KR102151272B1 - Method, apparatus and computer program for analyzing data using prediction model - Google Patents

Abstract

Provided are a data analysis method using prediction model learning, a device and a computer program. According to various embodiments of the present invention, the data analysis method, which is performed by a computing device, includes the following steps of: obtaining a plurality of pieces of initial data; creating basic statistical analysis data about each of the plurality of pieces of initial data or combinatorial data made by combining at least two of the plurality of pieces of initial data by using a basic statistical analysis method; and selecting one of a plurality of pre-stored data analysis models in accordance with the type of an analysis result to be extracted, and creating analysis result data by using the selected data analysis model and at least one piece of data among the plurality of initial data, the combinatorial data and the basic statistical analysis data. Therefore, the method is capable of providing information for monitoring a predetermined area from various angles.

Description

Data analysis method, device and computer program through predictive model learning {METHOD, APPARATUS AND COMPUTER PROGRAM FOR ANALYZING DATA USING PREDICTION MODEL}

Various embodiments of the present invention relate to a data analysis method, an apparatus, and a computer program for analyzing various types of data.

Smart city refers to a city where the infrastructure for tele-communication is connected to every corner of the city like a human neural network, so that the entire city can be controlled through a central control system.

The smart city market has recently shown great interest not only in the domestic market but also in the global market, and accordingly, the growth and spread of smart city technology is rapidly taking place. It is going on.

In general, the central control system of a smart city obtains data from the outside of the city (e.g., population data, CCTV video data, 112 data and 119 data), and employees who monitor the data collect these data output on the dashboard. By monitoring, various events (eg, fire, crime, incident, etc.) occurring in various places in the city can be detected, and various controls can be performed according to the detected events.

However, in the case of the conventional smart city central control system, data obtained from outside is used as it is to monitor or data processed through a simple processing procedure (e.g., basic statistical analysis, etc.). There is a problem that only one-dimensional information can be monitored.

In addition, in the case of the central control system of a conventional smart city, events that may occur in the future are used because previously acquired data (past data) or only real-time data (current data) is used based on the monitoring point. There is a problem that it is difficult to prevent them in advance.

Korean Patent Registration No. 10-2016661 (registered on August 30, 2019)

The problem to be solved by the present invention is a data analysis method capable of generating analysis result data including new information using a plurality of initial data obtained from the outside and processed data (eg, combined data and basic statistical analysis data) , Devices and computer programs.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems that are not mentioned will be clearly understood by those skilled in the art from the following description.

In the data analysis method according to an embodiment of the present invention for solving the above-described problem, in a method performed by a computing device, obtaining a plurality of initial data, each of the plurality of initial data using a basic statistical analysis method Or generating basic statistical analysis data for the combined data by combining two or more of the plurality of initial data, and selecting any one of a plurality of data analysis models previously stored according to the type of analysis result to be extracted. And generating analysis result data using the selected data analysis model and at least one of the plurality of initial data, the combined data, and the basic statistical analysis data.

A data analysis apparatus according to another embodiment of the present invention for solving the above-described problem may include a memory for storing one or more instructions and a processor for executing the one or more instructions stored in the memory, and the processor is the one Obtaining a plurality of initial data by executing the above instruction, generating basic statistical analysis data for each of the plurality of initial data or combined data combining two or more of the plurality of initial data using a basic statistical analysis method Selecting any one data analysis model from among a plurality of pre-stored data analysis models according to the step of performing and the type of analysis result to be extracted, and analyzing the selected data analysis model, the plurality of initial data, the combined data and the basic statistics A data analysis method including generating analysis result data using at least one of the data may be performed.

A data analysis computer program according to another embodiment of the present invention for solving the above-described problems is combined with a computer that is hardware to obtain a plurality of initial data, each of the plurality of initial data using a basic statistical analysis method. Or generating basic statistical analysis data for the combined data by combining two or more of the plurality of initial data, and selecting any one of a plurality of data analysis models previously stored according to the type of analysis result to be extracted. To perform a data analysis method comprising the step of generating analysis result data by selecting and using at least one of the selected data analysis model, the plurality of initial data, the combined data, and the basic statistical analysis data It may be stored in a computer-readable recording medium.

Other specific details of the present invention are included in the detailed description and drawings.

Analysis including new information by analyzing a plurality of initial data obtained from the outside according to various embodiments of the present invention, combined data combining two or more of the plurality of initial data, and basic statistical analysis data generated through basic statistical analysis By generating result data, there is an advantage in that information capable of monitoring a predetermined area from various angles can be provided.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a block diagram of a data analysis system according to an embodiment of the present invention.
2 is a hardware configuration diagram of a data analysis device according to another embodiment of the present invention.
3 is a flowchart of a data analysis method according to another embodiment of the present invention.
4 is a flowchart of a method of generating analysis result data and event occurrence prediction data for a vulnerable area of an event, according to various embodiments.
5 is a flowchart of a method of generating data as a result of predicting a floating population in various embodiments.
6 is a flowchart of a method of generating event occurrence pattern types and indicator data according to various embodiments.
7 is a flowchart of a method of generating event occurrence pattern analysis result data according to a population age in various embodiments.
8 is a flowchart of a method of generating data as a result of analyzing an increase or decrease in event occurrence according to event types, according to various embodiments.
9 is a flowchart of a method of generating analysis result data of a docking station installation area according to various embodiments.
10A to 10C are diagrams illustrating a user interface (UI) for providing analysis result data generated by a data analysis device in various embodiments.

Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in a variety of different forms, only the present embodiments are intended to complete the disclosure of the present invention, It is provided to fully inform the technician of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terms used in the present specification are for describing exemplary embodiments and are not intended to limit the present invention. In this specification, the singular form also includes the plural form unless specifically stated in the phrase. As used in the specification, “comprises” and/or “comprising” do not exclude the presence or addition of one or more other elements other than the mentioned elements. Throughout the specification, the same reference numerals refer to the same elements, and “and/or” includes each and all combinations of one or more of the mentioned elements. Although "first", "second", and the like are used to describe various elements, it goes without saying that these elements are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical idea of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used as meanings that can be commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in a commonly used dictionary are not interpreted ideally or excessively unless explicitly defined specifically.

The term "unit" or "module" used in the specification refers to a hardware component such as software, FPGA or ASIC, and the "unit" or "module" performs certain roles. However, "unit" or "module" is not meant to be limited to software or hardware. The “unit” or “module” may be configured to be in an addressable storage medium, or may be configured to reproduce one or more processors. Thus, as an example, "sub" or "module" refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, It includes procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays and variables. Components and functions provided within "sub" or "module" may be combined into a smaller number of components and "sub" or "modules" or into additional components and "sub" or "modules". Can be further separated.

Spatially relative terms "below", "beneath", "lower", "above", "upper", etc., as shown in the figure It can be used to easily describe the correlation between a component and other components. Spatially relative terms should be understood as terms including different directions of components during use or operation in addition to the directions shown in the drawings. For example, if a component shown in a drawing is turned over, a component described as "below" or "beneath" of another component will be placed "above" the other component. I can. Accordingly, the exemplary term “below” may include both directions below and above. Components may be oriented in other directions, and thus spatially relative terms may be interpreted according to orientation.

In the present specification, a computer refers to all kinds of hardware devices including at least one processor, and may be understood as encompassing a software configuration operating in a corresponding hardware device according to embodiments. For example, the computer may be understood as including all of a smartphone, a tablet PC, a desktop, a laptop, and a user client and an application running on each device, but is not limited thereto.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Each of the steps described herein is described as being performed by a computer, but the subject of each step is not limited thereto, and at least some of the steps may be performed by different devices according to embodiments.

1 is a block diagram of a data analysis system according to an embodiment of the present invention.

Referring to FIG. 1, a data analysis system according to an embodiment of the present invention may include a data analysis apparatus 100, an external terminal 200, and an external server 300.

Here, the data analysis system illustrated in FIG. 1 is according to an exemplary embodiment, and its components are not limited to the exemplary embodiment illustrated in FIG. 1, and may be added, changed, or deleted as necessary.

In one embodiment, the data analysis device 100 may obtain a plurality of raw data from the external server 300, and perform basic statistical analysis on each of the plurality of initial data obtained from the external server 300 Alternatively, basic statistical analysis data for the initial data may be generated by performing basic statistical analysis on the combined data generated by combining two or more of the plurality of initial data. For example, the data analysis apparatus 100 includes information on a total number of initial data or a total number of combined data, an average value, a standard deviation value, a median value, and a mode value using a pre-stored basic statistical analysis model. It is possible to generate basic statistical analysis data. However, it is not limited thereto.

In one embodiment, the data analysis device 100 analyzes any one of a plurality of pre-stored data analysis models according to the type of analysis result to be extracted (eg, the type of analysis result indicated by an input value input from a user). A model may be selected, and analysis result data may be generated by using the selected data analysis model and at least one of a plurality of initial data, combined data, and basic statistical analysis data.

Here, the data analysis model may include a learning model that has been trained using at least one of a plurality of pre-stored initial data, combined data, and basic statistical analysis data as training data, and the analysis result data is generated using the learning model. I can. For example, the learning model may learn a plurality of initial data, combined data, and basic statistical analysis data during a first period prior to the first time point as training data based on a first time point at which the analysis result data is requested. have. Thereafter, the learning model may extract a result value corresponding to the analysis result by using a plurality of initial data, combined data, and basic statistical analysis data during a second period after the first point in time. However, it is not limited thereto.

In various embodiments, the analysis result data generated by the data analysis apparatus 100 may include analysis result data for an event-prone area, event occurrence prediction data, floating population prediction result data, event occurrence pattern type and indicator data, and It may include at least one of event occurrence pattern analysis result data, event occurrence increase/decrease analysis result data, and data collection device (eg, drone) docking station installation area analysis result data. However, it is not limited thereto.

In one embodiment, the data analysis device 100 may visualize a plurality of initial data, combined data, basic statistical analysis data, and analysis result data, and the plurality of visualized initial data, combined data, basic statistical analysis data and analysis A UI (eg, 500 in FIGS. 10A to 10C) outputting at least one of the result data may be provided to the external terminal 200. For example, the data analysis apparatus 100 may provide a UI for providing initial data and combined data according to a user's request (eg, 510 in FIG. 10A), a UI providing basic statistical analysis data (eg, 520 in FIG. 10B) And a UI providing analysis result data (eg, 530 in FIG. 10C) may be provided. However, it is not limited thereto.

In an embodiment, the external terminal 200 may include a display in at least a portion, and may output a screen composed of data visualized through a UI provided from the data analysis apparatus 100 through the display. For example, the external terminal 200 is a dashboard provided in a control center operating a smart phone, a tablet PC, a laptop, a desktop, and a data providing system. It may include at least one of. However, it is not limited thereto.

In various embodiments, the external terminal 200 is connected to the data analysis device 100 by wired or wirelessly, so that a UI may be provided from the data analysis device 100, and a user input (eg, a plurality of visualized analysis) through the UI Among the result data, an input for requesting provision of specific analysis result data, an input for requesting more detailed data (eg, image data) for specific data output on the display may be provided to the data providing device 100. For example, a user may be wirelessly connected to the data analysis device 100 using a communication module included in his or her smartphone, and a UI provided from the data analysis device 100 may be output on the smartphone screen, and A user input indicating specific analysis result data may be provided to the data analysis apparatus 100 by a touch input to the UI output on the phone screen. However, it is not limited thereto.

In an embodiment, the external server 300 may store various types of data obtained from a predetermined area (eg, an area in which a smart city is established). For example, the external server 300 may be a server of the National Police Agency, and stores 112 data including data related to crime-related events and accidents occurring in a predetermined area, data related to crime-related events and accidents, and data related to police dispatch. I can. In addition, the external server 300 may be a 119 safety report center server, and includes data related to fire, rescue, first aid, disaster report, emergency medical care and hospital, reception data, and emergency vehicle dispatch-related data occurring in a predetermined area. 119 data can be stored. However, it is not limited thereto.

In various embodiments, the external server 300 may receive a plurality of initial data transmission requests from the data providing device 100 through the network 400, and a plurality of pre-stored data according to the transmission request of the plurality of initial data. Initial data may be provided to the data analysis device 100.

Here, the data analysis device 100 is described as receiving a plurality of initial data from the external server 300 and generating basic statistical analysis data and analysis result data using the received plurality of initial data, but limited to this. Instead, the data analysis device 100 transmits a control signal for generating data to an external server 300 (for example, a control signal instructing to generate basic statistics analysis data by performing basic statistics analysis and a previously stored data analysis model). It can perform only the function of transmitting a control signal instructing to generate the analysis result data), and the external server 300 uses the control signal received from the data analysis device 100 to provide data according to the control signal. Create can be done.

2 is a hardware configuration diagram of a data analysis device according to another embodiment of the present invention.

Referring to FIG. 2, a data analysis apparatus 100 (hereinafter, “computing apparatus 100”) according to another embodiment of the present invention may include a processor 110 and a memory 120. In various embodiments, the computing device 100 may further include a network interface (or communication interface) (not shown), a storage (not shown), and a bus (not shown). However, it is not limited thereto.

In one embodiment, the processor 110 may control the overall operation of each component of the computing device 100. The processor 110 may be configured to include a CPU (Central Processing Unit), MPU (Micro Processor Unit), MCU (Micro Controller Unit), or any type of processor well known in the technical field of the present invention.

In various embodiments, the processor 110 may perform an operation on at least one application or program for executing the method according to the embodiments of the present invention. In various embodiments, the processor 110 includes one or more cores (not shown) and a graphic processing unit (not shown), and/or a connection path (eg, a bus) for transmitting and receiving signals with other components. can do.

In various embodiments, the processor 110 temporarily and/or permanently stores a signal (or data) processed inside the processor 110, and a RAM (Random Access Memory, not shown) and a ROM (ROM). -Only Memory, not shown) may further be included. In addition, the processor 110 may be implemented in the form of a system on chip (SoC) including at least one of a graphic processing unit, RAM, and ROM.

In an embodiment, the processor 110 may perform a method (eg, a data analysis method) to be described with respect to FIGS. 3 to 9 by executing one or more instructions stored in the memory 120. For example, the processor 110 executes one or more instructions stored in the memory 120 to obtain a plurality of initial data, each of a plurality of initial data or two or more of a plurality of initial data using a basic statistical analysis method. According to the operation of generating basic statistical analysis data for the combined data that combines initial data and the type of analysis result to be extracted, one of the previously stored data analysis models is selected, and the selected data analysis model and the plurality A data analysis method including generating analysis result data using at least one of initial data, combined data, and basic statistical analysis data of may be performed.

In one embodiment, the memory 120 may store various types of data, commands, and/or information. The memory 120 may store programs (one or more instructions) for processing and controlling the processor 110. Programs stored in the memory 120 may be divided into a plurality of modules according to functions.

In various embodiments, steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. Software modules include Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains. Hereinafter, a data analysis method performed by the computing device 100 will be described with reference to FIGS. 3 to 9.

3 is a flowchart of a data analysis method according to another embodiment of the present invention.

Referring to FIG. 3, in step S110, the computing device 100 may acquire a plurality of initial data from the external server 300. For example, the computing device 100 from the external server 300 112 data, 119 data, population data, geographic data for a predetermined area, data including road traffic information including highways, progress in a predetermined area You can get data related to the construction in progress. However, the present invention is not limited thereto, and any data for monitoring the smart city and monitoring the state of the smart city may be applied.

In various embodiments, the computing device 100 may generate combined data by combining and merging two or more of the plurality of initial data. For example, event occurrence data for a predetermined area may be generated by merging 112 event data and 119 event data for a predetermined area. However, it is not limited thereto.

In various embodiments, the computing device 100 may update values of the plurality of initial data and combined data by acquiring a plurality of initial data from the external server 300 every preset period. In various embodiments, the computing device 100 obtains a plurality of initial data from the external server 300 in response to satisfying a preset condition (eg, that a plurality of initial data is changed). You can update the values of data and combined data. However, it is not limited thereto.

In step S120, the computing device 100 may generate basic statistical analysis data by using the plurality of initial data acquired in step S110. For example, the computing device 100 may generate basic statistical analysis data of combined data obtained by combining each of a plurality of initial data or two or more of the plurality of initial data using a pre-stored basic statistical analysis model.

Here, the basic statistical analysis method includes statistical data such as 112 data, 119 data, population data, geographic data for a predetermined area, data including road traffic information including highways, and data related to construction underway in a predetermined area. It refers to a statistical analysis method that organizes or summarizes data according to established criteria. For example, the basic statistical analysis data generated by the computing device 100 may include regional events (eg, crime-related incidents, fires, first aid, rescue-related incidents, etc.) occurrence data, event occurrence data per population, It may include at least one of event occurrence count data for each time slot, event occurrence count data for each day of the week, event occurrence count data for weekdays and weekends, event occurrence count data for each season, event occurrence count data for each event type, and progress time data for each event type. However, it is not limited thereto.

In various embodiments, when a plurality of initial data and combined data are updated, the computing device 100 may automatically perform basic statistical analysis again using the updated plurality of initial data and combined data. In various embodiments, the computing device 100 may repeatedly perform an operation of performing basic statistical analysis on a plurality of initial data and combined data every preset period, regardless of whether the plurality of initial data is updated. Through this, users who monitor various types of data provided by the computing device 100 in real time can monitor updated information, thereby obtaining more accurate information.

In step S130, the computing device 100 may select any one data analysis model from among a plurality of pre-stored data analysis models according to the type of analysis result to be extracted.

In various embodiments, the computing device 100 includes an event generation vulnerable area analysis model, an event occurrence prediction model, a floating population prediction model, an event occurrence pattern type and index model, an event occurrence pattern analysis model according to population age, and event occurrence by event type. A data analysis model that may include at least one data model of an increase/decrease analysis model and a docking station installation area analysis model of the data collection device, and matches the type of analysis result corresponding to the request to generate the analysis result data included in the user input You can choose. However, it is not limited thereto.

In step S140, the computing device 100 uses at least one of the data analysis model selected in step S130, a plurality of initial data and combined data obtained through step S110, and basic statistical analysis data generated in step S120. Analysis result data can be generated. Hereinafter, a method of generating analysis result data according to the type of analysis result to be extracted by the computing device 100 will be described with reference to FIGS. 4 to 9.

4 is a flowchart of a method of generating analysis result data and event occurrence prediction data for a vulnerable area of an event, according to various embodiments.

Referring to FIG. 4, in step S210, when the type of the analysis result to be extracted is the analysis result of an event-prone area among a predetermined area, among a plurality of initial data, combined data, and basic statistical analysis data, the computing device 100 Data related to event occurrence can be selected. For example, the computing device 100 includes 112 data and 119 data among a plurality of initial data, combined data, and basic statistical analysis data, and data related to a predetermined area (e.g., land area data by use, floating population data by time, resident registered population). Population data including information such as youth population, working age population, elderly population, etc., number and ratio data of buildings older than 30 years, publicly announced land price data, CCTV number data by use, average temperature and humidity data can be selected. It is not limited to this.

In step S220, the computing device 100 grids the data related to the occurrence of the event selected in step S210 to a preset unit size, and combines the data related to the occurrence of the gridded event with a basic grid representing a predetermined area to be based on a grid. Event occurrence data in the form of a statistical map can be created. For example, the computing device 100 grids the data related to the occurrence of the event selected in step S210 in units of 500m x 500m, and combines a predetermined area with a basic grid map that is gridd in units of 500m x 500m. Event occurrence data in the form of a statistical map can be created.

In various embodiments, the computing device 100 may use quantum GIS (QGIS) to generate event occurrence data in the form of a grid-based statistical map from data related to event occurrence. However, it is not limited thereto.

In operation S230, the computing device 100 may calculate an event occurrence data value for each of a plurality of grids included in the event occurrence data in the form of a grid-based statistical map. In various embodiments, when each of the plurality of grids has a plurality of event occurrence data values, the computing device 100 may set an average value of the plurality of event occurrence data values as the event occurrence data values of the corresponding grid.

In step S240, the computing device 100 may generate the result data of analyzing the vulnerable area of the event by analyzing the event occurrence data value calculated in step S230 using the zero excess negative binomial regression analysis model. However, the present invention is not limited thereto, and the computing device 100 may generate data as a result of analyzing the vulnerable area of an event by using at least one of an exploratory data analysis model, a correlation analysis model, a random forest model, and an XGBoost model.

In one embodiment, the computing device 100 is a zero inflation model that uses the elderly population data, the publicly announced land price data, the number of CCTVs by use (for example, the number of security CCTVs), the floating population data, and the green area variable data as independent variables. ) (Analytical model for the factors that affect belonging to the group of 0 where no crime will occur in the group with a dependent variable of 0), and the elderly population data, land use data, time zone data, publicly announced land price data, A count model using floating population data and total CCTV data as independent variables (analysis model for the effect of explanatory variables on the dependent variable (number of crime occurrences), which is an additive variable in a non-zero group) can be created.

In various embodiments, the computing device 100 determines whether a data value for each independent variable extracted through a zero inflation model has a meaningful value (eg, when a value less than or equal to a reference value appears) or multicollinearity. The suitability of the model can be judged according to whether or not this occurs, and independent variables can be corrected according to the judgment result.

In various embodiments, in consideration of the fact that the computing device 100 calculates an event occurrence value by taking a log value in the process of generating a zero-excess negative binomial regression model, all coefficients are used to analyze the regression coefficients. You can take an index.

Through the above method, the computing device 100 may calculate a probability of occurrence of an event in the binomial logistic regression analysis through Equation 1 below.

In this case, the computing device 100 may use Equation 2 below to calculate the linear regression equation (β) required for Equation 1 above.

In addition, the computing device 100 may calculate the number of occurrences in the count model using Equation 3 below.

Through this, the computing device may calculate a regression equation of the zero excess negative binomial regression analysis model as shown in Equation 4 below.

Here, the computing device 100 may set the value for the time zone as a dummy variable, and set it to 1 if the time zone is correct and 0 if the time zone is not correct. The computing device 100 may calculate the actual number of crime occurrences and the predicted number of crimes by calculating the finally calculated zero inflation model and the count model, respectively, and through this, indicate an area where crime is vulnerable among predetermined areas. It is possible to generate data as a result of analyzing the vulnerable area of an event including information on the grid.

In step S250, the computing device 100 may generate event occurrence prediction data for a predetermined area by using the data related to the event occurrence selected in step S210 and the analysis result data of the vulnerable area of the event generated in step S240.

In various embodiments, the computing device 100 uses a Long Short Term Memory (LSTM) model among learning models included in a plurality of data analysis models, and uses a plurality of grids from data related to event occurrence and analysis result data of an event vulnerable area. An event occurrence prediction value may be calculated for each, and event occurrence prediction data including an event occurrence prediction value for each of the calculated plurality of grids may be generated.

Here, the LSTM model is used for event occurrence-related data (e.g., 112 data, 119 data, and predetermined time period) for a first period (e.g., a certain period before the first time point based on the first time point at which the event occurrence prediction data is generated). It may be a model that learns (data related to the region of) as training data. The LSTM model may calculate an event occurrence prediction result value at a first time point by learning data related to an event occurrence during a first period. However, the present invention is not limited thereto, and event occurrence prediction data may be generated using a recurrent neural network (RNN) model.

In various embodiments, when receiving a user input for requesting event occurrence vulnerable area analysis result data and event occurrence prediction data from a user, the computing device 100 visualizes event occurrence vulnerable area analysis result data and event occurrence prediction data ( For example, a UI (eg, FIGS. 10A to 10C) that is output in the form of a grid-based statistical map may be provided to the external terminal 200. In this case, the computing device 100 may provide image data (eg, CCTV image data) for an area selected as an area vulnerable to an event according to an analysis result of an area vulnerable to occurrence of an event. However, it is not limited thereto.

5 is a flowchart of a method of generating data as a result of predicting a floating population in various embodiments.

5, in step S310, when the analysis result to be extracted is a floating population prediction result, the computing device 100 may select data related to the floating population among a plurality of initial data, combined data, and basic statistical analysis data. have. For example, the computing device 100 may select floating population data (eg, data representing a floating population value for a predetermined area obtained from a communication company server) among a plurality of initial data, combined data, and basic statistical analysis data. However, it is not limited thereto.

In step S320, the computing device 100 grids the data related to the floating population selected in step S310 to a predetermined unit size, and combines the grid-related data related to the floating population with a basic grid representing a predetermined area to be based on a grid. The floating population data in the form of a statistical map of can be generated (for example, step S220 in FIG. 4). For example, the computing device 100 may generate floating population data in the form of a grid-based statistical map by processing data related to the floating population using QGIS.

In step S330, the computing device 100 may calculate a floating population data value for each of a plurality of grids included in the floating population data in the form of a grid-based statistical map generated in step S320.

In various embodiments, when a plurality of floating population data values are included in one grid, the computing device 100 calculates an average of the plurality of floating population data values, and calculates the average of the calculated floating population data values of the corresponding grid. It can be set as a floating population data value.

In step S340, the computing device 100 may calculate a floating population prediction result value for each grid by performing time series analysis on each of the plurality of grids. For example, the computing device 100 may perform time series analysis on each of a plurality of grids using a prophet model. However, it is not limited thereto.

In step S350, the computing device 100 may generate floating population prediction result data including a floating population prediction result value calculated through time series analysis of each of the plurality of grids in step S340.

In various embodiments, when receiving a user input for requesting floating population prediction result data from a user, the computing device 100 visualizes the floating population prediction result data (eg, in the form of a grid-based statistical map) and outputs a UI ( Example: FIGS. 10A to 10C) may be provided to the external terminal 200. At this time, the computing device 100 visualizes data related to the area where the floating population is concentrated (eg, image data collected from CCTV provided in the area where the floating population is concentrated) according to the floating population prediction result. Can be printed with data.

6 is a flowchart of a method of generating event occurrence pattern types and indicator data according to various embodiments.

Referring to FIG. 6, in step S410, when the analysis result to be extracted is the type and index of the event occurrence pattern, the data related to the event occurrence among a plurality of initial data, combined data, and basic statistical analysis data You can choose For example, the computing device 100 may include event occurrence data for each administrative building, CCTV number data for each type, floating population data, resident registration population data in a predetermined area, and housing types for each administrative building among a plurality of initial data, combined data, and basic statistical analysis data. Among them, data on the number of apartments and detached houses and area data according to land use can be selected. However, it is not limited thereto.

In various embodiments, the computing device 100 may align and process data related to occurrence of a plurality of events selected in step S410 based on an administrative building in a predetermined area.

For example, the computing device 100 calculates the total number of events for each administrative building for 112, 119, and event occurrence data for the socially weak, included in the event occurrence data for each administrative dong, and then calculates the total number by season (spring, summer, Fall and winter).

In addition, the computing device 100 may calculate the total total of the floating population for each administrative building and then sort the total total by account.

In addition, the computing device 100 collects data on the number of residents registered in a predetermined area by administrative building by age group (e.g., the first-aged (0-19 years old), the middle-aged (20-39 years old), the elderly (40-59 years old), and the elderly (60). After separating according to)), the average number of residents registered by year can be calculated. Thereafter, the computing device 100 may divide the number of populations of each age group by the total unit area) and convert it into a ratio.

In addition, the computing device 100 grids data on the number of houses according to residential use, calculates the number of grids having the highest ratio of single-family houses and public houses among the plurality of grids, and reflects the difference in the number of grids by month. It can be used by calculating the average number of houses per star.

In addition, the computing device 100 uses the number of CCTVs for each type, 112, 119 for each administrative building, and CCTVs related to the occurrence of events of the socially disadvantaged (e.g., traffic, crime prevention, illegal parking control, safety management monitoring, and fire monitoring CCTV ) Can only be extracted.

In step S420, the computing device 100 uses a factor analysis method and a clustering analysis method to retrieve data related to the event occurrence selected in step S410 (or data related to the sorted and processed event occurrence). Can be categorized.

First, in order to categorize the event occurrence pattern, the computing device 100 includes a total of 19 indicators (e.g., the number of events per season and administrative dong, the number of floating populations, and the ratio of the resident registration population per unit area (first-aged, middle-aged, elderly, elderly, You can select the number of CCTVs (for traffic, crime prevention, illegal parking and stopping control, safety management monitoring, fire monitoring), and residential use (single housing, multi-family housing).

Next, the computing device 100 extracts a preset number of common factors that minimizes the value of Mean Square Of the Residuals (RMSR) and maximizes the value of Reliability. By integrating into common factors according to their properties, data related to event occurrence can be categorized according to event occurrence patterns.

Next, the computing device 100 is based on the factor loadings obtained by the factor analysis method, an elbow analysis method, a silhouette analysis method, a gap analysis method, and an NB Clust analysis method. The number of clusters (k) may be calculated using any one of the methods.

Next, the computing device 100 uses the calculated number of clusters (k) to calculate the regional classification and factor score average for each cluster by using k-means clustering, which is a non-hierarchical clustering method. I can.

In step S430, the computing device 100 may index data related to the event occurrence selected in step S410 (or data related to the sorted and processed event occurrence) by using a frequency analysis method.

First, the computing device 100 can calculate the event occurrence ratio to the resident population by dividing the total number of events per administrative building by the total number of residents, and dividing the total number of event occurrences by administrative building by the administrative unit area to determine the event occurrence ratio to the unit area. Can be calculated.

Next, the computing device 100 may divide the event occurrence rate relative to the resident population and the event occurrence rate relative to the unit area by a percentile and convert them into scores, and add the scores to set an influence measure.

In step S440, the computing device 100 may generate event occurrence pattern type and indicator data including information on data related to occurrence of the event typed in step S420 and information on data related to occurrence of the event indexed in step S430. .

In various embodiments, when receiving a user input requesting the type and indicator of the event occurrence pattern from the user, visualize the type of the event occurrence pattern and the type and indicator of the event occurrence pattern included in the indicator data, respectively (for example, a table ( A UI (eg, FIGS. 10A to 10C) that is output by performing a table) or a graph may be provided to the external terminal 200.

7 is a flowchart of a method of generating event occurrence pattern analysis result data according to a population age in various embodiments.

Referring to FIG. 7, in step S510, when the analysis result to be extracted is an event occurrence pattern according to the age of the population, data related to the population age among a plurality of initial data, combined data, and basic statistical analysis data And event occurrence data may be selected, and event occurrence data according to the population age may be generated by merging data related to the selected population age and data related to the event occurrence. However, it is not limited thereto.

In step S520, the computing device 100 may calculate the number of event occurrences per population number by analyzing event occurrence data according to the age of the population through the R data analysis module. For example, the computing device 100 calculates the number of annual events between administrative buildings, the total number of populations per 5 years old, the number of men at 5 years old, the number of women at 5 years old, the number of events per total population, and the number of occurrences per event subclass. can do.

In step S530, the computing device 100 may calculate an event occurrence pattern according to the age of the population by using the number of events per population calculated in step S520. For example, the computing device 100 includes the number of populations calculated in step S520 (e.g., the number of events per year between administrative buildings, the total number of people in 5 years old, the number of men in 5 years, the number of women in 5 years) and events By comparing the number (e.g., the number of events per total population, the number of occurrences by detailed event classification) with the number of populations and the number of events compared to the previous year and month, the pattern of the number of population and the number of event occurrences can be calculated.

In step S540, the computing device 100 may generate event occurrence pattern analysis result data according to the age of the population including information on the pattern of the number of populations and the number of events calculated in step S530.

In various embodiments, when the computing device 100 receives a user input requesting data from an event occurrence pattern analysis result according to a population age, the computing device 100 visualizes the event occurrence pattern analysis result data according to the population age (eg, a table and A UI (eg, FIGS. 10A to 10C) that is output through a graph may be provided to the external terminal 200.

8 is a flowchart of a method of generating data as a result of analyzing an increase or decrease in event occurrence according to event types, according to various embodiments.

Referring to FIG. 8, in step S610, when the analysis result to be extracted is an increase or decrease in event occurrence by event type, the computing device 100 selects data related to event occurrence among a plurality of initial data, combined data, and basic statistical data. I can. For example, the computing device 100 may select 112 event occurrence data and 119 event occurrence data among a plurality of initial data, combined data, and basic statistical data. However, it is not limited thereto.

In step S620, the computing device 100 grids the data related to the occurrence of the event selected in step S610 into a preset unit size, and combines the data related to the occurrence of the gridded event with a basic grid representing a predetermined area to be based on a grid. Event occurrence data in the form of a statistical map of may be generated (eg, step S220 of FIG. 4, step S320 of FIG. 5 ).

In operation S630, the computing device 100 may calculate an event increase/decrease data value for each of a plurality of grids included in the event occurrence data in the form of a grid-based statistical map through the R data analysis module.

Also, the computing device 100 may calculate an increase/decrease amount of event occurrence for each event type for each of the plurality of grids by year, month, and day by using event increase/decrease data values for each of the plurality of grids.

In step S640, the computing device 100 may generate event occurrence increase/decrease analysis result data for each event type including information on the event occurrence increase/decrease amount for each event type calculated in step S630.

In various embodiments, when receiving a user input requesting data from a user as a result of analyzing an increase or decrease in event occurrence, the computing device 100 visualizes and outputs the result data of an increase or decrease in event occurrence analysis (eg, tables and graphs). (Example: FIGS. 10A to 10C) may be provided to the external terminal 200.

9 is a flowchart of a method of generating analysis result data of a docking station installation area according to various embodiments.

Referring to FIG. 9, in step S710, the computing device 100 moves the analysis result to be extracted to a predetermined area, and a data collection device (eg, within a predetermined area) that collects data on the predetermined area. In the case of the analysis result of the installation area of the docking station of a drone that collects data on a predetermined area while driving, data related to a predetermined area can be selected from among a plurality of initial data, combined data, and basic statistical data. I can. For example, the computing device 100 may include 119 emergency dispatch data for a predetermined area among a plurality of initial data, combined data, and basic statistical data, traffic network GIS road node link data of the national traffic database, GIS of the national fire department and 119 safety center. You can select coordinate data. However, it is not limited thereto.

In step S720, the computing device 100 calculates the density of an unreachable area when the data collection device operates from a plurality of reference positions at a preset speed for a preset time using data related to a predetermined area selected in step S710. can do. For example, based on a plurality of fire stations located in a certain area, considering the road limit speed, deriving an area that can be reached when driving for 5 minutes at a speed of 40Km/h, 50Km/h or 60Km/h, Through this, the density of the unreachable area can be calculated.

In operation S730, the computing device 100 may select the docking station installation candidate area using the calculated density of the unreachable area. For example, the computing device 100 may select an area that occupies the widest area in a 1Km radius area among the top 25% of areas in which the density of the unreachable area is not reached as a candidate area for reviewing installation of the docking station.

In step S740, the computing device 100 may generate docking station installation area analysis result data including information related to the docking station installation candidate area selected in step S730.

In various embodiments, the computing device 100 is a UI that visualizes (for example, in the form of a density map) and outputs the analysis result data of the docking station installation area when receiving a user input requesting data from the user for analysis result data of the docking station installation area. (Example: FIGS. 10A to 10C) may be provided to the external terminal 200.

The above-described data analysis method has been described with reference to the flowchart shown in the drawings. For the sake of simplicity, the data analysis method has been described by showing a series of blocks, but the present invention is not limited to the order of the blocks, and some blocks are performed in an order different from those shown and described in this specification, or performed simultaneously. I can. In addition, a new block not described in the specification and drawings may be added, or some blocks may be deleted or changed.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or a combination thereof. The software module includes Random Access Memory (RAM), Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), Flash Memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer-readable recording medium well known in the art to which the present invention pertains.

Components of the present invention may be implemented as a program (or application) and stored in a medium in order to be combined with a computer as hardware to be executed. Components of the present invention may be implemented as software programming or software elements, and similarly, embodiments include various algorithms implemented with a combination of data structures, processes, routines or other programming elements, including C, C++ , Java, assembler, etc. may be implemented in a programming or scripting language. Functional aspects can be implemented with an algorithm running on one or more processors.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those of ordinary skill in the art to which the present invention pertains can be implemented in other specific forms without changing the technical spirit or essential features. You can understand. Therefore, the embodiments described above are illustrative in all respects, and should be understood as non-limiting.

100: data analysis device (or computing device)
200: external terminal
300: external server

Claims (11)

In the method performed by the computing device,
Obtaining a plurality of initial data;
Generating basic statistical analysis data for each of the plurality of initial data or combined data obtained by combining two or more of the plurality of initial data using a basic statistical analysis method; And
Select any one of a plurality of pre-stored data analysis models according to the type of analysis result to be extracted, and at least one of the selected data analysis model, the plurality of initial data, the combined data, and the basic statistical analysis data Including the step of generating analysis result data using one data,
The step of generating the analysis result data,
When the analysis result to be extracted moves to a specific area and is a result of analyzing a docking station installation area of a data collection device collecting data on the specific area, the plurality of initial data, the combined data, and the Selecting data related to the specific region from among basic statistical data;
Calculating a density of an unreachable area when the data collection device operates from a plurality of reference positions at a preset speed for a preset time using the data related to the selected specific area;
Selecting the docking station installation candidate area using the calculated density of the unreachable area; And
And generating docking station installation area analysis result data including information related to the docking station installation candidate area. The method of claim 1,
The data analysis model,
Includes a learning model learned as training data of at least one of a plurality of pre-stored initial data, combined data, and basic statistical analysis data,
The step of generating the analysis result data,
And generating the analysis result data using the learning model. The method of claim 2,
The step of generating the analysis result data,
When the type of the analysis result to be extracted is an analysis result of an event occurrence vulnerable area among a predetermined area, data related to the event occurrence among the plurality of initial data, the combined data, and the basic statistical analysis data are set to a predetermined unit size. Latticeizing;
Generating event occurrence data in the form of a grid-based statistical map by combining the gridd event occurrence data with a basic grid representing the predetermined area;
Calculating an event occurrence data value for each of a plurality of grids included in the event occurrence data in the form of the grid-based statistical map; And
Using a zero-excess negative binomial regression analysis model, from the calculated event occurrence data values for each of the plurality of grids, among the plurality of grids, a grid indicating a region where the event is vulnerable is selected, and information on the selected grid is A data analysis method through predictive model learning, comprising the step of generating data as a result of analyzing a vulnerable area of an event that includes. The method of claim 3,
Generating the analysis result data using the learning model,
An event occurrence prediction value for each of the plurality of grids is calculated from the data from the analysis result data of the event-prone area using a Long Short Term Memory (LSTM) model among the learning models, and an event occurs for each of the calculated plurality of grids. A method of analyzing data through predictive model learning, comprising the step of generating event occurrence prediction data including a predicted value. The method of claim 1,
The step of generating the analysis result data,
When the type of the analysis result to be extracted is the result of predicting the floating population for a predetermined area, data related to the floating population among the plurality of initial data, the combined data, and the basic statistical analysis data are gridded in a preset unit size. Step to do;
Generating floating population data in the form of a grid-based statistical map by combining the gridd floating population-related data with a basic grid representing the predetermined area;
Calculating a floating population data value for each of a plurality of grids included in the floating population data in the form of a grid-based statistical map; And
A time series analysis of the calculated floating population data values for each of the plurality of grids calculates a floating population prediction result value for the predetermined area, and generates floating population prediction result data including the calculated floating population prediction result value. A method of analyzing data through learning a predictive model, including the step of. The method of claim 1,
The step of generating the analysis result data,
When the analysis result to be extracted is an event occurrence pattern type and index, selecting data related to the event occurrence from among the plurality of initial data, the combined data, and the basic statistical analysis data;
Typing data related to the occurrence of the selected event using at least one of a factor analysis method and a clustering analysis method;
Indexing data related to the occurrence of the selected event using a frequency analysis method; And
And generating an event occurrence pattern type and indicator data including data related to the occurrence of the tangible event and data related to the occurrence of the indexed event. The method of claim 1,
The step of generating the analysis result data,
When the analysis result to be extracted is an event occurrence pattern according to the population age, data related to the population age and data related to the event occurrence among the plurality of initial data, the combined data, and the basic statistical analysis data are merged to Generating event occurrence data according to;
Calculating the number of event occurrences per population by analyzing the event occurrence data according to the population age through the R data analysis module, and calculating an event occurrence pattern according to the age of the population using the calculated number of event occurrences per population; And
A method of analyzing data through predictive model learning, comprising the step of generating event occurrence pattern analysis result data according to population age including information on the calculated event occurrence pattern according to the population age. In the method performed by the computing device,
Obtaining a plurality of initial data;
Generating basic statistical analysis data for each of the plurality of initial data or combined data obtained by combining two or more of the plurality of initial data using a basic statistical analysis method;
Select any one of a plurality of pre-stored data analysis models according to the type of analysis result to be extracted, and at least one of the selected data analysis model, the plurality of initial data, the combined data, and the basic statistical analysis data Generating analysis result data using one piece of data;
When the analysis result to be extracted is an increase or decrease in event occurrence by event type, gridding data related to event occurrence among the plurality of initial data, the combined data, and the basic statistical analysis data in a predetermined unit size;
Generating event occurrence data in the form of a grid-based statistical map by combining the gridd event occurrence data with a basic grid representing a predetermined area;
Calculating an event increase/decrease data value for each of a plurality of grids included in the event occurrence data in the form of the grid-based statistical map through an R data analysis module; And
Using the event increase/decrease data values for each of the plurality of grids, the event occurrence increase/decrease amount for each of the plurality of grids is calculated by year, month, and day, and the calculated event occurrence increase/decrease amount for each event type is calculated. A data analysis method through predictive model learning, comprising the step of generating data as a result of analyzing an increase or decrease in event occurrence for each event type including. delete A memory for storing one or more instructions; And
And a processor that executes the one or more instructions stored in the memory,
The processor executes the one or more instructions,
An apparatus for performing the method of claim 1 or 8. A computer program that is combined with a computer as hardware and stored in a recording medium readable by a computer to perform the method of claim 1 or 8.

Images