CN107145895A - Public security crime class case analysis method based on k means algorithms - Google Patents
Public security crime class case analysis method based on k means algorithms Download PDFInfo
- Publication number
- CN107145895A CN107145895A CN201710147542.5A CN201710147542A CN107145895A CN 107145895 A CN107145895 A CN 107145895A CN 201710147542 A CN201710147542 A CN 201710147542A CN 107145895 A CN107145895 A CN 107145895A
- Authority
- CN
- China
- Prior art keywords
- case
- mtd
- vector
- class
- msub
- 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.)
- Pending
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 15
- 239000013598 vector Substances 0.000 claims abstract description 75
- 239000011159 matrix material Substances 0.000 claims abstract description 30
- 239000000203 mixture Substances 0.000 claims abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 7
- 210000003746 feather Anatomy 0.000 abstract description 4
- 244000144992 flock Species 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 241000271566 Aves Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 241000894007 species Species 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Tourism & Hospitality (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Probability & Statistics with Applications (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)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Public security crime class case analysis method of the invention based on k means algorithms, steps of the method are:Collect in the nearest case information deposit database of having been solved a case at least above 1 month of client, define 6 dimension vector attributes;The case characteristic use bag of words for extracting case carry out attribute vector, obtain case matrix;Birdsed of the same feather flock together with k means algorithms, form class case storehouse, take the average value of the coordinate of all cases vector in each class set, be considered as barycenter Ai, the i=K optimal value of the class set, and by the barycenter composition of vector matrix A of K classifications;New case is inputted by user, correspondence case characteristic vector is determined by step 1 definition vector attribute, that is, inputs the distance of case vector and k class case collection, and pushes closest class case collection to policeman in charge of the case user, the general character of case is found, assists to solve a case.Achievable intelligent, automation of the invention, the carry out class case cluster of accuracy are studied and judged, so as to substantially reduce people's police's workload, improve efficiency of solving a case.
Description
Technical field
The present invention relates to a kind of public security crime class case analysis method based on k-means algorithms.
Background technology
Occur case for current social, can not position personnel's specifying information situation of committing a crime in investigation, pass through other class case
Cluster as a kind of similar criminal type, the class case clustered by these, with reference to other information (for example:Tourism, trip, communications and liaison
Base station information) find and occur personnel jointly, it is possible to land for emphasis suspect, be significant to solving a case.Pass through now
A kind of intelligent k-means algorithms cluster its Similarity Class case together, go to judge without human subjective.
Find to cluster for class case by inquiry, typically by manually poly- to its similar cases class case manually by experience
Class, introduces advanced intelligent, actual effect, automation, efficiently clusters analysis method.
Occurs crime case for current social, general public security police but has to extraction case relevant information is detected in the act
When some case investigations less than specific any perpetrator's relevant information, at this time people's police are manually by a period of time scope and periphery
Similar case (case type, crime time, crime object, crime place, crime personnel area and tool used in crime) occurs for city
Chosen and clustered as class case by human subjective, then the imaginary cluster case is that same person or the same group of people commit a crime, in knot
Other relevant information association collision refinements are closed, the specific people that commits a crime finally is navigated to.But by artificial on the positioning of class case is searched
Following Railway Project can be run into by studying and judging:1) case species is cumbersome;2) case data volume is big;3) case data are easily omitted;4) people
Work subjective judgement, increase cluster risk;5) real-time is than relatively low, it is impossible to quick class case cluster;6) artificial high-volume is put into, significantly
Increase the workload of a line policeman in charge of the case.
The content of the invention
This problem is directed in order to solve the above problems, the purpose of the present invention is to propose to a kind of intelligent, automation, accurately
Property carry out class case cluster study and judge, so as to substantially reduce people's police's workload, improve efficiency of solving a case based on improving k-means algorithms
Public security crime class case analysis method.
The technical scheme is that:Based on the public security crime class case analysis method for improving k-means algorithms, its feature exists
In this method is:Collect in the history case information deposit database that user has solved a case recently, define dimension vector attribute,
The case feature in database is extracted, the case characteristic use bag of words extracted are subjected to attribute vector, data are obtained
The case matrix of case feature, birdss of the same feather flock together with k-means algorithms to case matrix in storehouse, forms class case storehouse, takes each class
The average value of the coordinate for all cases vector concentrated, and by the barycenter composition of vector matrix A of K classifications;It is new by user's input
Case, the dimension vector attribute defined by step 1 determines case characteristic vector corresponding to new input case, and by new case
Part is denoted as vectorial B1;B1 is calculated into Euclidean distance with Ai in matrix A respectively, that is, inputs case vector and K class case collection
Distance, and push closest class case collection to user, find the general character of case, assist to solve a case.
Further, this method specifically includes following steps:
Step 1:Collect in the history case information deposit database that user has solved a case, define 6 dimension vector attributes;
Step 2:Extract database in case feature, by the case characteristic use bag of words extracted carry out attribute to
Quantify, obtain the case matrix of case feature in database;
Step 3:Case matrix is clustered with k-means algorithms, class case storehouse is formed, taken all in each class set
The average value of the coordinate of case vector, is considered as barycenter Ai, the i=K optimal value of the class set, and by the barycenter composition of vector of K classifications
Matrix A;
Step 4:New case is inputted by user, the 6 dimension vector attributes defined by step 1 determine new input case
Case characteristic vector corresponding to part, and new case is denoted as vectorial B1;B1 is calculated into Euclid with Ai in matrix A respectively
Distance, that is, input the distance of case vector and K class case collection, and pushes closest class case collection to user, finds case
General character, assists to solve a case.
Further, the history case solved a case is at least one month.
Further, the dimension vector attribute in the step 1 is respectively case type, crime time, crime object, crime
Place, crime personnel area and tool used in crime.
Further, in the step 2:The bag of words carry out concretely comprising the following steps for attribute vector:Calculate first every
One merit briefly explains CaseiIn keyword nameij(nameij represents attribute thresholding) relative to S in bag of words basic words
Similarity, S > 0, and select maximum similarity;Then the maximum similarity of selection is multiplied by keyword nameijIn bag of words
Order make its uniqueness, obtain nameijCorresponding numerical value xij;Each feature of all history cases carries out aforesaid operations, obtains
To case matrix, it is:
In formula:x11To compare basic similarity maximum, x in first case vector dimension-case type and bag of wordsi1
It is i-th case vector dimension-case type with comparing basic similarity maximum, X in bag of words1jFor the 1st case vector dimension
Degree-case type in bag of words with comparing basic similarity maximum, xi1For i-th of case vector dimension-case type and bag of words
Middle to compare basic similarity maximum, j is dimension values, and j=6, i is case number of packages, i>0.
Further, in the step 3:K values are defined as:
First, the span of the K values determined according to case type, K span is 1-100;
Secondly, the respective value that the round-robin algorithm of the scope combination silhouette coefficient of the K values of determination is obtained into each K values is respectively
Silhouette coefficient value, if one of silhouette coefficient value is closest to 1, the silhouette coefficient value is silhouette coefficient optimal value, the wheel
Wide coefficient optimal value correspondence K values are K optimal values, regard the K optimal values as input parameter.
The silhouette coefficient calculation is as follows:
For each sample i of each class set:
1) a (i)=average (distances of i other samples into all class sets that it belongs to) is calculated
2) b (i)=min (average distances of the i to the point of all non-place class set itself) is calculated
3) sample i silhouette coefficient is
4) silhouette coefficients of all samples is averaging, is exactly the total silhouette coefficient of the cluster result, between -1 to 1 it
Between, closer to 1,
The k-means algorithms are to the specific method birdsed of the same feather flock together:
Make data acquisition system X in a classes(a)For:X(a)={ x1 (a),x2 (a),…,xi (a),…,xm(a) (a),
Wherein, 1≤i≤m (a), m (a) are the data sample number in a class clusters;
Make i-th of sample in a classesFor:xi (a)={ xi,1 (a),xi,2 (a),…,xi,j (a),…,xi,n (a), wherein, 1≤j
≤ n, n are the characteristic attribute number of sample data.
The center c of a classes(a)For:C(a)={ c1 (a),c2 (a),…,cj (a),…,cn (a)}
Then the concentration class δ (a) of a classes data may be defined as:
Wherein, bf (xi,j (a)) it is constraint factorConstraint factor bf (xi,j (a)) work
With being to make δ (a) ∈ (0,1), data concentration class is by calculating the average value of each vector and the manhatton distance of central point in class set
To weigh, the influence of outlier is effectively reduced, during iteration, does not stop to detect classification state,
IfWhen the value tends to convergence, representing the effect of iteration tends to be optimal, stops iteration,
Algorithm terminates, and returns to final classification result.
This law invention beneficial effect be:Due to using above-mentioned technical proposal, the present invention can be achieved intelligent, automation,
The carry out class case cluster of accuracy is studied and judged, so as to substantially reduce people's police's workload, improves efficiency of solving a case.
Brief description of the drawings:
Fig. 1 is flow chart of the present invention based on the public security crime class case analysis method for improving k-means algorithms.
Embodiment
Technical scheme is described further with specific embodiment below in conjunction with the accompanying drawings.
As shown in figure 1, the present invention is based on the public security crime class case analysis method for improving k-means algorithms, this method is specific
Comprise the following steps:
Step 1:Collect in the history case information deposit database that user has solved a case at least above 1 month recently, it is fixed
Adopted 6 dimension vector attributes;
Step 2:Extract database in case feature, by the case characteristic use bag of words extracted carry out attribute to
Quantify, obtain the case matrix of case feature in database;
Step 3:Case matrix is birdsed of the same feather flock together with k-means algorithms, class case storehouse is formed, taken all in each class set
The average value of the coordinate of case vector, is considered as barycenter Ai, the i=K of the class set, and by the barycenter composition of vector matrix A of K classifications;
Step 4:New case is inputted by user, the 6 dimension vector attributes defined by step 1 determine new input case
Case characteristic vector corresponding to part, and new case is denoted as vectorial B1;B1 is calculated into Euclid with Ai in matrix A respectively
Distance, that is, input the distance of case vector and K class case collection, and pushes closest class case collection to user, finds case
General character, assists to solve a case.
Further, 6 dimension vector attributes in the step 1 include case type, crime time, crime object, work
Case place, crime personnel area and tool used in crime.
Further, in the step 2:The bag of words carry out concretely comprising the following steps for attribute vector:Calculate first every
One merit briefly explains CaseiIn nameijRelative to the similarity of S in bag of words basic words, and select maximum similarity;
Then the maximum similarity of selection is multiplied by keyword nameijOrder in bag of words makes its uniqueness, obtains nameijCorrespondence
Numerical value xij;Each feature of all history cases carries out aforesaid operations, obtains case matrix, is:
In formula:x11To compare basic similarity maximum, x in first case vector dimension-case type and bag of wordsi1
It is i-th case vector dimension-case type with comparing basic similarity maximum, X in bag of words1jFor the 1st case vector dimension
Degree-case type in bag of words with comparing basic similarity maximum, xi1For m-th of case vector dimension-case type and bag of words
Middle to compare basic similarity maximum, j is dimension values, and j=6, i is case number of packages.
First, the span of the K values determined according to case type, K span is 1-100;
Secondly, the respective value that the round-robin algorithm of the scope combination silhouette coefficient of the K values of determination is obtained into each K values is respectively
Silhouette coefficient value, if one of silhouette coefficient value is closest to 1, the silhouette coefficient value is silhouette coefficient optimal value, the wheel
Wide coefficient optimal value correspondence K values are K optimal values, regard the K optimal values as input parameter.
Embodiment 1:
Above-mentioned history magnanimity case attribute vector:
6 dimensional characteristics attributes are processed as, keyword similarity algorithm (algorithm steps are passed through with reference to bag of words:First
Calculate each merit brief description CaseiIn nameijRelative to the similarity of S in bag of words basic words, and select maximum
Similarity;Then the maximum similarity of selection is multiplied by keyword nameijOrder (making its uniqueness) in bag of words) so as to count
Each dimension vector value is calculated, so that case matrix is formed, it is as follows:
3) K value optimum choices in K-Means algorithms, obtain optimal K values:
By determining K scopes [1,100], i.e. 1<=k<=100, with reference to silhouette coefficient from 1 to 100, cycle calculations are optimal
Silhouette coefficient value, so as to obtain optimal profile coefficient correspondence K values, i.e. K optimal values.For example:Calculating obtains K optimal value=12,
It is 12 classes i.e. to history case cluster.12 class sets are judged to final result, take in each class set all cases vector
The average value of coordinate, is considered as the barycenter of the class set, and by the barycenter composition of vector matrix As of 12 classifications A1, A2, A3, A4, A5,
A6, A7, A8, A9 ... A12 }, barycenter Ai is to represent i-th of class case collection;
4) the class case of new input case is intelligently incorporated into
The nearest newly-increased case of user's input:
The case characteristic vector value is calculated by above-mentioned 6 dimensional characteristics vector attribute bag of words algorithm, and indicated
It is for vector:{18.0,57.3,109.2,96.2,18.6,49.5};Vector is calculated into Europe with Ai in above-mentioned case matrix A respectively
Distance is obtained in several, that is, calculates new case vector and the distance of 12 Ge Leianji centers center of mass point, minimum distance is calculated as the matter
Heart point class case, so as to push closest class case collection to policeman in charge of the case user, and then positions the general character of case, final to assist broken
Case.
Embodiment 2:
Client's public security nearest time (at least above 1 month) is collected first to have solved a case case information, carries out data processing,
Information is as follows after processing data:(6 dimensions of case)
To above-mentioned history magnanimity case attribute to:
6 dimensional characteristics vectors are processed as, keyword similarity algorithm (algorithm steps are passed through with reference to bag of words:First
Calculate each merit brief description CaseiIn nameijRelative to the similarity of S in bag of words basic words, and select maximum
Similarity;Then the maximum similarity of selection is multiplied by keyword nameijOrder (making its uniqueness) in bag of words is so as to count
Each dimension vector value is calculated, so that case matrix is formed, it is as follows:
3) K value optimum choices in K-Means algorithms, obtain optimal K values:
K scopes [1,100], i.e., 1 are obtained by case type<=k<=100, with reference to silhouette coefficient from 1 to 100, circulate
Optimal profile coefficient value is calculated, so as to obtain optimal profile coefficient correspondence K values, i.e. K optimal values.For example:It is optimal that calculating obtains K
Value=32, i.e., be 32 classes to history case cluster.32 class sets are judged to final result, all cases in each class set are taken
The average value of the coordinate of vector, is considered as the barycenter of the class set, and by the barycenter composition of vector matrix As of 32 classifications A1, A2, A3,
A4, A5, A6, A7, A8, A9 ... A32 }, barycenter Ai is to represent i-th of class case collection;The class case of new input case is intelligently incorporated into
The nearest newly-increased case of user's input:
The case characteristic vector value is calculated by above-mentioned 6 kinds of vector attributes bag of words algorithm, and is denoted as vector and is:
{21.0,37.8,71.2,109.2,28.7,99.1};By vector respectively with above-mentioned case matrix A Ai calculate Euclid away from
From calculating newly-increased case vector and the distance of 32 Ge Leianji centers center of mass point, minimum distance be calculated as into the center of mass point class
Case, so as to push closest class case collection to policeman in charge of the case user, and then positions the general character of case, final to assist to solve a case.
Claims (6)
1. based on the public security crime class case analysis method for improving k-means algorithms, it is characterised in that this method is:Collect user
In the history case information deposit database solved a case recently, dimension vector attribute is defined, the case extracted in database is special
Levy, the case characteristic use bag of words extracted are subjected to attribute vector, the case square of case feature in database is obtained
Battle array, is clustered with k-means algorithms to case matrix, forms class case storehouse, takes all cases in each class set vectorial
The average value of coordinate, and by the barycenter composition of vector matrix A of K classifications;New case is inputted by user, according to the dimension of definition
Vector attribute determines case characteristic vector corresponding to new input case, and new case is denoted as into vectorial B1;By B1 respectively with
Ai calculates Euclidean distance in matrix A, that is, inputs the distance of case vector and K class case collection, and pushes closest class
Case collection finds the general character of case, assists to solve a case to user.
2. public security crime class case analysis method according to claim 1, it is characterised in that this method specifically includes following step
Suddenly:
Step 1:Collect in the history case information deposit database that user has solved a case, define 6 dimension vector attributes;
Step 2:The case feature in database is extracted, the case characteristic use bag of words extracted are subjected to attribute vector
Change, obtain the case matrix of case feature in database;
Step 3:Case matrix is clustered with k-means algorithms, class case storehouse is formed, takes all cases in each class set
The average value of the coordinate of vector, is considered as barycenter Ai, the i=K optimal value of the class set, and by the barycenter composition of vector matrix of K classifications
A;
Step 4:New case is inputted by user, the 6 dimension vector attributes defined by step 1 determine new input case institute
Correspondence case characteristic vector, and new case is denoted as vectorial B1;By B1 respectively with matrix A Ai calculate Euclid away from
From, that is, the distance of case vector and K class case collection is inputted, and closest class case collection is pushed to policeman in charge of the case user, find
The general character of case, assists to solve a case.
3. public security crime class case analysis method according to claim 2, it is characterised in that the history case solved a case
Part is at least one month.
4. public security crime class case analysis method according to claim 2, it is characterised in that dimension in the step 1 to
Amount attribute is respectively case type, crime time, crime object, crime place, crime personnel area and tool used in crime.
5. public security crime class case analysis method according to claim 2, it is characterised in that in the step 2:The bag of words
Model carries out concretely comprising the following steps for attribute vector:Each merit brief description Case is calculated firstiIn keyword nameij
Relative to the similarity of S in bag of words basic words, S > 0, and select maximum similarity;Then the maximum similarity of selection is multiplied
With keyword nameijOrder in bag of words makes its uniqueness, obtains nameijCorresponding numerical value xij;All history cases it is every
Individual feature carries out aforesaid operations, obtains case matrix, is:
<mrow>
<mfenced open = "[" close = "]">
<mtable>
<mtr>
<mtd>
<msub>
<mi>x</mi>
<mn>11</mn>
</msub>
</mtd>
<mtd>
<msub>
<mi>x</mi>
<mn>12</mn>
</msub>
</mtd>
<mtd>
<mi>&Lambda;</mi>
</mtd>
<mtd>
<msub>
<mi>x</mi>
<mrow>
<mn>1</mn>
<mi>j</mi>
</mrow>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>x</mi>
<mn>21</mn>
</msub>
</mtd>
<mtd>
<msub>
<mi>x</mi>
<mn>22</mn>
</msub>
</mtd>
<mtd>
<mi>&Lambda;</mi>
</mtd>
<mtd>
<msub>
<mi>x</mi>
<mrow>
<mn>2</mn>
<mi>j</mi>
</mrow>
</msub>
</mtd>
</mtr>
<mtr>
<mtd>
<mi>M</mi>
</mtd>
<mtd>
<mi>M</mi>
</mtd>
<mtd>
<mi>&Lambda;</mi>
</mtd>
<mtd>
<mi>M</mi>
</mtd>
</mtr>
<mtr>
<mtd>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mn>1</mn>
</mrow>
</msub>
</mtd>
<mtd>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mn>2</mn>
</mrow>
</msub>
</mtd>
<mtd>
<mi>&Lambda;</mi>
</mtd>
<mtd>
<msub>
<mi>x</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>,</mo>
</mrow>
In formula:x11To compare basic similarity maximum, x in first case vector dimension-case type and bag of wordsi1For i-th
Individual case vector dimension-case type in bag of words with comparing basic similarity maximum, X1jFor the 1st case vector dimension-case
Part type in bag of words with comparing basic similarity maximum, xijIt is i-th of case vector dimension-case type with being compared in bag of words
Basic similarity maximum, j is dimension values, and j=6, j is case number of packages, j>0.
6. public security crime class case analysis method according to claim 2, it is characterised in that in the step 3:K values are really
It is set to:
First, the span of the K values determined according to case type, K span is 1-100;
Secondly, the round-robin algorithm of the scope combination silhouette coefficient of the K values of determination is obtained to the respective value respectively profile of each K values
Coefficient value, if one of silhouette coefficient value is closest to 1, the silhouette coefficient value is silhouette coefficient optimal value, the profile system
Number optimal value correspondence K values are K optimal values, regard the K optimal values as input parameter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710147542.5A CN107145895A (en) | 2017-03-13 | 2017-03-13 | Public security crime class case analysis method based on k means algorithms |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710147542.5A CN107145895A (en) | 2017-03-13 | 2017-03-13 | Public security crime class case analysis method based on k means algorithms |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107145895A true CN107145895A (en) | 2017-09-08 |
Family
ID=59783807
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710147542.5A Pending CN107145895A (en) | 2017-03-13 | 2017-03-13 | Public security crime class case analysis method based on k means algorithms |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107145895A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108874767A (en) * | 2018-05-04 | 2018-11-23 | 上海瀚所信息技术有限公司 | A kind of four dimension module intelligence Compare Systems and method for public security system |
CN109241320A (en) * | 2018-09-30 | 2019-01-18 | 电子科技大学 | The division methods of teenage crime area cluster based on Time Series Clustering |
CN110609961A (en) * | 2018-05-29 | 2019-12-24 | 南京大学 | Collaborative filtering recommendation method based on word embedding |
CN111198953A (en) * | 2018-11-16 | 2020-05-26 | 北京智慧正安科技有限公司 | Case text information based method and system for recommending cases and computer readable storage medium |
CN111488384A (en) * | 2019-01-29 | 2020-08-04 | 中国石油化工股份有限公司 | Intelligent drilling scheme recommendation method and system |
CN111753872A (en) * | 2020-05-12 | 2020-10-09 | 高新兴科技集团股份有限公司 | Method, device, equipment and storage medium for analyzing association of serial and parallel cases |
CN112905863A (en) * | 2021-03-19 | 2021-06-04 | 青岛檬豆网络科技有限公司 | Automatic customer classification method based on K-Means clustering |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346355A (en) * | 2013-07-26 | 2015-02-11 | 南京中兴力维软件有限公司 | Method and system for intelligent retrieval of series public security cases |
CN106127231A (en) * | 2016-06-16 | 2016-11-16 | 中国人民解放军国防科学技术大学 | A kind of crime individual discrimination method based on the information Internet |
CN106127241A (en) * | 2016-06-17 | 2016-11-16 | 中国电子科技集团公司第二十八研究所 | One is combined related cases sorting technique and categorizing system of combining related cases |
-
2017
- 2017-03-13 CN CN201710147542.5A patent/CN107145895A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104346355A (en) * | 2013-07-26 | 2015-02-11 | 南京中兴力维软件有限公司 | Method and system for intelligent retrieval of series public security cases |
CN106127231A (en) * | 2016-06-16 | 2016-11-16 | 中国人民解放军国防科学技术大学 | A kind of crime individual discrimination method based on the information Internet |
CN106127241A (en) * | 2016-06-17 | 2016-11-16 | 中国电子科技集团公司第二十八研究所 | One is combined related cases sorting technique and categorizing system of combining related cases |
Non-Patent Citations (2)
Title |
---|
卢睿: "刑事案件的属性约简聚类算法研究", 《中国人民公安大学学报( 自然科学版)》 * |
李亚 等: "基于改进K-Means 聚类和BP 神经网络的台区线损率计算方法", 《中国电机工程学报》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108874767A (en) * | 2018-05-04 | 2018-11-23 | 上海瀚所信息技术有限公司 | A kind of four dimension module intelligence Compare Systems and method for public security system |
CN108874767B (en) * | 2018-05-04 | 2022-04-05 | 上海瀚所信息技术有限公司 | Four-dimensional model intelligent comparison system and method for public security system |
CN110609961A (en) * | 2018-05-29 | 2019-12-24 | 南京大学 | Collaborative filtering recommendation method based on word embedding |
CN109241320A (en) * | 2018-09-30 | 2019-01-18 | 电子科技大学 | The division methods of teenage crime area cluster based on Time Series Clustering |
CN111198953A (en) * | 2018-11-16 | 2020-05-26 | 北京智慧正安科技有限公司 | Case text information based method and system for recommending cases and computer readable storage medium |
CN111488384A (en) * | 2019-01-29 | 2020-08-04 | 中国石油化工股份有限公司 | Intelligent drilling scheme recommendation method and system |
CN111753872A (en) * | 2020-05-12 | 2020-10-09 | 高新兴科技集团股份有限公司 | Method, device, equipment and storage medium for analyzing association of serial and parallel cases |
CN112905863A (en) * | 2021-03-19 | 2021-06-04 | 青岛檬豆网络科技有限公司 | Automatic customer classification method based on K-Means clustering |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107145895A (en) | Public security crime class case analysis method based on k means algorithms | |
CN101187927B (en) | Criminal case joint investigation intelligent analysis method | |
JP5665879B2 (en) | Method and system for sample image index creation and image filtering and retrieval | |
CN104462184B (en) | A kind of large-scale data abnormality recognition method based on two-way sampling combination | |
CN107122352A (en) | A kind of method of the extracting keywords based on K MEANS, WORD2VEC | |
CN106485262A (en) | A kind of bus load Forecasting Methodology | |
CN110348486A (en) | Based on sampling and feature brief non-equilibrium data collection conversion method and system | |
CN108805142A (en) | A kind of crime high-risk personnel analysis method and system | |
CN103336771A (en) | Data similarity detection method based on sliding window | |
CN107133257A (en) | A kind of similar entities recognition methods and system based on center connected subgraph | |
CN108805213A (en) | The electric load curve bilayer Spectral Clustering of meter and Wavelet Entropy dimensionality reduction | |
CN109784408A (en) | A kind of embedded time series Decision-Tree Method and system of marginal end | |
CN109686402A (en) | Based on key protein matter recognition methods in dynamic weighting interactive network | |
CN106202388B (en) | A kind of user gradation Automated Partition Method and system | |
CN110580510A (en) | clustering result evaluation method and system | |
CN102663723A (en) | Image segmentation method based on color sample and electric field model | |
CN116860840B (en) | Rapid retrieval method for highway pavement information | |
CN108304476A (en) | A kind of user's representation data integration method and system based on uncertain data table | |
CN109993042A (en) | A kind of face identification method and its device | |
CN116644184A (en) | Human Resource Information Management System Based on Data Clustering | |
CN110826449A (en) | Non-motor vehicle re-identification target retrieval method based on light convolutional neural network | |
CN104599062A (en) | Classification based value evaluation method and system for agricultural scientific and technological achievements | |
CN110176309A (en) | It is a kind of for predicting the medical data processing method of cardiovascular disease | |
CN109284409A (en) | Picture group geographic positioning based on extensive streetscape data | |
CN117493998A (en) | Questionnaire investigation event intelligent classification management method and system based on big data |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170908 |