CN110399537B - Artificial intelligence technology-based warning situation space-time prediction method - Google Patents
Artificial intelligence technology-based warning situation space-time prediction method Download PDFInfo
- Publication number
- CN110399537B CN110399537B CN201910660167.3A CN201910660167A CN110399537B CN 110399537 B CN110399537 B CN 110399537B CN 201910660167 A CN201910660167 A CN 201910660167A CN 110399537 B CN110399537 B CN 110399537B
- Authority
- CN
- China
- Prior art keywords
- alarm
- prediction
- time
- prediction model
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Business, Economics & Management (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Computational Linguistics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an artificial intelligence technology-based alarm situation space-time prediction method, which comprises the following steps: data integration, namely constructing a training sample of a prediction model according to the internet data resources such as alarm data, weather and the like; extracting features, namely calling a random forest algorithm in an artificial intelligence technology to extract feature variables; modeling data, and calling a logistic regression algorithm to perform alarm condition prediction model modeling; model checking and evaluating, namely selecting a test sample to check and evaluate the model; and (3) time prediction, namely establishing a historical alarm time sequence in the risk area, and predicting the occurrence time of the alarm through an autoregressive integration moving average model (ARIMA). According to the invention, an artificial intelligence algorithm is adopted, a prediction model is established based on historical alarm situation data, the alarm situation is predicted according to the latest feature element content, a 50m x 50m grid area can be predicted, the prediction result is accurate to a certain time period of a certain area, the prediction result is more objective, and the prediction area is more accurate.
Description
Technical Field
The invention relates to the technical field of warning condition prediction, in particular to a warning condition space-time prediction method based on an artificial intelligence technology.
Background
The continuously developing information technology enables the public security department to obtain more and more crime related (alarm) data, and how to design an effective and convenient method, by analyzing the data, the mode and trend behind the crime data are mined to help the police to better perform security control and crime prevention, which has become a very worthy of research, and the existing alarm prediction methods are mostly based on the alarm data which have occurred in the recent period of time, and the time aggregation degree and the space aggregation degree of the alarm are used for the future period of time, for example: the risk of possible alarm in tomorrow, the next week and the next month is evaluated. At present, the existing warning situation prediction means rely more on artificial experience and the degree of spatiotemporal aggregation to predict, and there is a great uncertainty in prediction accuracy, for example: the case of theft often happens in a certain district in recent period of time, which cannot explain that the case of theft also happens in tomorrow, and the following defects exist in the comprehensiveness of the prediction result: the prediction cannot be accurate to a time period, for example: predicting that the risk of alarm conditions in a certain area is greater at 3 to 6 pm; the prediction area is relatively large, and the risk assessment is performed by taking one street and a dispatched area as objects, so that further improvement is needed.
Disclosure of Invention
The invention aims to provide an alarm situation space-time prediction method based on an artificial intelligence technology, which adopts an artificial intelligence algorithm, establishes a prediction model based on historical alarm situation data, finds out characteristic elements influencing the occurrence of alarm situations in different regions from the perspective of historical rules, predicts the alarm situations according to the latest characteristic element content, meets the timeliness requirement of alarm situation prediction, has more objective prediction results and more accurate prediction regions, can predict 50m-50m grid regions, can predict alarm situation high-occurrence time periods, and can accurately predict the prediction results to a certain time period of a certain region so as to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
an artificial intelligence technology-based alert situation space-time prediction method comprises the following steps:
s1, integrating data, namely establishing a prediction model training data set and a sample of a test data set according to internet data resources such as alarm data, weather and the like, wherein distance attenuation and time intervals in the sample data are obtained by pre-calculation by using a nuclear density interpolation technology based on the alarm occurrence position and time;
s2, extracting features, namely calling a random forest algorithm in an artificial intelligence technology to extract feature variables;
s3, modeling data, and calling a logistic regression algorithm to perform alarm prediction model modeling;
s4, model checking and evaluating, namely selecting a sample of the test data set to check and evaluate the model;
and S5, time prediction, namely generating a time sequence of 8 time periods by taking 24 hours as a period and 3 hours as intervals according to the historical alarm condition of the predicted risk area, and acquiring the time most possibly generating the alarm condition by establishing an autoregressive integration moving average model (ARIMA).
Furthermore, an electronic map is used for dividing the police district into a plurality of 50m-50m grids, each grid unit is a risk area to be predicted, alarm condition data are gathered and integrated, and a training sample of a prediction model is constructed, wherein the training sample is composed of the training data set and the test data set.
Furthermore, a random forest algorithm is used for extracting characteristic variables, and because the calculation time of the random forest is long and the timeliness requirement of warning situation prediction is difficult to meet, the random forest is only used for extracting the characteristics and is not used for modeling.
Furthermore, a logistic regression algorithm is selected, a prediction model is automatically generated, and the situation of alarm is considered to be typical two-classification prediction, so that the situation of alarm is very suitable for two-classification characteristics of logistic regression; meanwhile, because the output of the logistic regression is a probability value, the basic-level policemen are easy to understand and accept; more importantly, the logistic regression allows more training data to be incorporated into the prediction model, and the dynamic updating requirement of the warning situation prediction model can be well met.
Further, a prediction model is examined, evaluated and optimized using a sample of the test data set.
And further, calling the prediction model according to the constructed prediction model, developing an application system, and predicting a risk area of future alarm based on the newly added data.
Further, a historical alarm time sequence in the risk area is established, and the occurrence time of the alarm is predicted through an autoregressive integration moving average model (ARIMA).
Furthermore, frequently changing interference measure data such as directional patrol and the like are timely brought into the training data set, and dynamic updating of the prediction model is achieved.
Compared with the prior art, the invention has the following beneficial effects:
1. an artificial intelligence algorithm is adopted, a prediction model is established based on historical warning situation data, characteristic elements which influence the occurrence of warning situations in different areas are found out from the perspective of historical rules, warning situations are predicted according to the latest characteristic element contents, and the timeliness requirement of warning situation prediction is met;
2. the prediction result is more objective, the prediction region is more accurate, the grid region of 50m × 50m can be predicted, the time interval of high occurrence of the alarm condition can be predicted, the prediction result is accurate to a certain time interval of a certain region, and the accuracy of alarm condition prediction is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention.
FIG. 1 is a flow chart of a method for predicting the warning situation space-time based on the artificial intelligence technology;
FIG. 2 is a detailed flowchart of the alert situation spatiotemporal prediction method based on the artificial intelligence technology of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, and it should be understood that the preferred embodiments described herein are merely for purposes of illustration and explanation and are not intended to limit the present invention.
The method is characterized in that an electronic map is used for dividing a police district into a plurality of 50m-50m grids, each grid unit is a risk area to be predicted, a random forest algorithm in an artificial intelligence technology is called, characteristic variables are extracted, a prediction model is established based on historical alarm situation data, a logistic regression algorithm is selected, an alarm situation prediction model is automatically generated, characteristic elements influencing the occurrence of alarm situations in different areas are found out from the perspective of historical rules, and the alarm situations are predicted according to the latest characteristic element content.
The first embodiment is as follows:
as shown in fig. 1-2, the present invention provides a technical solution: an artificial intelligence technology-based alert situation space-time prediction method comprises the following steps:
s1, integrating data, namely establishing a prediction model training data set and a sample of a test data set according to internet data resources such as alarm data, weather and the like, wherein distance attenuation and time interval in the sample data are obtained by using a nuclear density interpolation technology through precalculation based on alarm occurrence positions and time;
s2, extracting features, namely calling a random forest algorithm in an artificial intelligence technology to extract feature variables;
s3, modeling data, and calling a logistic regression algorithm to perform alarm prediction model modeling;
s4, model inspection and evaluation, wherein a sample of the test data set is selected to perform inspection and evaluation on the model;
and S5, time prediction, namely generating a time sequence of 8 time periods by taking 24 hours as a period and 3 hours as intervals according to the historical alarm condition of the predicted risk area, and acquiring the time most possibly generating the alarm condition by establishing an autoregressive integration moving average model (ARIMA).
In this embodiment, the specific steps of step S1 are: dividing the police district into a plurality of 50m-50m grids by using an electronic map, wherein each grid unit is a risk area to be predicted; and gathering and integrating alarm data, and constructing a training sample of a prediction model, wherein the training sample is composed of the training data set and the test data set.
In this embodiment, in the step S2, a random forest algorithm is used to extract the feature variables, and because the random forest has a relatively long calculation time and is difficult to meet the timeliness requirement of warning situation prediction, the random forest is only used to extract features and is not used for modeling.
In this embodiment, in the step S3, a logistic regression algorithm is selected, a prediction model is automatically generated, and the alarm prediction is considered to be a typical two-classification prediction, so that the logistic regression model is very suitable for the two-classification characteristic of logistic regression; meanwhile, because the output of the logistic regression is a probability value, the basic-level policemen are easy to understand and accept; more importantly, the logistic regression allows more training data to be incorporated into the prediction model, and the dynamic updating requirement of the warning situation prediction model can be well met.
In this embodiment, in step S4, a prediction model is checked, evaluated, and optimized by using the sample of the test data set, and the prediction model is called according to the constructed prediction model to develop an application system, and a risk region where a future alarm will occur is predicted based on newly added data.
In this embodiment, in step S5, a historical alarm time sequence in the risk area is established, the occurrence time of the alarm is predicted by an auto-regressive product-seeking moving average model (ARIMA), and frequently-changing interference measure data such as directional patrol is timely brought into the training data set, so as to dynamically update the prediction model.
Example two:
as shown in fig. 2, the present invention provides a warning situation spatiotemporal prediction method based on artificial intelligence technology, which specifically comprises the following steps:
step A, dividing a police district into a plurality of 50m grids by 50m grids by using an electronic map, wherein each grid unit is a risk area to be predicted;
step B, gathering and integrating alarm data, and constructing a training sample of a prediction model, wherein the training sample consists of a training data set and a test data set;
c, extracting characteristic variables by using a random forest algorithm, wherein the random forest is only used for extracting characteristics but not modeling because the calculation time of the random forest is long and the timeliness requirement of warning situation prediction is difficult to meet;
step D, selecting a logistic regression algorithm, automatically generating a prediction model, and considering that the alarm prediction is a typical two-classification prediction, so that the logistic regression two-classification prediction model is very suitable for the two-classification characteristic of logistic regression; meanwhile, because the output of the logistic regression is a probability value, the basic policemen are easy to understand and accept; more importantly, the logistic regression allows more training data to be brought into the prediction model, so that the dynamic updating requirement of the warning situation prediction model can be well met;
step E, utilizing the sample of the test data set to carry out inspection, evaluation and optimization on a prediction model;
step F, calling the prediction model according to the constructed prediction model, developing an application system, and predicting a risk area of future alarm based on newly added data;
step G, establishing a historical alarm time sequence in the risk area, and predicting the occurrence time of the alarm through an autoregressive quadrature moving average model (ARIMA);
and step I, timely bringing frequently-changing interference measure data such as directional patrol into the training data set, and realizing dynamic updating of the prediction model.
In this embodiment, an application system is developed through a built prediction model, a risk region where an incoming alarm occurs is predicted based on newly added data, the occurrence time of the alarm is predicted through an autoregressive quadrature moving average model (ARIMA), and the dynamic update of the prediction model is realized by including frequently changing interference measure data such as directional patrol in the training data set in time, so that various types of latest alarm can be acquired, and the occurrence of the alarm in a certain period of a certain region in a 50m × 50m grid region can be accurately predicted.
The invention is improved as follows: according to the alarm condition space-time prediction method based on the artificial intelligence technology, an artificial intelligence algorithm is adopted, a prediction model is built based on historical alarm condition data, characteristic elements influencing the occurrence of alarm conditions in different regions are found out from the perspective of historical rules, the alarm conditions are predicted according to the latest characteristic element content, the timeliness requirement of alarm condition prediction is met, the prediction result is more objective, the prediction region is more accurate, a 50m × 50m grid region can be predicted, the high-occurrence time period of the alarm conditions can be predicted, the prediction result is accurate to a certain time period of a certain region, and the accuracy of alarm condition prediction is improved.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. An artificial intelligence technology-based alarm situation space-time prediction method is characterized by comprising the following steps:
s1, integrating data, namely establishing a prediction model training data set and a sample of a test data set according to internet data resources such as alarm data, weather and the like, wherein distance attenuation and time intervals in the sample data are obtained by pre-calculation by using a nuclear density interpolation technology based on the alarm occurrence position and time;
s2, extracting features, namely calling a random forest algorithm in an artificial intelligence technology to extract feature variables;
s3, modeling data, and calling a logistic regression algorithm to perform alarm prediction model modeling;
s4, model checking and evaluating, namely selecting a sample of the test data set to check and evaluate the model;
and S5, time prediction, namely generating a time sequence of 8 time periods by taking 24 hours as a period and 3 hours as intervals according to the historical alarm condition of the predicted risk area, and acquiring the time most possibly generating the alarm condition by establishing an autoregressive integration moving average model (ARIMA).
2. The alarm situation spatiotemporal prediction method based on the artificial intelligence technology as claimed in claim 1, wherein the specific steps of the step S1 are as follows:
step S11, dividing the police district into a plurality of 50m grids by 50m grids by using an electronic map, wherein each grid unit is a risk area to be predicted;
and S12, aggregating and integrating the alarm data to construct a training sample of the prediction model, wherein the training sample consists of the training data set and the test data set.
3. The method for predicting the warning situation space-time based on the artificial intelligence technology as claimed in claim 1, wherein in the step S2, a random forest algorithm is used for extracting the characteristic variables, and because the random forest has longer calculation time and is difficult to meet the timeliness requirement of warning situation prediction, the random forest is only used for extracting the characteristics and is not used for modeling.
4. The method for predicting the alarm situation space-time based on the artificial intelligence technology as claimed in claim 1, wherein in the step S3, a logistic regression algorithm is selected, a prediction model is automatically generated, and the alarm situation prediction is considered to be a typical binary prediction, so that the prediction model is very suitable for the binary characteristics of logistic regression; meanwhile, because the output of the logistic regression is a probability value, the basic-level policemen are easy to understand and accept; more importantly, the logistic regression allows more training data to be incorporated into the prediction model, and the dynamic updating requirement of the warning situation prediction model can be well met.
5. The method according to claim 1, wherein in step S4, a prediction model is examined, evaluated and optimized using the sample of the test data set.
6. The method according to claim 4, wherein the prediction model is invoked according to the constructed prediction model, an application system is developed, and a risk area of future occurrence of an alarm is predicted based on newly added data.
7. The method as claimed in claim 1, wherein in step S5, a historical alarm time series in the risk area is established, and the occurrence time of the alarm is predicted by an auto-regressive integrated moving average model (ARIMA).
8. The artificial intelligence technology-based alarm situation space-time prediction method according to claim 2, wherein frequently changing interference measure data such as directional patrol are timely brought into the training data set to achieve dynamic update of a prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910660167.3A CN110399537B (en) | 2019-07-22 | 2019-07-22 | Artificial intelligence technology-based warning situation space-time prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910660167.3A CN110399537B (en) | 2019-07-22 | 2019-07-22 | Artificial intelligence technology-based warning situation space-time prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110399537A CN110399537A (en) | 2019-11-01 |
| CN110399537B true CN110399537B (en) | 2022-12-16 |
Family
ID=68324974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910660167.3A Active CN110399537B (en) | 2019-07-22 | 2019-07-22 | Artificial intelligence technology-based warning situation space-time prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110399537B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111597461B (en) * | 2020-05-08 | 2023-11-17 | 北京百度网讯科技有限公司 | Target object aggregation prediction method and device and electronic equipment |
| CN113610309B (en) * | 2021-08-13 | 2022-06-03 | 清华大学 | Fire station site selection method and device based on big data and artificial intelligence |
| CN114898889A (en) * | 2021-12-15 | 2022-08-12 | 南京行者易智能交通科技有限公司 | Design method of aggregative risk control model based on big data |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007005975A2 (en) * | 2005-07-01 | 2007-01-11 | Valen Technologies, Inc. | Risk modeling system |
| CN108701274B (en) * | 2017-05-24 | 2021-10-08 | 北京质享科技有限公司 | Urban small-scale air quality index prediction method and system |
| CN108052528B (en) * | 2017-11-09 | 2019-11-26 | 华中科技大学 | A kind of storage equipment timing classification method for early warning |
| CN108269000A (en) * | 2017-12-22 | 2018-07-10 | 武汉烽火众智数字技术有限责任公司 | Intelligent police deployment method and system based on alert big data space-time analysis |
| CN109472419B (en) * | 2018-11-16 | 2021-09-21 | 中山大学 | Method and device for establishing warning condition prediction model based on time and space and storage medium |
-
2019
- 2019-07-22 CN CN201910660167.3A patent/CN110399537B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110399537A (en) | 2019-11-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
| Al-Sharif et al. | A novel approach for predicting the spatial patterns of urban expansion by combining the chi-squared automatic integration detection decision tree, Markov chain and cellular automata models in GIS | |
| Chacon-Hurtado et al. | Rainfall and streamflow sensor network design: a review of applications, classification, and a proposed framework | |
| AU2018101946A4 (en) | Geographical multivariate flow data spatio-temporal autocorrelation analysis method based on cellular automaton | |
| CN110399537B (en) | Artificial intelligence technology-based warning situation space-time prediction method | |
| Jakaria et al. | Smart weather forecasting using machine learning: a case study in tennessee | |
| KR101242808B1 (en) | Method for equitable placement of a limited number of sensors for wide area surveillance | |
| KR101937940B1 (en) | Method of deciding cpted cctv position by big data | |
| CN110414715B (en) | Community detection-based passenger flow volume early warning method | |
| US20240060605A1 (en) | Method, internet of things (iot) system, and storage medium for smart gas abnormal data analysis | |
| Basak et al. | Analyzing the cascading effect of traffic congestion using LSTM networks | |
| CN116703004B (en) | Water system river basin intelligent patrol method and device based on pre-training model | |
| Serra et al. | Exploring countrywide spatial systems: Spatio-structural correlates at the regional and national scales | |
| KR102564191B1 (en) | Disaster response system that detects and responds to disaster situations in real time | |
| Karimi et al. | Developing a methodology for modelling land use change in space and time | |
| CN114970621A (en) | Method and device for detecting abnormal aggregation event, electronic equipment and storage medium | |
| WO2021102213A1 (en) | Data-driven determination of cascading effects of congestion in a network | |
| CN115037559A (en) | Data safety monitoring system based on flow, electronic equipment and storage medium | |
| Nagy et al. | Traffic congestion propagation identification method in smart cities | |
| Meng et al. | A dynamic emergency decision support model for emergencies in urban areas | |
| Barr et al. | Flood-prepared: a nowcasting system for real-time impact adaption to surface water flooding in cities | |
| CN113486754B (en) | Event evolution prediction method and system based on video | |
| CN117671961A (en) | Urban road traffic flow state prediction method and system based on block chain | |
| CN116108198A (en) | Water quality diagnosis method and storage medium for constructing knowledge graph based on big data AI | |
| CN113255593B (en) | Sensor information anomaly detection method facing space-time analysis model |
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 |