Make sure that user node v is at the check-in location
Figure BDA0003025133080000024
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000025

在影响力

Figure BDA0003025133080000026
的基础上，引入e指数时间衰减模型，并结合softmax函数进行用户节点u对用户节点v的影响力重分配，得到影响力
Figure BDA0003025133080000027
in influence
Figure BDA0003025133080000026
On the basis of , the e-exponential time decay model is introduced, and the softmax function is used to redistribute the influence of user node u to user node v to obtain influence.
Figure BDA0003025133080000027

根据地点签到记录Cu以及地点签到记录Cv，确定用户节点u与用户节点v之间的用户行为相似度Su，v；According to the location check-in record C u and the location check-in record C v , determine the user behavior similarity Su,v between the user node u and the user node v ;

将影响力

Figure BDA0003025133080000028
与用户行为相似度Su，v的比值，作为用户节点u对用户节点v的影响力输出。will influence
Figure BDA0003025133080000028
The ratio of the similarity S u, v to the user behavior is output as the influence of the user node u on the user node v.

结合本申请第一方面，在本申请第一方面第一种可能的实现方式中，基于签到地点

Figure BDA0003025133080000029
确定两者存在连边，包括：In combination with the first aspect of the present application, in the first possible implementation manner of the first aspect of the present application, based on the check-in location
Figure BDA0003025133080000029
Make sure there is a connection between the two, including:

若用户节点u在签到地点

Figure BDA00030251330800000210
的签到时间
Figure BDA00030251330800000211
晚于或者等于用户节点v在签到地点
Figure BDA00030251330800000212
的签到时间
Figure BDA00030251330800000213
则确定两者未存在连边；If user node u is at the check-in location
Figure BDA00030251330800000210
check-in time
Figure BDA00030251330800000211
later than or equal to user node v at the check-in location
Figure BDA00030251330800000212
check-in time
Figure BDA00030251330800000213
Then it is determined that there is no connection between the two;

若用户节点u在签到地点

Figure BDA00030251330800000214
的签到时间
Figure BDA00030251330800000215
早于用户节点v在签到地点
Figure BDA00030251330800000216
的签到时间
Figure BDA00030251330800000217
则确定两者存在连边。If user node u is at the check-in location
Figure BDA00030251330800000214
check-in time
Figure BDA00030251330800000215
earlier than user node v at the check-in location
Figure BDA00030251330800000216
check-in time
Figure BDA00030251330800000217
It is determined that there is a connection between the two.

结合本申请第一方面，在本申请第一方面第二种可能的实现方式中，计算签到地点c的位置流行度

Figure BDA00030251330800000218
包括：With reference to the first aspect of the present application, in the second possible implementation manner of the first aspect of the present application, the location popularity of the check-in location c is calculated
Figure BDA00030251330800000218
include:

基于过滤非首次配置签到地点

Figure BDA00030251330800000219
后的地点签到记录Cu以及过滤非首次配置签到地点
Figure BDA00030251330800000220
后的地点签到记录Cv，确定签到地点
Figure BDA00030251330800000221
在当前时间段内被所有用户签到的次数占总签到次数的比例，作为位置流行度
Figure BDA00030251330800000222
Based on filtering non-first-time configuration check-in locations
Figure BDA00030251330800000219
Check-in records C u of later locations and filter non-first-time check-in locations
Figure BDA00030251330800000220
Check-in record C v at the later location to determine the check-in location
Figure BDA00030251330800000221
The ratio of the number of check-ins by all users to the total number of check-ins in the current time period is used as the location popularity
Figure BDA00030251330800000222

位置流行度

Figure BDA00030251330800000223
Location popularity
Figure BDA00030251330800000223

δ为时段阈值，

Figure BDA00030251330800000224
表示用户节点v的在签到地点
Figure BDA00030251330800000225
的签到时刻
Figure BDA00030251330800000226
比用户节点u的签到时刻
Figure BDA00030251330800000227
早，||表示集合的大小。δ is the time period threshold,
Figure BDA00030251330800000224
Indicates the check-in location of user node v
Figure BDA00030251330800000225
check-in time
Figure BDA00030251330800000226
than the check-in time of user node u
Figure BDA00030251330800000227
Early, || indicates the size of the collection.

结合本申请第一方面，在本申请第一方面第三种可能的实现方式中，根据位置流行度

Figure BDA0003025133080000031
确定用户节点v在签到地点
Figure BDA0003025133080000032
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000033
包括：With reference to the first aspect of the present application, in a third possible implementation manner of the first aspect of the present application, according to the location popularity
Figure BDA0003025133080000031
Make sure that user node v is at the check-in location
Figure BDA0003025133080000032
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000033
include:

将预设的调节系数f，与位置流行度

Figure BDA0003025133080000034
的比值，作为用户节点v在签到地点
Figure BDA0003025133080000035
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000036
Compare the preset adjustment factor f with the location popularity
Figure BDA0003025133080000034
The ratio of , as the user node v at the check-in location
Figure BDA0003025133080000035
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000036

结合本申请第一方面，在本申请第一方面第四种可能的实现方式中，根据地点签到记录Cu以及地点签到记录Cv，确定用户节点u与用户节点v之间的用户行为相似度Su，v，包括：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 user behavior similarity between the user node u and the user node v is determined according to the location check-in record C u and the location check-in record C v Su, v , including:

根据地点签到记录Cu以及地点签到记录Cv，结合改进的杰卡德相似度系数，确定行为相似度Su，v，According to the location check-in record C u and the location check-in record C v , combined with the improved Jaccard similarity coefficient, determine the behavior similarity Su , v ,

行为相似度

Figure BDA0003025133080000037
behavioral similarity
Figure BDA0003025133080000037

Figure BDA00030251330800000316
为用户节点u在到达签到地点
Figure BDA0003025133080000038
之前的地点签到记录，lv(cv)为用户节点v在到达签到地点
Figure BDA0003025133080000039
之前的地点签到记录。
Figure BDA00030251330800000316
for the user node u to arrive at the check-in location
Figure BDA0003025133080000038
The previous location check-in record, l v (c v ) is the user node v arriving at the check-in location
Figure BDA0003025133080000039
Previous location check-in records.

结合本申请第一方面，在本申请第一方面第五种可能的实现方式中，e指数时间衰减模型的表达式为：In conjunction with the first aspect of the present application, in the fifth possible implementation manner of the first aspect of the present application, the expression of the e-exponential time decay model is:

Figure BDA00030251330800000310
Figure BDA00030251330800000310

P(v|u)为用户节点u当前对用户节点v的影响力，σ为衰减系数，

Figure BDA00030251330800000311
处于0与1之间，P′(v|u)表示经过时间衰减前用户节点u对用户节点v的影响力。P(v|u) is the current influence of user node u on user node v, σ is the attenuation coefficient,
Figure BDA00030251330800000311
Between 0 and 1, P'(v|u) represents the influence of user node u on user node v before time decay.

结合本申请第一方面，在本申请第一方面第六种可能的实现方式中，影响力

Figure BDA00030251330800000312
的表达式为：In conjunction with the first aspect of the present application, in the sixth possible implementation manner of the first aspect of the present application, the influence
Figure BDA00030251330800000312
The expression is:

Figure BDA00030251330800000313
Figure BDA00030251330800000313

第二方面，本申请提供了一种社交网络影响力的处理装置，装置包括：In a second aspect, the present application provides a device for processing social network influence, the device comprising:

构建单元，用于构建目标社交网络的初始社交网络数据，其中，初始社交网络数据以无向图G(V，E，C)配置，V为目标社交网络数据中包含的用户节点，E为无向图G(V，E，C)中的边集，C为用户节点的地点签到记录；The construction unit is used to construct the initial social network data of the target social network, wherein the initial social network data is configured with an undirected graph G(V, E, C), V is the user node included in the target social network data, and E is no To the edge set in the graph G(V, E, C), C is the location check-in record of the user node;

确定单元，用于遍历用户节点u的地点签到记录Cu以及用户节点v的地点签到记录Cv，确定用户节点u与用户节点v存在相同签到地点

Figure BDA00030251330800000314
并基于签到地点
Figure BDA00030251330800000315
确定两者存在连边；The determining unit is used to traverse the location check-in record C u of the user node u and the location check-in record C v of the user node v, and determine that the user node u and the user node v have the same check-in location
Figure BDA00030251330800000314
and based on the check-in location
Figure BDA00030251330800000315
Make sure there is a connection between the two;

过滤单元，用于在地点签到记录Cu以及地点签到记录Cv中，过滤非首次配置签到地点

Figure BDA0003025133080000041
的地点签到记录；The filtering unit is used to filter the non-first-time check-in locations in the location check-in record C u and the location check-in record C v
Figure BDA0003025133080000041
location check-in record;

确定单元，还用于计算签到地点

Figure BDA0003025133080000042
的位置流行度
Figure BDA0003025133080000043
并根据位置流行度
Figure BDA0003025133080000044
确定用户节点v在签到地点
Figure BDA0003025133080000045
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000046
Determine the unit, also used to calculate the check-in location
Figure BDA0003025133080000042
location popularity
Figure BDA0003025133080000043
and based on location popularity
Figure BDA0003025133080000044
Make sure that user node v is at the check-in location
Figure BDA0003025133080000045
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000046

重分配单元，用于在影响力

Figure BDA0003025133080000047
的基础上，引入e指数时间衰减模型，并结合softmax函数进行用户节点u对用户节点v的影响力重分配，得到影响力
Figure BDA0003025133080000048
redistribution unit, used in influence
Figure BDA0003025133080000047
On the basis of , the e-exponential time decay model is introduced, and the softmax function is used to redistribute the influence of user node u to user node v to obtain influence.
Figure BDA0003025133080000048

确定单元，还用于根据地点签到记录Cu以及地点签到记录Cv，确定用户节点u与用户节点v之间的用户行为相似度Su，v；The determining unit is also used to determine the user behavior similarity S u, v between the user node u and the user node v according to the location check-in record C u and the location check-in record C v ;

输出单元，用于将影响力

Figure BDA0003025133080000049
与用户行为相似度Su，v的比值，作为用户节点u对用户节点v的影响力输出。output unit, which is used to convert the influence
Figure BDA0003025133080000049
The ratio of the similarity S u, v to the user behavior is output as the influence of the user node u on the user node v.

结合本申请第二方面，在本申请第二方面第一种可能的实现方式中，确定单元，具体用于：In conjunction with the second aspect of the present application, in the first possible implementation manner of the second aspect of the present application, the determining unit is specifically used for:

若用户节点u在签到地点

Figure BDA00030251330800000410
的签到时间
Figure BDA00030251330800000411
晚于或者等于用户节点v在签到地点
Figure BDA00030251330800000412
的签到时间
Figure BDA00030251330800000413
则确定两者未存在连边；If user node u is at the check-in location
Figure BDA00030251330800000410
check-in time
Figure BDA00030251330800000411
later than or equal to user node v at the check-in location
Figure BDA00030251330800000412
check-in time
Figure BDA00030251330800000413
Then it is determined that there is no connection between the two;

若用户节点u在签到地点

Figure BDA00030251330800000414
的签到时间
Figure BDA00030251330800000415
早于用户节点v在签到地点
Figure BDA00030251330800000416
的签到时间
Figure BDA00030251330800000417
则确定两者存在连边。If user node u is at the check-in location
Figure BDA00030251330800000414
check-in time
Figure BDA00030251330800000415
earlier than user node v at the check-in location
Figure BDA00030251330800000416
check-in time
Figure BDA00030251330800000417
It is determined that there is a connection between the two.

结合本申请第二方面，在本申请第二方面第二种可能的实现方式中，确定单元，具体用于：In conjunction with the second aspect of the present application, in the second possible implementation manner of the second aspect of the present application, the determining unit is specifically used for:

基于过滤非首次配置签到地点

Figure BDA00030251330800000418
后的地点签到记录Cu以及过滤非首次配置签到地点
Figure BDA00030251330800000419
后的地点签到记录Cv，确定签到地点
Figure BDA00030251330800000420
在当前时间段内被所有用户签到的次数占总签到次数的比例，作为位置流行度
Figure BDA00030251330800000421
Based on filtering non-first-time configuration check-in locations
Figure BDA00030251330800000418
Check-in records C u of later locations and filter non-first-time check-in locations
Figure BDA00030251330800000419
Check-in record C v at the later location to determine the check-in location
Figure BDA00030251330800000420
The ratio of the number of check-ins by all users to the total number of check-ins in the current time period is used as the location popularity
Figure BDA00030251330800000421

位置流行度

Figure BDA00030251330800000422
Location popularity
Figure BDA00030251330800000422

δ为时段阈值，

Figure BDA00030251330800000423
表示用户节点v的在签到地点
Figure BDA00030251330800000424
的签到时刻
Figure BDA00030251330800000425
比用户节点u的签到时刻
Figure BDA00030251330800000426
早，||表示集合的大小。δ is the time period threshold,
Figure BDA00030251330800000423
Indicates the check-in location of user node v
Figure BDA00030251330800000424
check-in time
Figure BDA00030251330800000425
than the check-in time of user node u
Figure BDA00030251330800000426
Early, || indicates the size of the collection.

结合本申请第二方面，在本申请第二方面第三种可能的实现方式中，确定单元，具体用于：In conjunction with 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 used for:

将预设的调节系数f，与位置流行度

Figure BDA0003025133080000051
的比值，作为用户节点v在签到地点
Figure BDA0003025133080000052
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000053
Compare the preset adjustment factor f with the location popularity
Figure BDA0003025133080000051
The ratio of , as the user node v at the check-in location
Figure BDA0003025133080000052
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000053

结合本申请第二方面，在本申请第二方面第四种可能的实现方式中，确定单元，具体用于：In conjunction with 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 used for:

根据地点签到记录Cu以及地点签到记录Cv，结合改进的杰卡德相似度系数，确定行为相似度Su，v，According to the location check-in record C u and the location check-in record C v , combined with the improved Jaccard similarity coefficient, determine the behavior similarity Su , v ,

行为相似度

Figure BDA0003025133080000054
behavioral similarity
Figure BDA0003025133080000054

Figure BDA0003025133080000055
为用户节点u在到达签到地点
Figure BDA0003025133080000056
之前的地点签到记录，lv(cv)为用户节点v在到达签到地点
Figure BDA0003025133080000057
之前的地点签到记录。
Figure BDA0003025133080000055
for the user node u to arrive at the check-in location
Figure BDA0003025133080000056
The previous location check-in record, l v (c v ) is the user node v arriving at the check-in location
Figure BDA0003025133080000057
Previous location check-in records.

结合本申请第二方面，在本申请第二方面第五种可能的实现方式中，e指数时间衰减模型的表达式为：In conjunction with the second aspect of the present application, in the fifth possible implementation manner of the second aspect of the present application, the expression of the e-exponential time decay model is:

Figure BDA0003025133080000058
Figure BDA0003025133080000058

P(v|u)为用户节点u当前对用户节点v的影响力，σ为衰减系数，

Figure BDA0003025133080000059
处于0与1之间，P′(v|u)表示经过时间衰减前用户节点u对用户节点v的影响力。P(v|u) is the current influence of user node u on user node v, σ is the attenuation coefficient,
Figure BDA0003025133080000059
Between 0 and 1, P'(v|u) represents the influence of user node u on user node v before time decay.

结合本申请第二方面，在本申请第二方面第六种可能的实现方式中，影响力

Figure BDA00030251330800000510
的表达式为：In combination with the second aspect of the present application, in the sixth possible implementation manner of the second aspect of the present application, the influence
Figure BDA00030251330800000510
The expression is:

Figure BDA00030251330800000511
Figure BDA00030251330800000511

第三方面，本申请提供了一种处理设备，包括处理器和存储器，存储器中存储有计算机程序，处理器调用存储器中的计算机程序时执行本申请第一方面或者本申请第一方面任一种可能的实现方式提供的方法。In a third aspect, the present application provides a processing device, including a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the first aspect of the present application or any one of the first aspect of the present application when the processor calls the computer program in the memory methods provided by possible implementations.

第四方面，本申请提供了一种计算机可读存储介质，计算机可读存储介质存储有多条指令，指令适于处理器进行加载，以执行本申请第一方面或者本申请第一方面任一种可能的实现方式提供的方法。In a fourth aspect, the present application provides a computer-readable storage medium. The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the first aspect of the present application or any one of the first aspect of the present application. methods provided by a possible implementation.

从以上内容可得出，本申请具有以下的有益效果：It can be drawn from the above content that the present application has the following beneficial effects:

针对于用户间的社交网络影响力的捕捉，本申请首先以无向图的形式构建目标社交网络的初始社交网络数据，在该无向图中，配置了基于位置访问的社交网络形式，在基于地点签到记录确定用户节点u与用户节点v存在相同签到地点

Figure BDA0003025133080000061
并基于该签到地点
Figure BDA0003025133080000062
确定两者存在连边，即确定存在社交网络影响力后，在地点签到记录中过滤非首次配置签到地点
Figure BDA0003025133080000063
的地点签到记录，以排除用户个人习惯的干扰因素，接着计算该签到地点
Figure BDA0003025133080000064
的位置流行度
Figure BDA0003025133080000065
并根据位置流行度
Figure BDA0003025133080000066
确定用户节点v在签到地点
Figure BDA0003025133080000067
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000068
以排除社会大众的干扰因素，并继续引入e指数时间衰减模型，并结合soffmax函数进行用户节点u对用户节点v的影响力重分配，得到影响力
Figure BDA0003025133080000069
以排除用户自身社交圈的干扰因素，同时还根据地点签到记录确定用户节点u与用户节点v之间的用户行为相似度Su，v，此时，将影响力
Figure BDA00030251330800000610
与用户行为相似度Su，v的比值作为用户节点u对用户节点v的影响力进行输出，如此，从用户位置角度出发，通过多层的干扰因素排除机制，将用户间的社交网络影响力从复杂的混淆因素中分离出来，得到更为精确的用户间的社交网络影响力，对于目标社交网络的管控提供强有力的数据支持。In order to capture the influence of social networks among users, the present application first constructs the initial social network data of the target social network in the form of an undirected graph. The location check-in record confirms that user node u and user node v have the same check-in location
Figure BDA0003025133080000061
and based on that check-in location
Figure BDA0003025133080000062
It is determined that there is a connection between the two, that is, after determining that there is social network influence, filter the non-first-time check-in location in the location check-in record
Figure BDA0003025133080000063
check-in record of the location to exclude the interference factors of the user's personal habits, and then calculate the check-in location
Figure BDA0003025133080000064
location popularity
Figure BDA0003025133080000065
and based on location popularity
Figure BDA0003025133080000066
Make sure that user node v is at the check-in location
Figure BDA0003025133080000067
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000068
In order to eliminate the interference factors of the public, and continue to introduce the e exponential time decay model, and combine the soffmax function to redistribute the influence of the user node u to the user node v, and obtain the influence
Figure BDA0003025133080000069
In order to eliminate the interference factors of the user's own social circle, and also according to the location check-in record to determine the user behavior similarity S u, v between the user node u and the user node v, at this time, the influence
Figure BDA00030251330800000610
The ratio of similarity with user behavior Su , v is output as the influence of user node u on user node v. In this way, from the perspective of user location, through the multi-layer interference factor elimination mechanism, the social network influence between users is calculated. It can be separated from complex confounding factors to obtain more accurate social network influence among users, and provide strong data support for the management and control of target social networks.

附图说明Description of drawings

为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

图1为本申请社交网络影响力的处理方法的一种流程示意图；1 is a schematic flowchart of a method for processing social network influence of the application;

图2为本申请社交网络影响力的处理装置的一种结构示意图；FIG. 2 is a schematic structural diagram of an apparatus for processing social network influence of the present application;

图3为本申请处理设备的一种结构示意图。FIG. 3 is a schematic structural diagram of the processing equipment of the present application.

具体实施方式Detailed ways

下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本申请保护的范围。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. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或模块的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或模块，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或模块。在本申请中出现的对步骤进行的命名或者编号，并不意味着必须按照命名或者编号所指示的时间/逻辑先后顺序执行方法流程中的步骤，已经命名或者编号的流程步骤可以根据要实现的技术目的变更执行次序，只要能达到相同或者相类似的技术效果即可。The terms "first", "second" and the like in the description and claims of the present application and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It is to be understood that data so used may be interchanged under appropriate circumstances so that the embodiments described herein can be practiced in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having", and any variations thereof, are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device comprising a series of steps or modules is not necessarily limited to those expressly listed Rather, those steps or modules may include other steps or modules not expressly listed or inherent to the process, method, product or apparatus. The naming or numbering of the steps in this application does not mean that the steps in the method flow must be executed in the time/logical sequence indicated by the naming or numbering, and the named or numbered process steps can be implemented according to the The technical purpose is to change the execution order, as long as the same or similar technical effects can be achieved.

本申请中所出现的模块的划分，是一种逻辑上的划分，实际应用中实现时可以有另外的划分方式，例如多个模块可以结合成或集成在另一个系统中，或一些特征可以忽略，或不执行，另外，所显示的或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，模块之间的间接耦合或通信连接可以是电性或其他类似的形式，本申请中均不作限定。并且，作为分离部件说明的模块或子模块可以是也可以不是物理上的分离，可以是也可以不是物理模块，或者可以分布到多个电路模块中，可以根据实际的需要选择其中的部分或全部模块来实现本申请方案的目的。The division of modules in this application is a logical division. In practical applications, there may be other divisions. For example, multiple modules may be combined or integrated into another system, or some features may be ignored. , or not implemented, in addition, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, and the indirect coupling or communication connection between modules may be electrical or other similar forms. There are no restrictions in the application. In addition, 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 into multiple circuit modules, and some or all of them may be selected according to actual needs. module to achieve the purpose of the solution of this application.

在介绍本申请提供的社交网络影响力的处理方法之前，首先介绍本申请所涉及的背景内容。Before introducing the method for processing social network influence provided by this application, the background content involved in this application is first introduced.

本申请提供的社交网络影响力的处理方法、装置以及计算机可读存储介质，可应用于处理设备上，用于从用户位置角度出发，将用户间的社交网络影响力从复杂的混淆因素中分离出来，得到更为精确的用户间的社交网络影响力。The method, device and computer-readable storage medium for processing social network influence provided by the present application can be applied to processing equipment, and are used to separate the social network influence among users from complex confounding factors from the perspective of user location. Come out, get more accurate social network influence among users.

本申请提及的社交网络影响力的处理方法，其执行主体可以为社交网络影响力装置，或者集成了该社交网络影响力的处理装置的服务器、物理主机或者用户设备(UserEquipment，UE)等类型的处理设备。其中，社交网络影响力的处理装置可以采用硬件或者软件的方式实现，UE具体可以为智能手机、平板电脑、笔记本电脑、台式电脑或者个人数字助理(Personal Digital Assistant，PDA)等终端设备，处理设备可以通过设备集群的方式设置。For the method for processing social network influence mentioned in this application, the execution body may be a social network influence device, or a server, physical host, or user equipment (User Equipment, UE) that integrates the social network influence processing device. processing equipment. The device for processing social network influence may be implemented in hardware or software, and the UE may be a terminal device such as a smartphone, tablet computer, notebook computer, desktop computer, or personal digital assistant (PDA), and the processing device It can be set by means of device cluster.

下面，开始介绍本申请提供的社交网络影响力的处理方法。Next, the method for processing social network influence provided by the present application will be introduced.

首先，参阅图1，图1示出了本申请社交网络影响力的处理方法的一种流程示意图，本申请提供的社交网络影响力的处理方法，具体可包括如下步骤：First, referring to FIG. 1, FIG. 1 shows a schematic flowchart of a method for processing social network influence of the present application. The processing method for social network influence provided by the present application may specifically include the following steps:

步骤S101，构建目标社交网络的初始社交网络数据，其中，初始社交网络数据以无向图G(V，E，C)配置，V为目标社交网络数据中包含的用户节点，E为无向图G(V，E，C)中的边集，C为用户节点的地点签到记录；Step S101, constructing initial social network data of the target social network, wherein the initial social network data is configured with an undirected graph G(V, E, C), V is the user node included in the target social network data, and E is an undirected graph The edge set in G(V, E, C), C is the location check-in record of the user node;

步骤S102，遍历用户节点u的地点签到记录Cu以及用户节点v的地点签到记录Cv，确定用户节点u与用户节点v存在相同签到地点

Figure BDA0003025133080000081
并基于签到地点
Figure BDA0003025133080000082
确定两者存在连边；Step S102, traverse the location check-in record C u of the user node u and the location check-in record C v of the user node v, and determine that the user node u and the user node v have the same check-in location
Figure BDA0003025133080000081
and based on the check-in location
Figure BDA0003025133080000082
Make sure there is a connection between the two;

步骤S103，在地点签到记录Cu以及地点签到记录Cv中，过滤非首次配置签到地点

Figure BDA0003025133080000083
的地点签到记录；Step S103, in the location check-in record C u and the location check-in record C v , filter the non-first-time configuration check-in location
Figure BDA0003025133080000083
location check-in record;

步骤S104，计算签到地点

Figure BDA0003025133080000084
的位置流行度
Figure BDA0003025133080000085
并根据位置流行度
Figure BDA0003025133080000086
确定用户节点v在签到地点
Figure BDA0003025133080000087
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000088
Step S104, calculate the check-in location
Figure BDA0003025133080000084
location popularity
Figure BDA0003025133080000085
and based on location popularity
Figure BDA0003025133080000086
Make sure that user node v is at the check-in location
Figure BDA0003025133080000087
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000088

步骤S105，在影响力

Figure BDA0003025133080000089
的基础上，引入e指数时间衰减模型，并结合softmax函数进行用户节点u对用户节点v的影响力重分配，得到影响力
Figure BDA00030251330800000810
Step S105, in the influence
Figure BDA0003025133080000089
On the basis of , the e-exponential time decay model is introduced, and the softmax function is used to redistribute the influence of user node u to user node v to obtain influence.
Figure BDA00030251330800000810

步骤S106，根据地点签到记录Cu以及地点签到记录Cv，确定用户节点u与用户节点v之间的用户行为相似度Su，v；Step S106, according to the location check-in record C u and the location check-in record C v , determine the user behavior similarity Su,v between the user node u and the user node v ;

步骤S107，将影响力

Figure BDA00030251330800000811
与用户行为相似度Su，v的比值，作为用户节点u对用户节点v的影响力输出。In step S107, the influence
Figure BDA00030251330800000811
The ratio of the similarity S u, v to the user behavior is output as the influence of the user node u on the user node v.

从图1所示实施例可看出，针对于用户间的社交网络影响力的捕捉，本申请首先以无向图的形式构建目标社交网络的初始社交网络数据，在该无向图中，配置了基于位置访问的社交网络形式，在基于地点签到记录确定用户节点u与用户节点v存在相同签到地点

Figure BDA00030251330800000812
并基于该签到地点
Figure BDA00030251330800000813
确定两者存在连边，即确定存在社交网络影响力后，在地点签到记录中过滤非首次配置签到地点
Figure BDA00030251330800000814
的地点签到记录，以排除用户个人习惯的干扰因素，接着计算该签到地点
Figure BDA00030251330800000815
的位置流行度
Figure BDA0003025133080000091
并根据位置流行度
Figure BDA0003025133080000092
确定用户节点v在签到地点
Figure BDA0003025133080000093
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000094
以排除社会大众的干扰因素，并继续引入e指数时间衰减模型，并结合softmax函数进行用户节点u对用户节点v的影响力重分配，得到影响力
Figure BDA0003025133080000095
以排除用户自身社交圈的干扰因素，同时还根据地点签到记录确定用户节点u与用户节点v之间的用户行为相似度Su，v，此时，将影响力
Figure BDA0003025133080000096
与用户行为相似度Su，v的比值作为用户节点u对用户节点v的影响力进行输出，如此，从用户位置角度出发，通过多层的干扰因素排除机制，将用户间的社交网络影响力从复杂的混淆因素中分离出来，得到更为精确的用户间的社交网络影响力，对于目标社交网络的管控提供强有力的数据支持。It can be seen from the embodiment shown in FIG. 1 that, in order to capture the influence of social networks among users, the present application first constructs the initial social network data of the target social network in the form of an undirected graph. In the undirected graph, configure In the form of social network based on location access, it is determined that user node u and user node v have the same check-in location based on location check-in records.
Figure BDA00030251330800000812
and based on that check-in location
Figure BDA00030251330800000813
It is determined that there is a connection between the two, that is, after determining that there is social network influence, filter the non-first-time check-in location in the location check-in record
Figure BDA00030251330800000814
check-in record of the location to exclude the interference factors of the user's personal habits, and then calculate the check-in location
Figure BDA00030251330800000815
location popularity
Figure BDA0003025133080000091
and based on location popularity
Figure BDA0003025133080000092
Make sure that user node v is at the check-in location
Figure BDA0003025133080000093
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000094
In order to eliminate the interference factors of the public, and continue to introduce the e exponential time decay model, and combine the softmax function to redistribute the influence of the user node u to the user node v, and obtain the influence
Figure BDA0003025133080000095
In order to eliminate the interference factors of the user's own social circle, and also determine the user behavior similarity S u, v between the user node u and the user node v according to the location check-in record, at this time, the influence
Figure BDA0003025133080000096
The ratio of similarity with user behavior Su , v is output as the influence of user node u on user node v. In this way, from the perspective of user location, through the multi-layer interference factor elimination mechanism, the social network influence between users is calculated. It can be separated from complex confounding factors to obtain more accurate social network influence among users, and provide strong data support for the management and control of target social network.

下面则对上述图1所示实施例的各个步骤及其在实际应用中可能的实现方式进行详细阐述。The steps of the above-mentioned embodiment shown in FIG. 1 and possible implementations thereof in practical applications are described in detail below.

在本申请中，应用本申请提供的社交网络影响力的处理方法的处理设备，具体可以为目标社交网络中的设备，如此，对于目标社交网络的管理方，可从内部捕捉到更为精确的用户间的社交网络影响力，以此作为数据支持，实现对目标社交网络更为精确的数据研究、舆论管控等管理工作。In this application, the processing device that applies the method for processing social network influence provided by this application may specifically be a device in the target social network. In this way, for the manager of the target social network, more accurate information can be captured from the inside. The social network influence among users is used as data support to achieve more accurate data research, public opinion management and other management work on the target social network.

当然，处理设备具体也可以为目标社交网络外部的设备，在网络外部进行用户间的社交网络影响力的捕捉，为目标社交网络的管理工作提供数据支持。Of course, the processing device can also be a device outside the target social network, which captures the social network influence among users outside the network, and provides data support for the management of the target social network.

可以理解的是，对于这些用户间的社交网络影响力的捕捉，在实际应用中，是在满足用户隐私需求，或者得到用户确认后进行的。It can be understood that, in practical applications, the capture of the social network influence among these users is performed after satisfying the user's privacy requirements or obtaining confirmation from the user.

目标社交网络，可以为任意形式的社交网络，例如现今的微博、贴吧等社交网络，具体可随定义的社交网络形式调整，其中，目标社交网络不仅可以值得是单一的社交网络产品所构建的社交网络，还可以为不同社交网络产品所构建的社交网络，具体在此不做限定。The target social network can be any form of social network, such as today's social networks such as Weibo, Tieba, etc., which can be adjusted according to the defined social network form. Among them, the target social network can not only be worthy of being constructed by a single social network product The social network may also be a social network constructed for different social network products, which is not specifically limited here.

在现有的用户间的社交网络影响力判断机制中，主要通过用户之间的发生了相同行为来判定两用户之间存在影响力，而在很多情况下发现，不同时间、同一地点行为的用户之间，并不一定都存在影响力，甚至，大多数用户之间可能出现毫无关联的情况。In the existing social network influence judgment mechanism between users, the influence between two users is mainly determined by the same behavior between users. In many cases, it is found that users who behave at different times and at the same place There is not necessarily an influence between them, and even most users may be unrelated.

因此，本申请以用户位置出发，构建多层的干扰因素排除机制，以将用户间的社交网络影响力从复杂的混淆因素中分离出来，显著提高影响力的精度。Therefore, the present application builds a multi-layer interference factor elimination mechanism based on the user's location, so as to separate the social network influence among users from complex confounding factors, and significantly improve the accuracy of the influence.

首先，对于目标社交网络，构建一无向图G(V，E，C)，作为初始社交网络数据，在该无向图G(V，E，C)中，V为目标社交网络数据中包含的用户节点，E为无向图G(V，E，C)中的边集，C为用户节点的地点签到记录。First, for the target social network, construct an undirected graph G(V, E, C), as the initial social network data, in the undirected graph G(V, E, C), V is the target social network data contains The user node of , E is the edge set in the undirected graph G(V, E, C), and C is the location check-in record of the user node.

以用户节点u为例，Cu作为用户节点u的地点签到记录，其中可包括用户节点u在不同地点不同时间的签到记录，具体的，可表示为

Figure BDA0003025133080000101
表示任意用户u的签到记录，其中i表示用户节点的第i个签到记录，
Figure BDA0003025133080000102
表示用户节点u的第i个签到地点，
Figure BDA0003025133080000103
表示用户节点u在签到地点
Figure BDA0003025133080000104
的签到时间。Taking the user node u as an example, C u is the location check-in record of the user node u, which may include the check-in records of the user node u at different places and different times. Specifically, it can be expressed as
Figure BDA0003025133080000101
represents the check-in record of any user u, where i represents the ith check-in record of the user node,
Figure BDA0003025133080000102
represents the ith check-in location of user node u,
Figure BDA0003025133080000103
Indicates that the user node u is at the check-in location
Figure BDA0003025133080000104
check-in time.

该地点签到记录，可以理解为所在位置的定位的上报记录，可通过实时定位数据直接得到，也可以通过餐厅等地标间接得到。The check-in record of the location can be understood as the reporting record of the location's location, which can be obtained directly through real-time positioning data, or indirectly through landmarks such as restaurants.

可以理解的是，此时，无向图G(V，E，C)为初始状态，而在预期中，完成状态的无向图G(V，E，C)中，用户节点之间若存在社交网络影响力，其用户节点之间是存在连边情况的。It can be understood that at this time, the undirected graph G(V, E, C) is the initial state, and as expected, in the undirected graph G(V, E, C) of the completed state, if there is a user node The influence of a social network is connected between its user nodes.

从而，本申请遍历用户节点u的地点签到记录Cu以及用户节点v的地点签到记录Cv，以确定用户节点u与用户节点v存在相同签到地点

Figure BDA0003025133080000105
并基于该签到地点
Figure BDA0003025133080000106
确定两者存在连边，此时方可进行后续影响力的捕捉。Therefore, the present application traverses the location check-in record C u of the user node u and the location check-in record C v of the user node v to determine that the user node u and the user node v have the same check-in location.
Figure BDA0003025133080000105
and based on that check-in location
Figure BDA0003025133080000106
After confirming that there is a connection between the two, the subsequent influence can be captured only at this time.

可以理解的是，若两用户节点的地点签到记录中存在相同签到地点

Figure BDA0003025133080000107
可直接认定两者之间存在社交网络影响力。It is understandable that if the same check-in location exists in the location check-in records of the two user nodes
Figure BDA0003025133080000107
It can be directly identified that there is social network influence between the two.

而在进一步的研究中，本申请还认为还可对该判断进行优化，即，在存在相同签到地点

Figure BDA0003025133080000108
的情况下，继续结合签到时间
Figure BDA0003025133080000109
提高社交网络影响力判断结果的精度。In further research, the present application also believes that the judgment can be optimized, that is, when the same check-in location exists
Figure BDA0003025133080000108
In the case of , continue to combine the check-in time
Figure BDA0003025133080000109
Improve the accuracy of social network influence judgment results.

举例而言，若用户节点u在签到地点

Figure BDA00030251330800001010
的签到时间
Figure BDA00030251330800001011
晚于或者等于用户节点v在签到地点
Figure BDA00030251330800001012
的签到时间
Figure BDA00030251330800001013
则确定两者未存在连边；For example, if user node u is at the check-in location
Figure BDA00030251330800001010
check-in time
Figure BDA00030251330800001011
later than or equal to user node v at the check-in location
Figure BDA00030251330800001012
check-in time
Figure BDA00030251330800001013
Then it is determined that there is no connection between the two;

若用户节点u在签到地点

Figure BDA00030251330800001014
的签到时间
Figure BDA00030251330800001015
早于用户节点v在签到地点
Figure BDA00030251330800001016
的签到时间
Figure BDA00030251330800001017
则确定两者存在连边。If user node u is at the check-in location
Figure BDA00030251330800001014
check-in time
Figure BDA00030251330800001015
earlier than user node v at the check-in location
Figure BDA00030251330800001016
check-in time
Figure BDA00030251330800001017
It is determined that there is a connection between the two.

在判断用户节点u对用户节点v的社交网络影响力时，可限定用户节点u需早于用户节点v到达签到地点

Figure BDA00030251330800001018
在该情况下，用户节点u方可在签到地点
Figure BDA00030251330800001019
对用户节点v产生社交网络影响力，即：When judging the social network influence of user node u on user node v, it can be limited that user node u needs to arrive at the check-in location earlier than user node v
Figure BDA00030251330800001018
In this case, the user node u can be at the check-in location
Figure BDA00030251330800001019
Generate social network influence on user node v, namely:

Figure BDA0003025133080000111
Figure BDA0003025133080000111

若两用户节点均在同一地点签到，且用户节点u的签到时间早于用户节点v的签到时间，则存在连边，表明用户节点u对用户节点v基于某行为产生了一次影响。If both user nodes check in at the same place, and the check-in time of user node u is earlier than the check-in time of user node v, there is an edge, indicating that user node u has an impact on user node v based on a certain behavior.

而若两用户节点在同一地点签到，但用户节点u的签到时间晚于用户节点v的签到时间，或者两用户节点没有在同一地点签到，则不存在连边，影响力为0。If two user nodes check in at the same place, but the check-in time of user node u is later than the check-in time of user node v, or the two user nodes do not check in at the same place, there is no connection and the influence is 0.

在确定存在连边、确定存在影响力后，则可结合本申请构建的多层干扰因素排除机制，进行影响力的具体测量。After it is determined that there is a connection and that there is influence, the specific measurement of influence can be carried out in combination with the multi-layer interference factor elimination mechanism constructed in this application.

具体的，包括干扰因素排除机制包括三层，分别排除用户个人习惯的干扰因素、排除社会大众的干扰因素、排除用户自身社交圈的干扰因素。Specifically, the interference factor exclusion mechanism includes three layers, which respectively exclude interference factors of the user's personal habits, exclude the interference factors of the general public, and exclude the interference factors of the user's own social circle.

在用户个人习惯的干扰因素排除机制中，本申请认为，用户自身的行为习惯也可能造成用户的地点签到行为，比如某用户在单位上班，由于重复特性，他可能经常在此进行地点签到行为，在该情况下，本申请考虑将相同的签到地点c发起的初次地点签到行为作为有效的地点签到行为，即，过滤非首次配置签到地点c的地点签到记录。In the mechanism for eliminating interference factors of the user's personal habits, this application believes that the user's own behavioral habits may also cause the user's location check-in behavior. In this case, the present application considers the initial location check-in behavior initiated by the same check-in location c as a valid location check-in behavior, that is, filtering location check-in records that are not configured for the first time check-in location c.

该过滤，继续以用户节点u为例，根据地点签到记录

Figure BDA0003025133080000112
的签到时间
Figure BDA0003025133080000113
提取出用户节点u在同一签到地点
Figure BDA0003025133080000114
最早的签到记录，然后删去用户节点u签到地点在
Figure BDA0003025133080000115
且签到时间在
Figure BDA0003025133080000116
之后的签到记录，构建得到用户节点u初次地点签到行为记录集合，可表示为：For this filtering, continue to take the user node u as an example, according to the location check-in record
Figure BDA0003025133080000112
check-in time
Figure BDA0003025133080000113
Extract the user node u at the same check-in location
Figure BDA0003025133080000114
The earliest check-in record, and then delete the user node u check-in location at
Figure BDA0003025133080000115
and the check-in time is
Figure BDA0003025133080000116
The subsequent check-in records are constructed to obtain the first-time check-in behavior record set of the user node u, which can be expressed as:

Figure BDA0003025133080000117
Figure BDA0003025133080000117

其中，n表示用户节点u的总签到记录次数。Among them, n represents the total number of check-in records of user node u.

在社会大众的干扰因素排除机制中，本申请认为，有别于用户自身社交圈(所有一阶朋友)的影响，社会大众为用户的非直接朋友，例如，某地一家新网红餐厅受到大众的热情追捧，在“大众点评”等网站的主页上被广泛点评并分享，并被这些网站推荐给用户A，这时如果用户A也到该餐厅打卡，那么就表明A受到了来自社会大众的影响。In the mechanism of eliminating interference factors of the public, this application believes that, different from the influence of the user's own social circle (all first-order friends), the public is the indirect friend of the user. For example, a new Internet celebrity restaurant in a certain place is affected by the public. It is widely commented and shared on the homepages of websites such as "Dianping", and is recommended to user A by these websites. If user A also checks in at the restaurant, it means that A has received favorable comments from the general public. influences.

社会大众的干扰因素，具体可用位置流行度

Figure BDA0003025133080000121
来表示，该位置流行度H是指位置在当前时段受到社会大众追捧的程度，该带来的影响力与社会大众影响因素呈现反比关系。Interference factors of the general public, popularity of specific available locations
Figure BDA0003025133080000121
It means that the popularity H of the location refers to the degree that the location is sought after by the public at the current time period, and the influence brought by it is inversely proportional to the influencing factors of the public.

示例性的，位置流行度

Figure BDA0003025133080000122
基于可以为签到地点c在当前时间段内被所有用户签到的次数占总签到次数的比例，具体可以表示为：Exemplary, location popularity
Figure BDA0003025133080000122
Based on the ratio of the number of times the check-in location c is checked in by all users to the total number of check-ins in the current time period, it can be specifically expressed as:

位置流行度

Figure BDA0003025133080000123
Location popularity
Figure BDA0003025133080000123

δ为时段阈值，

Figure BDA0003025133080000124
表示用户节点v的在签到地点
Figure BDA0003025133080000125
的签到时刻
Figure BDA0003025133080000126
比用户节点u的签到时刻
Figure BDA0003025133080000127
早，||表示集合的大小。δ is the time period threshold,
Figure BDA0003025133080000124
Indicates the check-in location of user node v
Figure BDA0003025133080000125
check-in time
Figure BDA0003025133080000126
than the check-in time of user node u
Figure BDA0003025133080000127
Early, || indicates the size of the collection.

其中，采用的地点签到记录，具体可以为上面提及的过滤非首次配置所述签到地点

Figure BDA0003025133080000128
后的地点签到记录Cu以及过滤非首次配置签到地点
Figure BDA0003025133080000129
后的所述地点签到记录Cv。Among them, the location check-in record used may be the above-mentioned filter non-first-time configuration of the check-in location.
Figure BDA0003025133080000128
Check-in records C u of later locations and filter non-first-time check-in locations
Figure BDA0003025133080000129
The later said location check-in record C v .

而在确定位置流行度

Figure BDA00030251330800001210
后，可根据位置流行度
Figure BDA00030251330800001211
确定用户节点v在签到地点
Figure BDA00030251330800001212
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA00030251330800001213
While determining location popularity
Figure BDA00030251330800001210
After that, the popularity of the location can be
Figure BDA00030251330800001211
Make sure that user node v is at the check-in location
Figure BDA00030251330800001212
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA00030251330800001213

在上文已提及，位置流行度

Figure BDA00030251330800001214
带来的影响力与社会大众影响因素呈现反比关系，而在具体的量化过程中，还可以引入调节系数，以更适宜地量化其带来的影响力，As mentioned above, location popularity
Figure BDA00030251330800001214
The influence brought by it is inversely proportional to the social influence factors, and in the specific quantification process, an adjustment coefficient can also be introduced to more appropriately quantify the influence it brings.

将预设的调节系数f，与位置流行度

Figure BDA00030251330800001215
的比值，作为用户节点v在签到地点
Figure BDA00030251330800001216
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA00030251330800001217
可表示为：Compare the preset adjustment factor f with the location popularity
Figure BDA00030251330800001215
The ratio of , as the user node v at the check-in location
Figure BDA00030251330800001216
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA00030251330800001217
can be expressed as:

Figure BDA00030251330800001218
Figure BDA00030251330800001218

其中，f为调节参数，

Figure BDA00030251330800001219
表示在用户节点v发生相同签到地点
Figure BDA00030251330800001220
签到之前已经处于激活态的朋友用户。Among them, f is the adjustment parameter,
Figure BDA00030251330800001219
Indicates that the same check-in location occurs at user node v
Figure BDA00030251330800001220
Friend users who have been active before signing in.

在用户自身社交圈的干扰因素排除机制中，具体是在确定的用户节点v在签到地点c排除社会影响因素后从朋友用户节点收到的影响力

Figure BDA00030251330800001221
的基础上，继续引入e指数时间衰减模型，并结合softmax函数进行用户节点u对用户节点v的影响力重分配，得到的影响力
Figure BDA00030251330800001222
In the interference factor exclusion mechanism of the user's own social circle, it is specifically the influence received from the friend user node after the determined user node v excludes social influence factors at the check-in location c
Figure BDA00030251330800001221
On the basis of , continue to introduce the e exponential time decay model, and combine the softmax function to redistribute the influence of user node u to user node v, and obtain the influence of
Figure BDA00030251330800001222

本申请认为，当一种地点打卡行为开始在A的朋友圈(直接朋友)里流行，而最后A亦发生了该行为，那么就认为A受到社交圈子的影响，社交圈子由目标用户节点v的多个一阶朋友组成，首先根据用户之间的行为影响力随时间间隔的延长而衰减理论，建立影响力时间衰减模型，可表示为：This application believes that when a location check-in behavior begins to become popular in A's circle of friends (direct friends), and finally A also occurs this behavior, then it is considered that A is affected by the social circle, and the social circle is determined by the target user node v. It consists of multiple first-order friends. First, according to the theory that the behavioral influence between users decays with the extension of time interval, the influence time decay model is established, which can be expressed as:

Figure BDA0003025133080000131
Figure BDA0003025133080000131

P(v|u)为用户节点u当前对用户节点v的影响力，σ为衰减系数，

Figure BDA0003025133080000132
处于0与1之间，P′(v|u)表示经过时间衰减前用户节点u对用户节点v的影响力。P(v|u) is the current influence of user node u on user node v, σ is the attenuation coefficient,
Figure BDA0003025133080000132
Between 0 and 1, P'(v|u) represents the influence of user node u on user node v before time decay.

在影响力时间衰减模型下，对于社交圈子中不同用户节点对目标用户节点的影响力大小，因激活时间间隔不同而不同，根据softmax函数进行u对v的影响力重分配，得到的排除社会大众的干扰因素后用户节点u对用户节点v的影响力

Figure BDA0003025133080000133
可表示为：Under the influence time decay model, the influence of different user nodes in the social circle on the target user node is different due to the different activation time intervals. According to the softmax function, the influence of u to v is redistributed, and the result is obtained by excluding the public. The influence of user node u on user node v after the interference factor of
Figure BDA0003025133080000133
can be expressed as:

Figure BDA0003025133080000134
Figure BDA0003025133080000134

此外，除了考虑到上面提及的三种干扰因素，本申请还考虑了用户在相位上的相似度带来的影响，本申请亦认为该行为相似度带来的影响力，与其本身呈现反比关系。In addition, in addition to taking into account the three interference factors mentioned above, the application also considers the influence of the user's similarity in phase, and the application also considers that the influence of the behavior similarity is inversely proportional to itself .

本申请考虑到行为相似度是随时间变化的，在初期用户的位置行为记录较少，而后期的位置行为记录则较多，这前后的行为相似度计算结果也是不一样的，考虑到行为相似度随时间变化，因此在地点签到记录的基础上，采用改进的杰卡德相似度系数，确定行为相似度Su，v，具体的，可表示为：This application considers that the behavior similarity changes with time, and the user's location behavior records are few in the initial stage, and the location behavior records are more in the later stage. The calculation results of the behavior similarity before and after are also different. Considering that the behavior is similar The degree changes with time, so on the basis of the location check-in record, the improved Jaccard similarity coefficient is used to determine the behavior similarity S u, v , specifically, it can be expressed as:

行为相似度

Figure BDA0003025133080000135
behavioral similarity
Figure BDA0003025133080000135

其中，

Figure BDA0003025133080000136
为用户节点u在到达签到地点c之前的地点签到记录，
Figure BDA0003025133080000137
lv(cv)为用户节点v在到达签到地点c之前的地点签到记录，
Figure BDA0003025133080000138
in,
Figure BDA0003025133080000136
is the check-in record of the user node u before arriving at the check-in location c,
Figure BDA0003025133080000137
l v (c v ) is the check-in record of the user node v before arriving at the check-in place c,
Figure BDA0003025133080000138

结合上面多层干扰因素的考虑，排除用户个人习惯的干扰因素、社会大众的干扰因素、用户自身社交圈的干扰因素以及行为相似度的干扰因素，此时可将将影响力

Figure BDA0003025133080000139
与用户行为相似度Su，v的比值，作为用户节点u对用户节点v的影响力，可表示为：Combined with the consideration of the above multi-layered interference factors, the interference factors of the user's personal habits, the interference factors of the public, the interference factors of the user's own social circle, and the interference factors of the behavior similarity can be excluded.
Figure BDA0003025133080000139
The ratio of user behavior similarity S u, v , as the influence of user node u on user node v, can be expressed as:

Figure BDA0003025133080000141
Figure BDA0003025133080000141

以上是本申请提供社交网络影响力的处理方法的介绍，为便于更好的实施本申请提供的社交网络影响力的处理方法，本申请还提供了社交网络影响力的处理装置。The above is the introduction of the method for processing social network influence provided by the present application. In order to facilitate better implementation of the processing method for social network influence provided by the present application, the present application also provides an apparatus for processing social network influence.

参阅图2，图2为本申请社交网络影响力的处理装置的一种结构示意图，在本申请中，社交网络影响力的处理装置200具体可包括如下结构：Referring to FIG. 2, FIG. 2 is a schematic structural diagram of an apparatus for processing social network influence in the present application. In this application, the apparatus 200 for processing social network influence may specifically include the following structure:

构建单元201，用于构建目标社交网络的初始社交网络数据，其中，初始社交网络数据以无向图G(V，E，C)配置，V为目标社交网络数据中包含的用户节点，E为无向图G(V，E，C)中的边集，C为用户节点的地点签到记录；The construction unit 201 is used for constructing initial social network data of the target social network, wherein the initial social network data is configured with an undirected graph G(V, E, C), V is the user node included in the target social network data, E is The edge set in the undirected graph G(V, E, C), C is the location check-in record of the user node;

确定单元202，用于遍历用户节点u的地点签到记录Cu以及用户节点v的地点签到记录Cv，确定用户节点u与用户节点v存在相同签到地点

Figure BDA0003025133080000142
并基于签到地点
Figure BDA0003025133080000143
确定两者存在连边；The determining unit 202 is used to traverse the location check-in record C u of the user node u and the location check-in record C v of the user node v, and determine that the user node u and the user node v have the same check-in location
Figure BDA0003025133080000142
and based on the check-in location
Figure BDA0003025133080000143
Make sure there is a connection between the two;

过滤单元203，用于在地点签到记录Cu以及地点签到记录Cv中，过滤非首次配置签到地点

Figure BDA0003025133080000144
的地点签到记录；The filtering unit 203 is used to filter the non-first-time configuration check-in locations in the location check-in record C u and the location check-in record C v
Figure BDA0003025133080000144
location check-in record;

确定单元202，还用于计算签到地点

Figure BDA0003025133080000145
的位置流行度
Figure BDA0003025133080000146
并根据位置流行度
Figure BDA0003025133080000147
确定用户节点v在签到地点
Figure BDA0003025133080000148
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000149
The determining unit 202 is also used to calculate the check-in location
Figure BDA0003025133080000145
location popularity
Figure BDA0003025133080000146
and based on location popularity
Figure BDA0003025133080000147
Make sure that user node v is at the check-in location
Figure BDA0003025133080000148
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000149

重分配单元204，用于在影响力

Figure BDA00030251330800001410
的基础上，引入e指数时间衰减模型，并结合softmax函数进行用户节点u对用户节点v的影响力重分配，得到影响力
Figure BDA00030251330800001411
The redistribution unit 204 is used in the influence
Figure BDA00030251330800001410
On the basis of , the e-exponential time decay model is introduced, and the softmax function is used to redistribute the influence of user node u to user node v to obtain influence.
Figure BDA00030251330800001411

确定单元202，还用于根据地点签到记录Cu以及地点签到记录Cv，确定用户节点u与用户节点v之间的用户行为相似度Su，v；The determining unit 202 is further configured to determine the user behavior similarity Su, v between the user node u and the user node v according to the location check-in record C u and the location check-in record C v ;

输出单元205，用于将影响力

Figure BDA00030251330800001412
与用户行为相似度Su，v的比值，作为用户节点u对用户节点v的影响力输出。The output unit 205 is used to convert the influence
Figure BDA00030251330800001412
The ratio of the similarity S u, v to the user behavior is output as the influence of the user node u on the user node v.

在又一种示例性的实现方式中，确定单元202，具体用于：In yet another exemplary implementation, the determining unit 202 is specifically configured to:

若用户节点u在签到地点

Figure BDA00030251330800001413
的签到时间
Figure BDA00030251330800001414
晚于或者等于用户节点v在签到地点
Figure BDA00030251330800001415
的签到时间
Figure BDA00030251330800001416
则确定两者未存在连边；If user node u is at the check-in location
Figure BDA00030251330800001413
check-in time
Figure BDA00030251330800001414
later than or equal to user node v at the check-in location
Figure BDA00030251330800001415
check-in time
Figure BDA00030251330800001416
Then it is determined that there is no connection between the two;

若用户节点u在签到地点

Figure BDA0003025133080000151
的签到时间
Figure BDA0003025133080000152
早于用户节点v在签到地点
Figure BDA0003025133080000153
的签到时间
Figure BDA0003025133080000154
则确定两者存在连边。If user node u is at the check-in location
Figure BDA0003025133080000151
check-in time
Figure BDA0003025133080000152
earlier than user node v at the check-in location
Figure BDA0003025133080000153
check-in time
Figure BDA0003025133080000154
It is determined that there is a connection between the two.

在又一种示例性的实现方式中，确定单元202，具体用于：In yet another exemplary implementation, the determining unit 202 is specifically configured to:

基于过滤非首次配置签到地点

Figure BDA0003025133080000155
后的地点签到记录Cu以及过滤非首次配置签到地点
Figure BDA0003025133080000156
后的地点签到记录Cv，确定签到地点
Figure BDA0003025133080000157
在当前时间段内被所有用户签到的次数占总签到次数的比例，作为位置流行度
Figure BDA0003025133080000158
Based on filtering non-first-time configuration check-in locations
Figure BDA0003025133080000155
Check-in records C u of later locations and filter non-first-time check-in locations
Figure BDA0003025133080000156
Check-in record C v at the later location to determine the check-in location
Figure BDA0003025133080000157
The ratio of the number of check-ins by all users to the total number of check-ins in the current time period is used as the location popularity
Figure BDA0003025133080000158

位置流行度

Figure BDA0003025133080000159
Location popularity
Figure BDA0003025133080000159

δ为时段阈值，

Figure BDA00030251330800001510
表示用户节点v的在签到地点
Figure BDA00030251330800001511
的签到时刻
Figure BDA00030251330800001512
比用户节点u的签到时刻
Figure BDA00030251330800001513
早，||表示集合的大小。δ is the time period threshold,
Figure BDA00030251330800001510
Indicates the check-in location of user node v
Figure BDA00030251330800001511
check-in time
Figure BDA00030251330800001512
than the check-in time of user node u
Figure BDA00030251330800001513
Early, || indicates the size of the collection.

在又一种示例性的实现方式中，确定单元202，具体用于：In yet another exemplary implementation, the determining unit 202 is specifically configured to:

将预设的调节系数f，与位置流行度

Figure BDA00030251330800001514
的比值，作为用户节点v在签到地点
Figure BDA00030251330800001515
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA00030251330800001516
Compare the preset adjustment factor f with the location popularity
Figure BDA00030251330800001514
The ratio of , as the user node v at the check-in location
Figure BDA00030251330800001515
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA00030251330800001516

在又一种示例性的实现方式中，确定单元202，具体用于：In yet another exemplary implementation, the determining unit 202 is specifically configured to:

根据地点签到记录Cu以及地点签到记录Cv，结合改进的杰卡德相似度系数，确定行为相似度Su，v，According to the location check-in record C u and the location check-in record C v , combined with the improved Jaccard similarity coefficient, determine the behavior similarity Su , v ,

行为相似度

Figure BDA00030251330800001517
behavioral similarity
Figure BDA00030251330800001517

Figure BDA00030251330800001518
为用户节点u在到达签到地点
Figure BDA00030251330800001519
之前的地点签到记录，lv(cv)为用户节点v在到达签到地点
Figure BDA00030251330800001520
之前的地点签到记录。
Figure BDA00030251330800001518
for the user node u to arrive at the check-in location
Figure BDA00030251330800001519
The previous location check-in record, l v (c v ) is the user node v arriving at the check-in location
Figure BDA00030251330800001520
Previous location check-in records.

在又一种示例性的实现方式中，e指数时间衰减模型的表达式为：In yet another exemplary implementation, the expression of the e-exponential time decay model is:

Figure BDA00030251330800001521
Figure BDA00030251330800001521

P(v|u)为用户节点u当前对用户节点v的影响力，σ为衰减系数，

Figure BDA00030251330800001522
处于0与1之间，P′(v|u)表示经过时间衰减前用户节点u对用户节点v的影响力。P(v|u) is the current influence of user node u on user node v, σ is the attenuation coefficient,
Figure BDA00030251330800001522
Between 0 and 1, P'(v|u) represents the influence of user node u on user node v before time decay.

在又一种示例性的实现方式中，影响力

Figure BDA00030251330800001523
的表达式为：In yet another exemplary implementation, the influence
Figure BDA00030251330800001523
The expression is:

Figure BDA00030251330800001524
Figure BDA00030251330800001524

本申请还提供了处理设备，参阅图3，图3示出了本申请处理设备的一种结构示意图，具体的，本申请处理设备可包括处理器301、存储器302以及输入输出设备303，处理器301用于执行存储器302中存储的计算机程序时实现如图1对应实施例中社交网络影响力的处理方法的各步骤；或者，处理器301用于执行存储器302中存储的计算机程序时实现如图2对应实施例中各单元的功能，存储器302用于存储处理器301执行上述图1对应实施例中社交网络影响力的处理方法所需的计算机程序。The present application also provides a processing device, please refer to FIG. 3, which 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 and output device 303. The processor 301 is used to implement each step of the method for processing social network influence in the embodiment corresponding to FIG. 1 when executing the computer program stored in the memory 302; or, when the processor 301 is used to execute the computer program stored in the memory 302, it is implemented as shown in FIG. 2 Corresponding to the functions of each unit in the embodiment, the memory 302 is used for storing the computer program required by the processor 301 to execute the method for processing social network influence in the embodiment corresponding to FIG. 1 .

示例性的，计算机程序可以被分割成一个或多个模块/单元，一个或者多个模块/单元被存储在存储器302中，并由处理器301执行，以完成本申请。一个或多个模块/单元可以是能够完成特定功能的一系列计算机程序指令段，该指令段用于描述计算机程序在计算机装置中的执行过程。Exemplarily, the computer program may be divided into one or more modules/units, and the one or more modules/units are stored in the memory 302 and executed by the processor 301 to complete the present application. One or more modules/units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in a computer apparatus.

处理设备可包括，但不仅限于处理器301、存储器302、输入输出设备303。本领域技术人员可以理解，示意仅仅是处理设备的示例，并不构成对处理设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如处理设备还可以包括网络接入设备、总线等，处理器301、存储器302、输入输出设备303以及网络接入设备等通过总线相连。Processing devices may include, but are not limited to, processor 301 , memory 302 , and input and output devices 303 . Those skilled in the art can understand that the illustration is only an example of a processing device, and does not constitute a limitation on the processing device, and may include more or less components than the one shown in the figure, or combine some components, or different components, such as processing The device may also include a network access device, a bus, etc., and the processor 301, the memory 302, the input and output device 303, and the network access device, etc. are connected through the bus.

处理器301可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，处理器是处理设备的控制中心，利用各种接口和线路连接整个设备的各个部分。The processor 301 may be a central processing unit (Central Processing Unit, CPU), other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), an off-the-shelf processor Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The processor is the control center of the processing device, and uses various interfaces and lines to connect various parts of the entire device.

存储器302可用于存储计算机程序和/或模块，处理器301通过运行或执行存储在存储器302内的计算机程序和/或模块，以及调用存储在存储器302内的数据，实现计算机装置的各种功能。存储器302可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序等；存储数据区可存储根据处理设备的使用所创建的数据等。此外，存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如硬盘、内存、插接式硬盘，智能存储卡(Smart Media Card，SMC)，安全数字(SecureDigital，SD)卡，闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。The memory 302 can 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 calling data stored in the memory 302. The memory 302 may mainly include a storage program area and a storage data area, wherein the storage program 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 and the like created according to the use of the processing device. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, internal memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.

处理器301用于执行存储器302中存储的计算机程序时，具体可实现以下功能：When the processor 301 is configured to execute the computer program stored in the memory 302, the following functions can be specifically implemented:

遍历用户节点u的地点签到记录Cu以及用户节点v的地点签到记录Cv，确定用户节点u与用户节点v存在相同签到地点

Figure BDA0003025133080000171
并基于签到地点
Figure BDA0003025133080000172
确定两者存在连边；Traverse the location check-in record C u of the user node u and the location check-in record C v of the user node v, and determine that the user node u and the user node v have the same check-in location
Figure BDA0003025133080000171
and based on the check-in location
Figure BDA0003025133080000172
Make sure there is a connection between the two;

在地点签到记录Cu以及地点签到记录Cv中，过滤非首次配置签到地点

Figure BDA0003025133080000173
的地点签到记录；In the location check-in record C u and the location check-in record C v , filter the non-first-time configuration check-in location
Figure BDA0003025133080000173
location check-in record;

计算签到地点

Figure BDA0003025133080000174
的位置流行度
Figure BDA0003025133080000175
并根据位置流行度
Figure BDA0003025133080000176
确定用户节点v在签到地点
Figure BDA0003025133080000177
排除社会影响因素后从朋友用户节点收到的影响力
Figure BDA0003025133080000178
Calculate check-in location
Figure BDA0003025133080000174
location popularity
Figure BDA0003025133080000175
and based on location popularity
Figure BDA0003025133080000176
Make sure user node v is at the check-in location
Figure BDA0003025133080000177
Influence received from friend user nodes after exclusion of social influence factors
Figure BDA0003025133080000178

在影响力

Figure BDA0003025133080000179
的基础上，引入e指数时间衰减模型，并结合softmax函数进行用户节点u对用户节点v的影响力重分配，得到影响力
Figure BDA00030251330800001710
in influence
Figure BDA0003025133080000179
On the basis of , the e-exponential time decay model is introduced, and the softmax function is used to redistribute the influence of user node u to user node v to obtain influence.
Figure BDA00030251330800001710

根据地点签到记录Cu以及地点签到记录Cv，确定用户节点u与用户节点v之间的用户行为相似度Su，v；According to the location check-in record C u and the location check-in record C v , determine the user behavior similarity Su,v between the user node u and the user node v ;

将影响力

Figure BDA00030251330800001711
与用户行为相似度Su，v的比值，作为用户节点u对用户节点v的影响力输出。will influence
Figure BDA00030251330800001711
The ratio of the similarity S u, v to the user behavior is output as the influence of the user node u on the user node v.

所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的社交网络影响力的处理装置、处理设备及其相应单元的具体工作过程，可以参考如图1对应实施例中社交网络影响力的处理方法的说明，具体在此不再赘述。Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the social network influence processing device, processing equipment and corresponding units described above may refer to the social network in the corresponding embodiment of FIG. 1 . The description of the method for processing network influence will not be repeated here.

本领域普通技术人员可以理解，上述实施例的各种方法中的全部或部分步骤可以通过指令来完成，或通过指令控制相关的硬件来完成，该指令可以存储于一计算机可读存储介质中，并由处理器进行加载和执行。Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above-mentioned embodiments can be completed by instructions, or by instructions that control relevant hardware, and the instructions can be stored in a computer-readable storage medium, and loaded and executed by the processor.

为此，本申请提供一种计算机可读存储介质，其中存储有多条指令，该指令能够被处理器进行加载，以执行本申请如图1对应实施例中社交网络影响力的处理方法中的步骤，具体操作可参考如图1对应实施例中社交网络影响力的处理方法的说明，在此不再赘述。To this end, the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the processing method of social network influence in the embodiment corresponding to FIG. 1 of the present application. Steps, and specific operations can refer to the description of the processing method of social network influence in the embodiment corresponding to FIG. 1 , which will not be repeated here.

其中，该计算机可读存储介质可以包括：只读存储器(Read Only Memory，ROM)、随机存取记忆体(Random Access Memory，RAM)、磁盘或光盘等。Wherein, the computer-readable storage medium may include: a read only memory (Read Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and the like.

由于该计算机可读存储介质中所存储的指令，可以执行本申请如图1对应实施例社交网络影响力的处理方法中的步骤，因此，可以实现本申请如图1对应实施例中社交网络影响力的处理方法所能实现的有益效果，详见前面的说明，在此不再赘述。Due to the instructions stored in the computer-readable storage medium, the steps in the method for processing social network influence in the embodiment corresponding to FIG. 1 of the present application can be executed. Therefore, the social network influence in the embodiment corresponding to FIG. 1 of the present application can be realized. The beneficial effects that can be achieved by the force processing method are detailed in the foregoing description, and will not be repeated here.

以上对本申请提供的社交网络影响力的处理方法、装置、处理设备以及计算机可读存储介质进行了详细介绍，本文中应用了具体个例对本申请的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本申请的方法及其核心思想；同时，对于本领域的技术人员，依据本申请的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本申请的限制。The processing method, device, processing device, and computer-readable storage medium for social network influence provided by the present application have been described in detail above. The principles and implementations of the present application are described with specific examples. The description is only used to help understand the method of the present application and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, there will be changes in the specific embodiment and the scope of application. In summary, The contents of this specification should not be construed as limiting the 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 place
Figure FDA0003025133070000011
And based on the check-in place
Figure FDA0003025133070000012
Determining that the two have a connecting edge;
check-in record C at the placeuAnd the place check-in record CvIn filtering the check-in place
Figure FDA0003025133070000013
The location check-in record;
calculating the check-in location
Figure FDA0003025133070000014
Location popularity of
Figure FDA0003025133070000015
And according to the position popularity
Figure FDA0003025133070000016
Determining that the user node v is at the check-in place
Figure FDA0003025133070000017
Influence received from friend user node after social influence factor is eliminated
Figure FDA0003025133070000018
Under the influence of
Figure FDA0003025133070000019
On 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
Figure FDA00030251330700000110
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；
Will influence the force
Figure FDA00030251330700000111
Similarity to the user behavior Su,vIs used as the influence output of the user node u on the user node v.
2. The method of claim 1, wherein the checking-in based on the check-in location
Figure FDA00030251330700000112
Determining that the two have the connecting edge, including:
if the user node u is at the check-in place
Figure FDA00030251330700000113
Time of attendance
Figure FDA00030251330700000114
Later than or equal to the check-in place of the user node v
Figure FDA00030251330700000115
Time of attendance
Figure FDA00030251330700000116
Determining that the two have no connecting edge;
if the user node u is at the check-in place
Figure FDA00030251330700000117
Time of attendance
Figure FDA00030251330700000118
Earlier than the user node v at the check-in place
Figure FDA00030251330700000119
Time of attendance
Figure FDA00030251330700000120
It is determined that there is a continuous edge between the two.
3. The method of claim 1, wherein calculating the location popularity of the check-in place c
Figure FDA00030251330700000121
The method comprises the following steps:
sign-in place based on filtering non-first configuration
Figure FDA00030251330700000122
The post place check-in record CuAnd filtering the non-first configuration check-in places
Figure FDA00030251330700000123
The post place check-in record CvDetermining the check-in location
Figure FDA00030251330700000124
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 is used as the position popularity
Figure FDA0003025133070000021
The location popularity
Figure FDA0003025133070000022
Delta is the time period threshold value and is,
Figure FDA0003025133070000023
representing said user node v at said check-in place
Figure FDA0003025133070000024
Time of attendance
Figure FDA0003025133070000025
The check-in time of the user node u
Figure FDA0003025133070000026
Early, | | represents the size of the set.
4. The method of claim 1, wherein the popularity is based on the location
Figure FDA0003025133070000027
Determining that the user node v is at the check-in place
Figure FDA0003025133070000028
Influence received from friend user node after social influence factor is eliminated
Figure FDA0003025133070000029
The method comprises the following steps:
the preset adjusting coefficient f is compared with the position popularity
Figure FDA00030251330700000210
As the ratio of said user node v at said check-in location
Figure FDA00030251330700000211
Influence received from friend user node after social influence factor is eliminated
Figure FDA00030251330700000212
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，
The behavioral similarity
Figure FDA00030251330700000213
Figure FDA00030251330700000214
For the user node u to arrive at the check-in place
Figure FDA00030251330700000215
Previous location check-in record, |v(cv) For the user node v arriving at the check-in place
Figure FDA00030251330700000216
Previous location check-in records.
6. The method of claim 1, wherein the e-exponential time decay model is expressed by:
Figure FDA00030251330700000217
p (v | u) is the current influence of the user node u on the user node v, σ is an attenuation coefficient,
Figure FDA00030251330700000218
between 0 and 1, P' (v | u) represents the influence of the user node u on the user node v before the time decay.
7. The method of claim 1, wherein the influencing force is
Figure FDA00030251330700000219
The expression of (a) is:
Figure FDA0003025133070000031
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 place
Figure FDA0003025133070000032
And based on the check-in place
Figure FDA0003025133070000033
Determining 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 place
Figure FDA0003025133070000034
The location check-in record;
the determining unit is further used for calculating the check-in place
Figure FDA0003025133070000035
Location popularity of
Figure FDA0003025133070000036
And according to the position popularity
Figure FDA0003025133070000037
Determining that the user node v is at the check-in place
Figure FDA0003025133070000038
Removing social influence factorsInfluence received by friend user node
Figure FDA0003025133070000039
A redistribution unit for redistributing the influence
Figure FDA00030251330700000310
On 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
Figure FDA00030251330700000311
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；
An output unit for outputting the influence
Figure FDA00030251330700000312
Similarity to the user behavior Su,vIs used as the influence output of the user node u on the user node 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.
CN202110414260.3A 2021-04-16 2021-04-16 Processing method, device and processing equipment for social network influence Withdrawn CN113239285A (en)

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)

* Cited by examiner, † Cited by third party
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 factorization POI recommendation method based on geographic location 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

Patent Citations (6)

* Cited by examiner, † Cited by third party
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 factorization POI recommendation method based on geographic location 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
CN104281882B (en) The method and system of prediction social network information stream row degree based on user characteristics
CN104809408B (en) A kind of histogram dissemination method based on difference privacy
CN105335409B (en) A kind of determination method, equipment and the network server of target user
US10243967B2 (en) Method, apparatus and system for detecting fraudulant software promotion
CN103970879B (en) Method and system for regulating storage positions of data blocks
CN108834077B (en) Tracking area division method, device and electronic device based on user movement characteristics
CN108920601B (en) A data matching method and device
WO2021047021A1 (en) Information mining method and apparatus, device, and storage medium
WO2020088262A1 (en) Data analysis method and device, and storage medium
CN110516752A (en) Cluster quality evaluation method, device and equipment and storage medium
JP2023535851A (en) METHOD, DEVICE, TERMINAL DEVICE, AND STORAGE MEDIUM FOR DATA PROCESSING MODEL BY PRIVACY PROTECTION
CN114120221A (en) Environment checking method based on deep learning, electronic equipment and storage medium
CN114281648A (en) A data collection method, device, electronic device and storage medium
CN107659430B (en) A kind of Node Processing Method, device, electronic equipment and computer storage medium
CN111966842A (en) Multimedia file pushing method, device and equipment
CN111475540A (en) A method and device for generating a user relationship network
CN104572687B (en) The key user's recognition methods and device that microblogging is propagated
CN114546360A (en) Code program recommendation method and device, electronic equipment and storage medium
CN113239285A (en) Processing method, device and processing equipment for social network influence
CN115098795B (en) Data analysis method, data analysis device, equipment and storage medium
CN105243122A (en) Social software based data acquisition method and apparatus
CN107426133B (en) Method and device for identifying user identity information
CN109241419B (en) ID data network data analysis method, device and computing equipment
CN114429177A (en) Equipment fingerprint feature screening method and device, electronic equipment and storage medium
CN109241421B (en) ID data network processing method, device, computing device and computer storage medium

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20210810