CN110750609A - Method for predicting number of criminal cases based on space-time data and neural network - Google Patents
Method for predicting number of criminal cases based on space-time data and neural network Download PDFInfo
- Publication number
- CN110750609A CN110750609A CN201910973743.XA CN201910973743A CN110750609A CN 110750609 A CN110750609 A CN 110750609A CN 201910973743 A CN201910973743 A CN 201910973743A CN 110750609 A CN110750609 A CN 110750609A
- Authority
- CN
- China
- Prior art keywords
- region
- neural network
- area
- crime
- hidden layer
- 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
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 14
- 238000003062 neural network model Methods 0.000 claims abstract description 20
- 238000012360 testing method Methods 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims abstract description 14
- 239000011159 matrix material Substances 0.000 claims description 18
- 150000001875 compounds Chemical class 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 6
- 238000011160 research Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 3
- 238000012886 linear function Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 8
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 241001123248 Arma Species 0.000 description 1
- 241000728173 Sarima Species 0.000 description 1
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/082—Learning methods modifying the architecture, e.g. adding, deleting or silencing nodes or connections
-
- 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)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Remote Sensing (AREA)
- Economics (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
A crime case quantity prediction method based on space-time data and a neural network comprises the following steps of firstly, recalculating the proximity relation between areas by using taxi flow data on the basis of the traditional distance calculation; then, considering that crimes have the property of adjacent propagation in time and space, taking the number of crime cases of the area and the adjacent area in the current and past periods as input, and taking the number of crime cases of the area in the next period as output, and constructing a BP neural network model for each area; and determining the range of the superior adjacent region, the range of the past time period and the node number of the hidden layer of the neural network by training the BP neural network of each region and calculating the average absolute error of the test set. The method has reasonable model, improves the accuracy of predicting the number of crime cases by analyzing the crime record data and the taxi flow data from the space-time perspective, and provides guidance for optimizing and deploying the patrol police strength of each area.
Description
Technical Field
The invention relates to the field of crime prediction, in particular to a crime case quantity prediction method based on space-time data and a neural network.
Background
The crime prediction is to apply a scientific method, judge the conditions, structures, development trends and the like of crime phenomena possibly occurring in a specific space-time range in the future according to the existing crime data and the analysis and research of various relevant factors possibly influencing crimes, and is an important scientific basis for making crime prevention strategies and tactical measures. The industry and academia have been studying crime prediction problems from different perspectives. The method and the system have the advantages that the number of crime cases is predicted, namely the number of crime cases in a certain region in a future specified time period is predicted according to crime record data and other available data, guidance can be provided for optimizing and deploying patrol police force, and the method and the system are more and more favored.
Currently, the research on the prediction of the number of criminal cases is mainly started from the following three aspects: (1) the method is characterized in that the criminal quantity of a certain area in the next period is predicted according to the criminal quantity of the area in the past period by utilizing the regularity of the criminal behaviors distributed in time, and a commonly used prediction method is time sequence analysis (such as an ARMA model and a SARIMA model), a statistical regression algorithm and the like. (2) The method utilizes the spatial regularity of the distribution of the criminal behaviors, and predicts the number of crimes in a certain area according to the criminal conditions in the adjacent area of the certain area and by combining the characteristic data such as the population structure, the economic level, the education degree and the like in each area. In recent years, with the development of big data technology, the characteristic data in each area is also increased by human activity data, such as taxi traffic and people traffic in various places, and related research results can be found in document [1 ]: wang H, Kifer D, Graif C, equivalent.Crime Rate with Big Data [ C ]// Proceedings of the 22nd ACMInternal Conference on Knowledge Discovery and Data mining.ACM,2016: 635-: wang H, Li Z. registration retrieval learning via Mobility Flow [ C ]// Proceedings of the 26th ACM International conference on Information and Knowledge management. ACM,2017: 237-: kadar C, Iria J, Pletikosa I. expanding Foursquare-derivedfectures for crime prediction in New York City [ C ]// Urban Computing Workshop-KDD (i.e., Kadar C, Iria J, Pletikosa I. exploring the effects of feature data provided by Foursquare-the ubiquitous Computing Workshop of International knowledge discovery and data conference, N.Y. [ C ]// 22nd ACM, 2016). (3) And simultaneously analyzing from the time and space angles, namely predicting the crime number of a certain area in the next period according to the crime number of the area and the adjacent area in the past period. The results of the more recent studies can be found in the literature [4 ]: liumeilin, rare, Huanghong, etc., criminal intelligence prediction analysis [ J ] intelligence journal, 2018,37(9):27-37 ] based on space-time sequence mixing model.
The research results provide powerful theoretical support for the prediction of the number of crime cases. However, they either analyze from a single temporal or spatial perspective or use only a single source of comparative data, historical crime record data, without mining the impact of other spatiotemporal data on crimes. It is a significant problem how to use various data sources and make criminal case quantity prediction from a spatiotemporal perspective to improve prediction accuracy.
Disclosure of Invention
In order to overcome the defects that the conventional crime case quantity prediction method is single in data source, does not well excavate the property that crimes have adjacent propagation in time and space and the like, the invention provides the crime case quantity prediction method based on space-time data and neural networks, which is reasonable in model, uses crime record data and taxi traffic data and analyzes from the space-time perspective on the basis of reference and absorption of the conventional research results, so as to achieve the purpose of improving the crime case quantity prediction accuracy.
The technical scheme adopted by the invention for solving the technical problems is as follows:
a criminal case quantity prediction method based on spatio-temporal data and a neural network comprises the following steps:
(1) dividing a region (a city or a county) to be researched into geographic spaces according to local patrol or administrative district division, wherein the patrol or administrative district is called as a 'region', and the number of the region is represented by M;
(2) dividing crime record data in each region into time sequences with the same time interval, and constructing a test set and a training set, wherein the time interval is set by a user as required;
(3) constructing a BP neural network model for each region i, wherein the BP neural network model is composed of an input layer, a hidden layer and an output layer, an input vector is composed of the number of crime cases of the region i and m adjacent regions of the region i in a time period s and n past time periods (i.e. s-1, s-2, …, s-n), the output is the number of crime cases of the region in a time period s +1, and the number of nodes of the hidden layer is c;
(4) training and testing the BP neural network model of each region by using a training set and a testing set respectively;
(5) for each region, predicting the number of crime cases of the region in the next time period by using the BP neural network model of the region trained in the step (4).
Further, in the step (3), the m neighboring areas of the area i are defined by:
s3-1: constructing a normalized spatial weight matrix W based on distanced：
In the formula (I), the compound is shown in the specification,
wherein d isijRepresents the center distance between the region i and the region j, C is a constant, exp (·) represents an exponential function based on a natural number e;
s3-2: construction of normalized spatial weight matrix W based on taxi traffict：
In the formula (I), the compound is shown in the specification,
wherein f isijRepresenting the total taxi flow taking the area i as a starting point and the area j as a target point;
s3-3: constructing a normalized mixed space weight matrix W:
in the formula (I), the compound is shown in the specification,
thereinAnd
respectively, the distance-based normalized spatial weight matrix WdAnd the normalized space weight matrix W based on taxi traffictIs/are as followsAnd
a and b are weight parameters and satisfy a + b ═ 1;
s3-4: the m neighboring regions of the region i are the m regions corresponding to the first m large elements in the ith row of the normalized hybrid spatial weight matrix W.
Still further, in the step (3), the BP neural network model of each region i is represented as:
in the formula (I), the compound is shown in the specification,
represents the set of the region i and its m neighbors, zj(s) represents the number of crime cases for zone j over time period s,
representing the connection weight between the input node and the hidden layer node, αkRepresenting the connection weight between the hidden layer node and the output node, βkAnd representing a threshold value of a hidden layer node, wherein a hidden layer activation function F is a sigmoid function, and an output layer activation function F is a linear function.
Further, in the step (4), the values of m, n and c in the "m neighboring regions", "n time periods in the past", and "the number of nodes of the hidden layer" in the step (3) are continuously changed, the BP neural network model of each of the regions is trained and tested using the training set and the test set, and the average absolute error of the test set is calculated, thereby determining the superior values of m, n and c.
The technical conception of the invention is as follows: according to the method, on the basis of traditional distance calculation, the proximity relation between areas is recalculated by using taxi flow data; considering that crimes have the property of adjacent propagation in space and time, taking the number of crime cases of an area and the adjacent area in the current and past periods as input, taking the number of crime cases of the area in the next period as output, and constructing a BP neural network model for each area; and determining the range of the superior adjacent region, the range of the past time period and the node number of the hidden layer of the neural network by training the BP neural network of each region and calculating the average absolute error of the test set.
According to the technical scheme, the beneficial effects of the invention are mainly shown in that: the invention uses the crime record data and the taxi flow data and analyzes from the space-time perspective, improves the accuracy of predicting the number of crime cases and provides guidance for optimizing and deploying the patrol police force of each area.
Drawings
Fig. 1 is a structural diagram of a BP neural network model constructed for each region according to the present invention.
Detailed Description
In order that the objects, aspects and advantages of the present invention will become more apparent, the invention is further described with reference to the following detailed description and the accompanying drawings.
Referring to fig. 1, a method for predicting the number of criminal cases based on spatio-temporal data and a neural network first divides a region (prefecture or county) under study into geographical spaces. A simple mesh method can be used for segmentation. In this embodiment, the region under study (city or county) is divided into geographical spaces according to the local patrol or administrative district division, which is called "region". This geospatial segmentation approach conforms to the actual patrol or management model. The number of said regions is denoted by M. Then, dividing the crime record data in each region into time sequences of the same time interval, and constructing a test set and a training set. The time interval may be set by the user as desired, such as a month, a week, a day.
According to the window breaking theory in crime, crimes have a proximity propagation property in both time and space. The starting point of the present invention is also based on this property, which can be summarized as: the number of crime cases in the next period of time for the area is predicted by calculating the number of crime cases in the current period of time and in the past period of time in the area and the adjacent areas of the area. Therefore, how to select the neighborhood of the region is the basis and key of the present invention. Particularly with the development of traffic and the frequent human travel activities, the proximity relationship between the areas is not merely spatially adjacent. To solve the problem to a certain extent, the invention recalculates the proximity relation between areas by using taxi flow data based on the traditional distance calculation, and comprises the following steps:
s3-1: constructing a normalized spatial weight matrix W based on distanced：
In the formula (I), the compound is shown in the specification,wherein d isijRepresents the center distance between the region i and the region j, C is a constant, exp (·) represents an exponential function based on a natural number e;
s3-2: construction of normalized spatial weight matrix W based on taxi traffict：
In the formula (I), the compound is shown in the specification,
wherein f isijRepresenting the total taxi flow taking the area i as a starting point and the area j as a target point;
s3-3: constructing a normalized mixed space weight matrix W:
in the formula (I), the compound is shown in the specification,
thereinAnd
respectively, the distance-based normalized spatial weight matrix WdAnd the normalized space weight matrix W based on taxi traffictIs/are as follows
And
a and b are weight parameters and satisfy a + b ═ 1;
s3-4: the m neighboring regions of the region i are the m regions corresponding to the first m large elements in the ith row of the normalized hybrid spatial weight matrix W.
And after the definition of m adjacent areas of the area i is completed, constructing a BP neural network model for each area i. The BP neural network model is composed of an input layer, a hidden layer and an output layer, as shown in fig. 1. Wherein the input vector is composed of the number of crime cases of the region i and m adjacent regions thereof in a time period s and n time periods (i.e. s-1, s-2, …, s-n) in the past, and the number of input nodes is (1+ m) × (1+ n); the output is the number of crime cases in the time period s +1 of the area, and the number of output nodes is 1; the number of nodes of the hidden layer is c. The BP neural network model for each region i may be represented as:
in the formula (I), the compound is shown in the specification,
represents the set of the region i and its m neighbors, zj(s) represents the number of crime cases for zone j over time period s,
representing the connection weight between the input node and the hidden layer node, αkRepresenting the connection weight between the hidden layer node and the output node, βkAnd representing a threshold value of a hidden layer node, wherein a hidden layer activation function F is a sigmoid function, and an output layer activation function F is a linear function.
And then, continuously changing the values of m, n and c in the m adjacent regions, the n time periods in the past and the node number of the hidden layer as c, training and testing the BP neural network model of each region by using the training set and the test set, and calculating the average absolute error of the test set so as to determine the superior values of m, n and c. In this embodiment, the value sets of m, n and c are [0,10], [1,15], {2,3,4,5,10,50,100}, respectively, i.e. there are 1155 different value combinations of m, n and c, and the mean absolute error of the test set is the smallest when m, n and c are 6, 2 and 10, respectively.
Next, for each of the areas, the trained BP neural network model of the area may be used to predict the number of crime cases for the area in the next time period.
Claims (4)
1. A criminal case quantity prediction method based on spatio-temporal data and a neural network is characterized by comprising the following steps: the method comprises the following steps:
(1) dividing a geographical space of a research area according to local patrol or management jurisdiction division, wherein the patrol or management jurisdiction is called as a 'region', and the number of the regions is expressed by M;
(2) dividing crime record data in each region into time sequences with the same time interval, and constructing a test set and a training set, wherein the time interval is set by a user as required;
(3) and constructing a BP neural network model for each region i. The BP neural network model is composed of an input layer, a hidden layer and an output layer. Wherein, the input vector is composed of the number of crime cases of the area i and m adjacent areas thereof in a time period s and n past time periods (i.e. s-1, s-2, …, s-n), the output is the number of crime cases of the area in a time period s +1, and the number of nodes of the hidden layer is c;
(4) training and testing the BP neural network model of each region by using a training set and a testing set respectively;
(5) for each region, predicting the number of crime cases of the region in the next time period by using the BP neural network model of the region trained in the step (4).
2. The method of claim 1, wherein the method comprises the following steps: in the step (3), the m neighboring areas of the area i are defined by:
s3-1: constructing a normalized spatial weight matrix W based on distanced：
In the formula (I), the compound is shown in the specification,
wherein d isijRepresents the center distance between the region i and the region j, C is a constant, exp (·) represents an exponential function based on a natural number e;
s3-2: construction of normalized spatial weight matrix W based on taxi traffict：
In the formula (I), the compound is shown in the specification,
wherein f isijRepresenting the total taxi flow taking the area i as a starting point and the area j as a target point;
s3-3: constructing a normalized mixed space weight matrix W:
in the formula (I), the compound is shown in the specification,
therein
And
respectively, the distance-based normalized spatial weight matrix WdAnd the normalized space weight matrix W based on taxi traffictIs/are as followsAnd
a and b are weight parameters and satisfy a + b ═ 1;
s3-4: the m neighboring regions of the region i are the m regions corresponding to the first m large elements in the ith row of the normalized hybrid spatial weight matrix W.
3. A method for predicting the number of criminal cases based on spatio-temporal data and neural networks as claimed in claim 1 or 2, characterized in that: in step 3, the BP neural network model of each region i is represented as:
in the formula (I), the compound is shown in the specification,
represents the set of the region i and its m neighbors, zj(s) represents the number of crime cases for zone j over time period s,
representing connection rights between an input node and a hidden layer nodeValue, αkRepresenting the connection weight between the hidden layer node and the output node, βkAnd representing a threshold value of a hidden layer node, wherein a hidden layer activation function F is a sigmoid function, and an output layer activation function F is a linear function.
4. A method for predicting the number of criminal cases based on spatio-temporal data and neural networks as claimed in claim 1 or 2, characterized in that: in the step (4), the values of m, n and c in the "m adjacent regions", "n time periods in the past" and "the number of nodes of the hidden layer" in the step (3) are continuously changed, a training set and a test set are used for training and testing the BP neural network model of each region, and the average absolute error of the test set is calculated, so that the superior values of m, n and c are determined.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910973743.XA CN110750609A (en) | 2019-10-14 | 2019-10-14 | Method for predicting number of criminal cases based on space-time data and neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910973743.XA CN110750609A (en) | 2019-10-14 | 2019-10-14 | Method for predicting number of criminal cases based on space-time data and neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110750609A true CN110750609A (en) | 2020-02-04 |
Family
ID=69278205
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910973743.XA Pending CN110750609A (en) | 2019-10-14 | 2019-10-14 | Method for predicting number of criminal cases based on space-time data and neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110750609A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112308316A (en) * | 2020-10-30 | 2021-02-02 | 浙江警察学院 | Crime number prediction method based on linear regression algorithm |
| CN112906970A (en) * | 2021-03-03 | 2021-06-04 | 上海融港网络科技有限公司 | Crime prediction method, crime prediction device, crime prediction terminal and storage medium |
| CN113919160A (en) * | 2021-10-14 | 2022-01-11 | 南京审计大学 | Fine-grained urban crime prediction method and system |
| CN114418244A (en) * | 2022-03-29 | 2022-04-29 | 北京零点远景网络科技有限公司 | Case prediction analysis method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150073759A1 (en) * | 2013-09-06 | 2015-03-12 | Public Engines, Inc. | Prediction accuracy in a spatio-temporal prediction system |
| US9129219B1 (en) * | 2014-06-30 | 2015-09-08 | Palantir Technologies, Inc. | Crime risk forecasting |
| CN107563122A (en) * | 2017-09-20 | 2018-01-09 | 长沙学院 | The method of crime prediction of Recognition with Recurrent Neural Network is locally connected based on interleaving time sequence |
| CN108805345A (en) * | 2018-06-01 | 2018-11-13 | 广西师范学院 | A kind of crime space-time Risk Forecast Method based on depth convolutional neural networks model |
| CN108876062A (en) * | 2018-08-13 | 2018-11-23 | 湖北经济学院 | A kind of big data method and device of crime dramas intelligent predicting |
| CN110309935A (en) * | 2019-03-26 | 2019-10-08 | 浙江工业大学 | A kind of method of crime prediction based on improvement STARMA model |
-
2019
- 2019-10-14 CN CN201910973743.XA patent/CN110750609A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150073759A1 (en) * | 2013-09-06 | 2015-03-12 | Public Engines, Inc. | Prediction accuracy in a spatio-temporal prediction system |
| US9129219B1 (en) * | 2014-06-30 | 2015-09-08 | Palantir Technologies, Inc. | Crime risk forecasting |
| CN107563122A (en) * | 2017-09-20 | 2018-01-09 | 长沙学院 | The method of crime prediction of Recognition with Recurrent Neural Network is locally connected based on interleaving time sequence |
| CN108805345A (en) * | 2018-06-01 | 2018-11-13 | 广西师范学院 | A kind of crime space-time Risk Forecast Method based on depth convolutional neural networks model |
| CN108876062A (en) * | 2018-08-13 | 2018-11-23 | 湖北经济学院 | A kind of big data method and device of crime dramas intelligent predicting |
| CN110309935A (en) * | 2019-03-26 | 2019-10-08 | 浙江工业大学 | A kind of method of crime prediction based on improvement STARMA model |
Non-Patent Citations (2)
| Title |
|---|
| MEILIN LIU ET AL: "A Hybrid Model of Crime Prediction", 《JOURNAL OF PHYSICS: CONFERENCE SERIES》 * |
| 李卫红 等: "改进的GA-BP神经网络模型在财产犯罪预测中的应用", 《武汉大学学报• 信息科学版》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112308316A (en) * | 2020-10-30 | 2021-02-02 | 浙江警察学院 | Crime number prediction method based on linear regression algorithm |
| CN112906970A (en) * | 2021-03-03 | 2021-06-04 | 上海融港网络科技有限公司 | Crime prediction method, crime prediction device, crime prediction terminal and storage medium |
| CN112906970B (en) * | 2021-03-03 | 2022-12-30 | 上海融港网络科技有限公司 | Crime prediction method, crime prediction device, crime prediction terminal and storage medium |
| CN113919160A (en) * | 2021-10-14 | 2022-01-11 | 南京审计大学 | Fine-grained urban crime prediction method and system |
| CN114418244A (en) * | 2022-03-29 | 2022-04-29 | 北京零点远景网络科技有限公司 | Case prediction analysis method and device, electronic equipment and storage medium |
| CN114418244B (en) * | 2022-03-29 | 2022-07-08 | 北京零点远景网络科技有限公司 | Case prediction analysis method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110750609A (en) | Method for predicting number of criminal cases based on space-time data and neural network | |
| Stec et al. | Forecasting crime with deep learning | |
| Bogomolov et al. | Moves on the street: Classifying crime hotspots using aggregated anonymized data on people dynamics | |
| CN107977673B (en) | Economic activity population identification method based on big data | |
| CN107609682B (en) | Medium-short term early warning method for population aggregation in big data environment | |
| CN108650614B (en) | Mobile user position prediction method and device for automatically deducing social relationship | |
| Yu et al. | Crime forecasting using spatio-temporal pattern with ensemble learning | |
| Ghurye et al. | A framework to model human behavior at large scale during natural disasters | |
| US8838134B2 (en) | Method and computer programs for the construction of communting matrices using call detail records and a use for providing user's mobility information | |
| Chen et al. | Designing efficient and balanced police patrol districts on an urban street network | |
| CN113505314A (en) | Position track analysis system for space-time complex network clustering | |
| CN112801411B (en) | Network traffic prediction method based on generation of countermeasure network | |
| Roland et al. | Modeling and predicting vehicle accident occurrence in Chattanooga, Tennessee | |
| Huang et al. | Mining location-based social networks for criminal activity prediction | |
| Cao et al. | Understanding metropolitan crowd mobility via mobile cellular accessing data | |
| Zuo et al. | Detection and Analysis of Urban Area Hotspots Based on Cell Phone Traffic. | |
| Gorr et al. | Chronic and temporary crime hot spots | |
| Ueda et al. | Spatio-temporal multidimensional collective data analysis for providing comfortable living anytime and anywhere | |
| Kurilkin et al. | Evaluation of urban mobility using surveillance cameras | |
| CN110399537A (en) | A kind of alert spatio-temporal prediction method based on artificial intelligence technology | |
| CN116502763A (en) | Crowd congestion prediction method and system based on space-time diagram neural network | |
| Frei et al. | Modeling spatial embedded social networks | |
| Rumi et al. | Modelling regional crime risk using directed graph of check-ins | |
| CN113689957A (en) | Construction method of fine-grained infectious disease simulation model | |
| CN113220675A (en) | College student behavior analysis system and method based on WiFi positioning 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: 20200204 |