CN111008316B - Clue collection system - Google Patents
Clue collection system Download PDFInfo
- Publication number
- CN111008316B CN111008316B CN201911342783.0A CN201911342783A CN111008316B CN 111008316 B CN111008316 B CN 111008316B CN 201911342783 A CN201911342783 A CN 201911342783A CN 111008316 B CN111008316 B CN 111008316B
- Authority
- CN
- China
- Prior art keywords
- clue
- graph
- information
- algorithm
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a thread collection system comprising: the initial clue storage module is used for storing clue information provided by each person, wherein the clue information comprises personnel information of each person, initial clues and alarm information; the graph algorithm module is connected with the initial clue storage module and comprises a plurality of graph algorithm units, each graph algorithm unit is respectively provided with a graph algorithm model in advance, and each graph algorithm model is respectively used for processing according to clue information to obtain a corresponding clue grid graph; the graph algorithm module is used for outputting all the clue network graphs; the clue mining module is connected with the graph algorithm module, a plurality of clue mining models are preset in the clue mining module, each clue mining model corresponds to one type of clue network graph, and the clue mining module is used for inputting the clue network graph of the corresponding type into the corresponding clue mining model so as to obtain a target clue. The invention has the beneficial effects that: the method realizes deep fusion of the graph calculation technology and the clue discovery in the public security field.
Description
Technical Field
The invention relates to the technical field of urban safety, in particular to a clue collecting system.
Background
At present, a great deal of data accumulated by public security authorities have the problems of dispersion and disorder, such as a great deal of case records, citizen identities, social activities, lodging and the like, which cause the incidence relation of the data and hidden effective clues to be not found and utilized efficiently.
Therefore, in the prior art, a large amount of data accumulated by public security authorities are processed by adopting thread discovery software, for example, an 'eye' data visualization cognitive analysis platform of an intelligent cloud company, a 'astronomical network' distributed big data cognitive analysis platform, a 'honeycomb' knowledge graph database of an explicit data company and the like, and the thread discovery software mainly has the functions of assisting the public security authorities to collect, store, manage and analyze related data, and generally supports searching character relation network threads with a certain artificial core, and performing visual analysis and research on a social relation network of people, things, places and objects, so that deep threads cannot be mined.
In addition, the clue discovery system adopted in the prior art sacrifices the pertinence of specific service scenes, so that the deep value of data cannot be mined in practical application. In the prior art, mainly the management, storage, display and simple association analysis of data are focused on, the business characteristics in the public security field are not considered, the method has strong universality, and in the specific case data analysis process, deep knowledge cannot be mined from the data, namely hidden clues cannot be provided for civil police. In addition, the construction of the graph calculation model on the basis of the distributed graph database belongs to a frontier technology, large data technologies based on Hadoop, apache TinkerPop, spark and the like are needed, and particularly, performance needs to be further researched, and mature products are lacking in the field at present.
Disclosure of Invention
In view of the foregoing problems in the prior art, a thread collection system is now provided that aims to deeply fuse graph computing technology with thread discovery in the public security domain.
The specific technical scheme is as follows:
a thread collection system, comprising:
the initial clue storage module is used for storing clue information provided by each person, wherein the clue information comprises personnel information of each person, initial clues provided by each person and alarm information;
the graph algorithm module is connected with the initial clue storage module and comprises a plurality of graph algorithm units, each graph algorithm unit is respectively provided with a graph algorithm model in advance, and each graph algorithm model is respectively used for processing according to clue information to obtain a corresponding clue grid graph;
the graph algorithm module is used for outputting all the clue network graphs;
the clue mining module is connected with the graph algorithm module, a plurality of clue mining models are preset in the clue mining module, each clue mining model corresponds to one type of clue network graph, and the clue mining module is used for inputting the clue network graph of the corresponding type into the corresponding clue mining model so as to obtain a target clue.
Preferably, the thread collection system, wherein the graph algorithm module comprises:
the link map algorithm unit is preset with a first map algorithm model which is used for carrying out algorithm processing according to the clue information to obtain a link map;
the link map comprises a plurality of link nodes obtained by clue information, and every two link nodes with link relations are connected.
Preferably, the thread collection system, wherein the thread mining module comprises:
and the link connection discovery model is connected with the link connection graph algorithm unit and is used for inputting the link connection graph into the link connection discovery model, acquiring a target clue according to the connection relation between each link connection node, wherein the target clue is an intermediate link in the hidden link connection.
Preferably, the thread collection system, wherein the thread mining module comprises:
the illicit person discovery model is connected with the link map algorithm unit and is used for inputting the link map into the illicit person discovery model, acquiring a target clue according to the connection relation between each link node, wherein the target clue is a link node with the connection relation of a plurality of link nodes being a circulating relation.
Preferably, the thread collection system, wherein the graph algorithm module comprises:
the central graph algorithm unit is preset with a second graph algorithm model which is used for carrying out algorithm processing according to the clue information to obtain a shortest associated path graph;
the shortest association path diagram comprises a shortest association path between every two person relationship nodes which are not directly connected, wherein the shortest association path comprises a plurality of person relationship nodes obtained by personnel information and/or initial clues and/or alarm information, and every two person relationship nodes with association relationship are connected.
Preferably, the thread collection system, wherein the thread mining module comprises:
the center character acquisition model is connected with the center graph algorithm unit and is used for inputting the shortest association path graph into the center character acquisition model, calculating corresponding association coefficients according to the times of character relation nodes passing by each shortest association path, and acquiring target clues according to all the association coefficients.
Preferably, the thread collection system, wherein the graph algorithm module comprises:
the community algorithm unit is preset with a third graph algorithm model, and the third graph algorithm model is used for carrying out algorithm processing according to the clue information to obtain a relationship graph between the characters and the community;
the relationship diagram between the person and the community comprises a plurality of person relationship nodes obtained by personnel information and/or initial clues and/or alarm information, and each person relationship node is provided with corresponding community information.
Preferably, the thread collection system, wherein the thread mining module comprises:
the community acquisition model is connected with the community algorithm unit and is used for inputting a relationship diagram between people and communities into the central person acquisition model, calculating the same quantity of community information according to community information corresponding to each person relationship node to acquire target clues, wherein the target clues are the community information exceeding the preset quantity.
Preferably, the thread collection system, wherein the graph algorithm module comprises:
the social circle algorithm unit is pre-provided with a fourth graph algorithm model, and the fourth graph algorithm model is used for carrying out algorithm processing according to the clue information to obtain a social graph of each personnel information;
the social graph comprises person relation nodes which are obtained by the personnel information and/or the initial clues and/or the alarm information and are associated with the corresponding personnel information.
Preferably, the thread collection system, wherein the thread mining module comprises:
the core character acquisition model is connected with the social circle algorithm unit and is used for inputting the social graph of each piece of personal information into the core character acquisition model, calculating the core coefficient of each person relation node according to the association relation between each person relation node in the social graph of each piece of personal information, and acquiring the target clue according to all the core coefficients.
The technical scheme has the following advantages or beneficial effects: the clue information is input into a graph algorithm model in a corresponding graph algorithm unit through a graph algorithm module to obtain a corresponding clue network graph, and then the clue network graph is input into a corresponding clue mining model to obtain a target clue, so that the depth fusion of a graph computing technology and clue discovery in the public security field is realized.
Drawings
Embodiments of the present invention will now be described more fully with reference to the accompanying drawings. The drawings, however, are for illustration and description only and are not intended as a definition of the limits of the invention.
FIG. 1 is a schematic diagram of a thread collection system according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a second embodiment of a thread collection system according to the present invention;
FIG. 3 is a diagram of a link diagram of an embodiment of a thread collection system according to the present invention;
FIG. 4 is a diagram of a link diagram of a thread collection system according to an embodiment of the present invention;
FIG. 5 is a shortest association path diagram of an embodiment of a thread collection system according to the present invention;
FIG. 6 is a social graph of an embodiment of the thread collection system of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be noted that, without conflict, the embodiments of the present invention and features of the embodiments may be combined with each other.
The invention is further described below with reference to the drawings and specific examples, which are not intended to be limiting.
The present invention includes a thread collection system, as shown in FIG. 1, comprising:
the initial clue storage module 1 is used for storing clue information provided by each person, wherein the clue information comprises personnel information of each person, initial clues provided by each person and alarm information;
the graph algorithm module 2 is connected with the initial clue storage module 1 and comprises a plurality of graph algorithm units, each graph algorithm unit is respectively provided with a graph algorithm model in advance, and each graph algorithm model is respectively used for processing according to clue information to obtain a corresponding clue grid graph;
the graph algorithm module 2 is used for outputting all the clue network graphs;
the clue mining module 3 is connected with the graph algorithm module 2, a plurality of clue mining models are preset in the clue mining module 3, each clue mining model corresponds to one type of clue network graph, and the clue mining module 3 is used for inputting the clue network graph of the corresponding type into the corresponding clue mining model so as to obtain a target clue.
In the above embodiment, the corresponding clue network diagram is obtained by the graph algorithm module 2 according to the graph algorithm model in the corresponding graph algorithm unit, and then the clue network diagram is input into the corresponding clue mining model to obtain the target clue, so as to realize the deep fusion of the graph computing technology and the clue discovery in the public security domain.
In the above embodiment, the corresponding clue network diagram is obtained by the graph algorithm unit in the graph algorithm module 2, and the clue network diagram is input into the clue mining model corresponding to the graph algorithm unit, so that the mapping between the graph algorithm unit and the clue mining model is realized. The police personnel can apply a more complex graph algorithm in the actual case analysis to find deeper case clues.
In this case, as a preferred embodiment, the algorithm in the graph algorithm unit can be implemented using gremlin-python, so that the graph algorithm unit can support infinite expansion.
As a preferred embodiment, the initial clue storage module 1 may be a graph database, and the personnel information, the initial clue, and the alarm information are stored in the graph database.
Further, in the above embodiment, as shown in fig. 2, the graph algorithm module 2 includes:
the link map association algorithm unit 201 is preset with a first map algorithm model, and the first map algorithm model is used for carrying out algorithm processing according to personnel information and/or initial clues and/or alarm information in clue information to obtain a link map association;
the link map comprises a plurality of link nodes which are obtained by personnel information and/or initial clues and/or alarm information in clue information, and every two link nodes with link relations are connected.
As a preferred embodiment, the contact link map algorithm unit 201 may include:
the association clue extraction component is used for providing various association ways which occur in personnel information and/or initial clues and/or alarm information in clue information, wherein the association ways comprise QQ numbers, micro signals, various mailbox numbers and mobile phone numbers;
the association node establishing component is connected with the association clue extracting component and is used for taking the association account number in each type of association way as an association node;
and the contact link establishing component is connected with the contact node establishing component and is used for connecting two contact nodes which are mutually connected to create a contact link diagram.
Further, in the above embodiment, as shown in fig. 2, the thread mining module 3 includes:
the link discovery model 301 is connected to the link map algorithm unit 201, and is configured to input a link map into the link discovery model 301, and obtain a target clue according to a connection relationship between each link node, where the target clue is an intermediate contact in the hidden link.
In the above embodiment, the hidden contact link is the shortest contact path between two contact nodes, and the shortest contact path is the shortest contact path formed by two contact nodes, that is, the contact node between the contact paths is the least.
As a preferred embodiment, as shown in fig. 3, there may be no direct contact between two cheating persons 11 and 12 related to the same case, but a hidden contact link hidden by the cheating person 11 and the cheating person 12 can be found by calculating the shortest path between them, so as to obtain an intermediate contact in the hidden contact link;
for example, there are a contact node 13, a contact node 14, a contact node 15, a contact node 16 and a contact node 17 between the fraudster 11 and the fraudster 12, wherein the fraudster 11 and the contact node 13 are in bidirectional contact, the fraudster 11 and the contact node 14 are in bidirectional contact, the contact node 13 and the contact node 15 are in bidirectional contact, the contact node 15 and the contact node 16 are in bidirectional contact, the contact node 15 and the contact node 17 are in bidirectional contact, the contact node 16 and the contact node 17 are in bidirectional contact, the contact node 13 is in unidirectional contact with the contact node 16, the contact node 16 is in unidirectional contact with the contact node 14, and the contact node 17 is in unidirectional contact with the fraudster 12;
obtaining the shortest path between fraudster 11 and fraudster 12 includes:
first, fraudster 11 to contact node 13 to contact node 15 to contact node 17 to fraudster 12;
second, fraudster 11 to contact node 13 to contact node 16 to contact node 17 to fraudster 12;
thus, the intermediate contacts in the hidden contact link may be one or more of contact node 13, contact node 15, contact node 16 and contact node 17, which may then be further confirmed.
Wherein the contact node 14 is not in the hidden contact link, but the contact node 14 appears in the contact link diagram and the contact node 14 is contacted by both the fraudster 11 and the contact node 16, respectively, so that the contact node 14 is also likely to be an intermediate contact in the hidden contact link.
Further, in the above embodiment, as shown in fig. 2, the thread mining module 3 includes:
the illicit person discovery model 302 is connected to the link map algorithm unit 201, and is configured to input the link map into the illicit person discovery model 302, obtain a target clue according to a connection relationship between each of the link nodes, where the target clue is a link node whose connection relationship between a plurality of link nodes is a cyclic relationship.
In the above embodiment, as shown in fig. 4, the connection relationship of the target clues for the plurality of connection nodes is ase:Sub>A connection node of the circulation relationship, including the circulation relationship of "ase:Sub>A-B-ase:Sub>A", for example, two cheating persons 21 and 22 related to the same case are in upper and lower level connection, and possible other upper and lower level persons can be found through the illegal person discovery model 302; for example, fraudster 21 and fraudster 22 are interconnected, fraudster 22 and contact node 23 are interconnected, fraudster 22 and contact node 24 are interconnected, contact node 24 and contact node 25 are interconnected, contact node 25 is interconnected with contact node 26 and contact node 27, respectively, so that fraudster 21, fraudster 22 and contact node 23 form an "A-B-A" contact, contact node 25, contact node 26 and contact node 27 form an "A-B-A" contact, and fraudster 22, contact node 24 and contact node 25 also form an "A-B-A" contact; namely contact node 23, contact node 24, contact node 25, contact node 26 and contact node 27 are possible illicit persons.
Further, in the above embodiment, as shown in fig. 2, the graph algorithm module 2 includes:
the central graph algorithm unit 202 is pre-provided with a second graph algorithm model, wherein the second graph algorithm model is used for carrying out algorithm processing according to personnel information and/or initial clues and/or alarm information in clue information to obtain a shortest associated path graph;
the shortest association path diagram comprises a shortest association path between every two person relationship nodes which are not directly connected, wherein the shortest association path comprises a plurality of person relationship nodes obtained by personnel information and/or initial clues and/or alarm information in clue information, and every two person relationship nodes with association relationship are connected.
As a preferred embodiment, as shown in fig. 2, the centrogram algorithm unit 202 may include:
the central clue extraction component is used for providing personnel information in clue information and/or various associated information appearing in initial clues and/or alarm information, and the associated information comprises person relations and relation ways;
the center node establishing component is connected with the center cue extracting component and is used for taking each piece of associated information in each type of associated information as a person relationship node;
the central link establishing component is connected with the central node establishing component and is used for connecting the two associated information which are associated with each other and obtaining the shortest associated path between every two non-direct-connection person relationship nodes so as to form a shortest associated path diagram through each shortest associated path.
Further, in the above embodiment, as shown in fig. 2, the thread mining module 3 includes:
the center character acquisition model 303 is connected to the center graph algorithm unit 202, and is configured to input the shortest association path graph into the center character acquisition model 303, calculate corresponding association coefficients according to the number of times each shortest association path passes through the character relationship nodes, and acquire target cues according to all the association coefficients.
As a preferred embodiment, the central persona acquisition model 303 may calculate the number of connection times of each persona relationship node according to the number of times of the persona relationship node passing through the shortest association path, and calculate the association coefficient of the corresponding persona relationship node according to the number of connection times. When the association coefficient of a person relationship node is higher, the association relationship with the person relationship node is more, namely, the connection line of the person relationship node is more. It should be noted that the person corresponding to the person relationship node with higher association coefficient is the necessary person for establishing the connection of the social circle, which is a key clue of case investigation.
For example, as shown in fig. 5, the shortest association path a includes a person relationship node 311, a person relationship node 312, a person relationship node 313, a person relationship node 314, and a person relationship node 315;
the shortest association path B includes a personal relationship node 321, a personal relationship node 312, a personal relationship node 322, a personal relationship node 323, and a personal relationship node 324;
the shortest association path C includes a personal relationship node 331, a personal relationship node 312, a personal relationship node 322, a personal relationship node 333, and a personal relationship node 334;
the shortest association path a, the shortest association path B and the shortest association path C all pass through the person relationship node 312, and the relationship number of the person relationship node 312 can be recorded as 3;
the shortest association path B and the shortest association path C both pass through the person relationship node 322, and the relationship number of the person relationship node 322 can be recorded as 2;
thus, the persona relationship node 312 and persona relationship node 322 are the necessary personas for the social circle to establish a connection, and the persona relationship node 312 is more important because the association coefficient of persona relationship node 312 is higher.
In the above embodiment, the central persona acquisition model 303 may employ Betweenness Centrality algorithm, so the association coefficient may be Betweenness Centrality value, and the magnitude of Betweenness Centrality value reflects the necessity of each persona relationship node in the persona relationship graph.
Further, in the above embodiment, as shown in fig. 2, the graph algorithm module 2 includes:
the community algorithm unit 203 is preset with a third graph algorithm model, and the third graph algorithm model is used for performing algorithm processing according to personnel information and/or initial clues and/or alarm information in clue information to obtain a relationship graph between the person and the community;
the relationship graph between the person and the community comprises a plurality of person relationship nodes obtained by person information and/or initial clues and/or alarm information in clue information, and each person relationship node is provided with corresponding community information.
Further, in the above embodiment, as shown in fig. 2, the thread mining module 3 includes:
the community acquisition model 304 is connected to the community algorithm unit 203, and is configured to input a relationship diagram between a person and a community into the central person acquisition model 303, calculate the same number of community information according to community information corresponding to each person relationship node, so as to acquire a target clue, where the target clue is the community information exceeding the preset number.
As a preferred embodiment, community acquisition model 304 employs a louvain community discovery algorithm.
Further, in the above embodiment, as shown in fig. 2, the graph algorithm module 2 includes:
the social circle algorithm unit 204 is preset with a fourth graph algorithm model, and the fourth graph algorithm model is used for performing algorithm processing according to personnel information and/or initial clues and/or alarm information in clue information to obtain a social graph of each personnel information;
the social graph comprises person relation nodes which are obtained by personnel information in the clue information and/or initial clues and/or alarm information and are associated with corresponding personnel information.
Further, in the above embodiment, as shown in fig. 2, the thread mining module 3 includes:
the core personage acquisition model 305 is connected to the social circle algorithm unit 204, and is configured to input a social graph of each personage information into the core personage acquisition model 305, calculate a core coefficient of each personage relationship node according to an association relationship between each personage relationship node in the social graph of each personage information, and acquire a target cue according to all the core coefficients.
As a preferred embodiment, the core character acquisition model 305 may employ a PageRank algorithm,
the core persona acquisition model 305 may calculate the number of connections passing through each persona relationship node according to the association relationship between each persona relationship node in the social graph of each persona information, and calculate the core coefficient (PageRank value) of the corresponding persona relationship node according to the number of connections. When the core coefficient of a person relation node is higher, the association relation between the description and the person relation node is more, namely the connection line of the person relation node is more. It should be noted that the person corresponding to the person relationship node with higher association coefficient is the core person for establishing the connection of the social circle, which is a key clue for case investigation.
For example, as shown in fig. 6, the core character 4a in the social graph 41 has the largest number of connections, the core character 4b in the social graph 42 has the largest number of connections, and the core character 4c in the social graph 43 has the largest number of connections.
The foregoing is merely illustrative of the preferred embodiments of the present invention and is not intended to limit the embodiments and scope of the present invention, and it should be appreciated by those skilled in the art that equivalent substitutions and obvious variations may be made using the description and illustrations of the present invention, and are intended to be included in the scope of the present invention.
Claims (8)
1. A thread collection system, comprising:
the system comprises an initial clue storage module, a first clue storage module and a second clue storage module, wherein the initial clue storage module is used for storing clue information provided by each person, and the clue information comprises personnel information of each person, and initial clue and alarm information provided by each person;
the graph algorithm module is connected with the initial clue storage module and comprises a plurality of graph algorithm units, each graph algorithm unit is respectively provided with a graph algorithm model in advance, and each graph algorithm model is respectively used for processing according to clue information to obtain a corresponding clue grid graph;
the graph algorithm module is used for outputting all clue network graphs;
the clue mining module is connected with the graph algorithm module, a plurality of clue mining models are preset in the clue mining module, each clue mining model corresponds to one type of clue network graph, and the clue mining module is used for inputting the clue network graph of the corresponding type into the corresponding clue mining model so as to obtain a target clue;
the graph algorithm module comprises:
the link map algorithm unit is connected with a first map algorithm model in advance, and the first map algorithm model is used for carrying out algorithm processing according to the clue information to obtain a link map;
wherein the link map comprises a plurality of link nodes obtained by the clue information, and every two link nodes with link relations are connected;
the thread mining module includes:
and the link connection discovery model is connected with the link connection graph algorithm unit and is used for inputting the link connection graph into the link connection discovery model, acquiring the target clue according to the connection relation between each link connection node, and the target clue is an intermediate link in the hidden link connection.
2. The thread collection system of claim 1, wherein the thread mining module comprises:
and the illicit person discovery model is connected with the link map algorithm unit and is used for inputting the link map into the illicit person discovery model, acquiring the target clue according to the connection relation between each link node, wherein the target clue is the link nodes with the connection relation of a plurality of link nodes being circulation relations.
3. The thread collection system of claim 1, wherein the graph algorithm module comprises:
the central graph algorithm unit is preset with a second graph algorithm model which is used for carrying out algorithm processing according to the clue information to obtain a shortest associated path graph;
the shortest association path diagram comprises a shortest association path between every two person relationship nodes which are not directly connected, wherein the shortest association path comprises a plurality of person relationship nodes obtained by the personnel information and/or the initial clue and/or the alarm information, and every two person relationship nodes with association relationship are connected.
4. The thread collection system of claim 3, wherein the thread mining module comprises:
and the center character acquisition model is connected with the center figure algorithm unit and is used for inputting the shortest association path diagram into the center character acquisition model, calculating corresponding association coefficients according to the times of character relation nodes passed by each shortest association path, and acquiring the target clue according to all the association coefficients.
5. The thread collection system of claim 4, wherein the graph algorithm module comprises:
the community algorithm unit is preset with a third graph algorithm model which is used for carrying out algorithm processing according to the clue information to obtain a relationship graph between the characters and the community;
the relationship diagram between the person and the community comprises a plurality of person relationship nodes obtained by the person information and/or the initial clue and/or the alarm information, and each person relationship node is provided with corresponding community information.
6. The thread collection system of claim 5, wherein the thread mining module comprises:
the community acquisition model is connected with the community algorithm unit and is used for inputting the relationship diagram between the characters and communities into the central character acquisition model, calculating the same quantity of community information according to the community information corresponding to each character relationship node to acquire the target clue, wherein the target clue is the community information exceeding the preset quantity.
7. The thread collection system of claim 1, wherein the graph algorithm module comprises:
the social circle algorithm unit is pre-provided with a fourth graph algorithm model, and the fourth graph algorithm model is used for carrying out algorithm processing according to the clue information to obtain a social graph of each piece of personnel information;
the social graph comprises person relation nodes which are obtained by the personnel information and/or the initial clues and/or the alarm information and are associated with the corresponding personnel information.
8. The thread collection system of claim 7, wherein the thread mining module comprises:
the core character acquisition model is connected with the social circle algorithm unit and is used for inputting the social graph of each piece of personnel information into the core character acquisition model, calculating the core coefficient of each person relation node according to the association relation between each person relation node in the social graph of each piece of personnel information, and acquiring the target clue according to all the core coefficients.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911342783.0A CN111008316B (en) | 2019-12-23 | 2019-12-23 | Clue collection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911342783.0A CN111008316B (en) | 2019-12-23 | 2019-12-23 | Clue collection system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111008316A CN111008316A (en) | 2020-04-14 |
| CN111008316B true CN111008316B (en) | 2023-09-01 |
Family
ID=70116058
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911342783.0A Active CN111008316B (en) | 2019-12-23 | 2019-12-23 | Clue collection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111008316B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114676288A (en) * | 2022-03-17 | 2022-06-28 | 北京悠易网际科技发展有限公司 | ID pull-through method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103489036A (en) * | 2013-09-27 | 2014-01-01 | 上海交通大学 | Emergent pre-warning grid dividing system and method based on hybrid space correlation analysis |
| CN107016457A (en) * | 2017-03-14 | 2017-08-04 | 上海市精神卫生中心 | One kind realizes community's hazardous act pre-warning system and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110078188A1 (en) * | 2009-09-28 | 2011-03-31 | Microsoft Corporation | Mining and Conveying Social Relationships |
-
2019
- 2019-12-23 CN CN201911342783.0A patent/CN111008316B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103489036A (en) * | 2013-09-27 | 2014-01-01 | 上海交通大学 | Emergent pre-warning grid dividing system and method based on hybrid space correlation analysis |
| CN107016457A (en) * | 2017-03-14 | 2017-08-04 | 上海市精神卫生中心 | One kind realizes community's hazardous act pre-warning system and method |
Non-Patent Citations (1)
| Title |
|---|
| 基于社交网络大数据线索分析平台研究及应用;钱汉伟等;《中国人民公安大学学报(自然科学版)》;20180515(第02期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111008316A (en) | 2020-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Musaev et al. | LITMUS: a multi-service composition system for landslide detection | |
| WO2016177066A1 (en) | Employee potential relationship analysis method and device | |
| CN105095211B (en) | The acquisition methods and device of multi-medium data | |
| Nagle et al. | Power-sharing after civil war: Thirty years since Lebanon’s Taif agreement | |
| CN104408149A (en) | Criminal suspect mining association method and system based on social network analysis | |
| CN111427968A (en) | Key person holographic archive construction method and device based on knowledge graph | |
| Lwowski et al. | Geospatial event detection by grouping emotion contagion in social media | |
| CN112231750B (en) | Multi-mode privacy protection method | |
| CN104636408A (en) | News authentication early warning method and system based on user generated content | |
| Sheth et al. | Semantic filtering for social data | |
| CN103077207A (en) | Method and system for analyzing microblog happiness index | |
| CN106934592A (en) | A kind for the treatment of method and apparatus of report information | |
| CN111008316B (en) | Clue collection system | |
| CN109800251A (en) | A kind of relationship discovery method and apparatus, computer readable storage medium | |
| CN109325661A (en) | A kind of community management method, apparatus and electronic equipment | |
| Brush et al. | Gender embeddedness of women entrepreneurs: An empirical test of the 5 “M” Framework | |
| Feng et al. | Chasing total solar eclipses on twitter: Big social data analytics for once-in-a-lifetime events | |
| CN116028467A (en) | Intelligent service big data modeling method, system, storage medium and computer equipment | |
| Epafras et al. | Religious blasphemy and monitory society in Indonesian digital age | |
| CN106021441B (en) | A kind of mechanism friend recommendation method and system | |
| CN100367295C (en) | Intelligent imaging implicit writting analytical system based on three-layer frame | |
| CN114186029A (en) | Analysis method of information leakage behavior, related device and computer storage medium | |
| Burnay | Privacy and surveillance in a digital era: transnational implications of China’s surveillance state | |
| CN112416922A (en) | Group partner association data mining method, device, equipment and storage medium | |
| Abdalla et al. | Towards emotional and opinion-based layers in city GIS |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |