• Title/Summary/Keyword: crime data

Search Result 386, Processing Time 0.02 seconds

Crime amount prediction based on 2D convolution and long short-term memory neural network

  • Dong, Qifen;Ye, Ruihui;Li, Guojun
    • ETRI Journal
    • /
    • 제44권2호
    • /
    • pp.208-219
    • /
    • 2022
  • Crime amount prediction is crucial for optimizing the police patrols' arrangement in each region of a city. First, we analyzed spatiotemporal correlations of the crime data and the relationships between crime and related auxiliary data, including points-of-interest (POI), public service complaints, and demographics. Then, we proposed a crime amount prediction model based on 2D convolution and long short-term memory neural network (2DCONV-LSTM). The proposed model captures the spatiotemporal correlations in the crime data, and the crime-related auxiliary data are used to enhance the regional spatial features. Extensive experiments on real-world datasets are conducted. Results demonstrated that capturing both temporal and spatial correlations in crime data and using auxiliary data to extract regional spatial features improve the prediction performance. In the best case scenario, the proposed model reduces the prediction error by at least 17.8% and 8.2% compared with support vector regression (SVR) and LSTM, respectively. Moreover, excessive auxiliary data reduce model performance because of the presence of redundant information.

한국에서 빅데이터를 활용한 범죄예방시스템 구축을 위한 연구 (A Study on Construction of Crime Prevention System using Big Data in Korea)

  • 김성준
    • 한국인터넷방송통신학회논문지
    • /
    • 제17권5호
    • /
    • pp.217-221
    • /
    • 2017
  • 범죄는 사전적 예방이 중요하다. 과거 범죄는 사후적으로 대처하고 이를 처벌하는데 집중하였다. 그러나 빅데이터 기술을 적용하면 범죄는 사전적으로 예방될 수 있다. 빅데이터는 범죄자 또는 잠재적 범죄자의 행동을 예측할 수 있기 때문이다. 이 글은 범죄예방을 위해 빅데이터 시스템을 어떻게 구축할지에 대해 논의한다. 구체적으로는 빅데이터의 비정형 데이터와 기본의 정형데이터를 결합하는 방식을 다루고 그 결과로서 범죄예방시스템을 설계한다. 이 연구를 통해 범죄 예방을 위해 빅데이터가 활용되는 가능성을 지문을 통해 기술하였고 이를 기초로 향후 범죄예방프로그램 및 연구에 도움을 줄 것으로 기대된다.

Crime hotspot prediction based on dynamic spatial analysis

  • Hajela, Gaurav;Chawla, Meenu;Rasool, Akhtar
    • ETRI Journal
    • /
    • 제43권6호
    • /
    • pp.1058-1080
    • /
    • 2021
  • Crime is not a completely random event but rather shows a pattern in space and time. Capturing the dynamic nature of crime patterns is a challenging task. Crime prediction models that rely only on neighborhood influence and demographic features might not be able to capture the dynamics of crime patterns, as demographic data collection does not occur frequently and is static. This work proposes a novel approach for crime count and hotspot prediction to capture the dynamic nature of crime patterns using taxi data along with historical crime and demographic data. The proposed approach predicts crime events in spatial units and classifies each of them into a hotspot category based on the number of crime events. Four models are proposed, which consider different covariates to select a set of independent variables. The experimental results show that the proposed combined subset model (CSM), in which static and dynamic aspects of crime are combined by employing the taxi dataset, is more accurate than the other models presented in this study.

수사단서를 이용한 동일 사이버범죄 판단기법 (Technique for Indentifying Cyber Crime Using Clue)

  • 김주희
    • 정보보호학회논문지
    • /
    • 제25권4호
    • /
    • pp.767-780
    • /
    • 2015
  • 최근 몇 년간 스마트폰의 보급률이 폭발적으로 증가하면서 사이버범죄는 기존의 수사체계의 한계를 넘어서는 새로운 형태의 수사단서들을 쏟아내고 있다. 일선 경찰관서에서는 사건 접수 시 피해자로부터 이러한 형태의 수사단서를 수집하여 방대하게 축적하고 있으나, 이를 체계적으로 관리하고 있지 않아 많은 데이터 속에서 이것이 내포하고 있는 숨은 의미를 지나치는 경우가 많다. 사이버범죄에서 주 범행 도구인 컴퓨터 시스템의 특성상 기계적이고 복잡한 단서가 대량 생성되므로, 수집된 수사단서를 체계적으로 분류, 단순화하여 분석할 필요가 있다. 본 논문에서는 국내에서 발생되는 사이버범죄 유형에 따른 수사단서를 체계적으로 분류, 단순화하여 주요수사단서를 선정하고, 데이터 마이닝 및 시각화를 통해 사건 수사단서 간 상호 연관성을 확인할 수 있었다. 이러한 사이버범죄 데이터 활용을 통해 범죄 조기차단 및 중복수사를 방지하여 수사의 효율성을 증대하고 사이버범죄 예방을 도모하고자 한다.

소셜 네트워크 서비스의 단어 빈도와 범죄 발생과의 관계 분석 (An Analysis of Relationship Between Word Frequency in Social Network Service Data and Crime Occurences)

  • 김용우;강행봉
    • 정보처리학회논문지:컴퓨터 및 통신 시스템
    • /
    • 제5권9호
    • /
    • pp.229-236
    • /
    • 2016
  • 기존의 범죄 예측 방법들은 범죄 발생을 예측하기 위해 기존 기록을 이용하였다. 그러나 이러한 범죄 예측 모델은 데이터를 갱신하는데 어려움이 있다. 범죄 예측을 향상시키기 위해서 소셜 네트워크 서비스(SNS)를 이용하여 범죄를 예측하는 연구들이 진행되었지만, SNS 데이터와 범죄 기록 사이의 관계에 대한 연구는 미흡하다. 따라서, 본 논문에서는 SNS 데이터와 범죄 발생 사이의 관계를 범죄 예측의 관점에서 분석하였다. 잠재 디리클레 할당(LDA)을 이용하여 범죄 발생과 관련된 단어를 포함하는 트윗을 추출하였고, 범죄 기록에 따른 트윗 빈도의 변화를 분석하였다. 범죄 관련 단어를 포함하는 트윗의 빈도를 계산하고, 범죄 발생에 따라서 트윗 빈도를 분석하였다. 범죄가 발생하였을 때, 범죄와 관련된 트윗의 빈도가 변화하였다. 게다가, 범죄 발생 전후에 트윗 빈도가 특정 패턴을 보이기 때문에 SNS 데이터가 범죄 예측 모델에 도움이 될 것이다.

범죄예측에서의 데이터마이닝 적용 가능성 연구 : 절도범죄를 중심으로 (A Study on the Applicability of Data Mining for Crime Prediction : Focusing on Burglary)

  • 방승환;김태훈;조현보
    • 한국컴퓨터정보학회논문지
    • /
    • 제19권12호
    • /
    • pp.309-317
    • /
    • 2014
  • 최근, 범죄가 증가함에 따라 범죄를 예측하고 예방하는 것은 사회의 중요한 이슈이며 정부 및 지자체는 다양한 방법론을 활용하여 범죄를 사전에 막고자 노력하고 있다. 데이터마이닝은 범죄예측 및 예방에 활용되는 대표적인 방법론이며, 범죄 패턴 분석, 범죄 발생 예측 등 다양한 분야에서 연구되고 있다. 그러나 데이터마이닝의 결과가 범죄학에서의 범죄 환경요소와 어떤 관련이 있는지 혹은, 사건해결에 어떤 도움을 줄 수 있는지에 대한 연구는 이루어지고 있지 않다. 따라서 본 논문에서는, 범죄학에서 범죄의 발생과 범죄 환경요소들의 상호 관련성을 보이고 범죄 발생과 관련된 환경요소와 데이터마이닝에 활용되는 변수 간의 관계를 정의하고자 하였다. 또한, 국내 보호관찰소에서 보관되고 있는 절도범죄 데이터를 사용하여 실제로 데이터마이닝의 결과가 범죄 환경요소와 어떤 관련이 있는지를 보이기 위해 군집분석을 적용하였다. 그 결과 각 군집별로 범죄가 발생하는 환경에 차이가 있었으며, 이를 활용하여 데이터마이닝이 범죄학관점에서 범죄 예측 및 예방 활용에 유의미함을 보였다.

Artificial-Neural-Network-based Night Crime Prediction Model Considering Environmental Factors

  • Lee, Juwon;Jeong, Yongwook;Jung, Sungwon
    • Architectural research
    • /
    • 제24권1호
    • /
    • pp.1-11
    • /
    • 2022
  • As the occurrence of a crime is dependent on different factors, their correlations are beyond the ordinary cognitive range. Owing to this limitation, systems face difficulty in correlating various factors, thereby requiring the assistance of artificial intelligence (AI) to overcome such limitations. Therefore, AI has become indispensable for crime prediction. Crimes can cause severe and irrevocable damage to a society. Recently, big data has been introduced for developing highly accurate models for crime prediction. Prediction of night crimes should be given significant consideration, because crimes primarily occur during nights, when the spatiotemporal characteristics become vulnerable to crimes. Many environmental factors that influence crime rate are applied for crime prediction, and their influence on crime rate may differ based on temporal characteristics and the nature of crime. This study aims to identify the environmental factors that influence sex and theft crimes occurring at night and proposes an artificial neural network (ANN) model to predict sex and theft crimes at night in random areas. The crime data of A district in Seoul for 12 years (2004-2015) was used, and environmental factors that influence sex and theft crimes were derived through multiple regression analysis. Two types of crime prediction models were developed: Type A using all environmental factors as input data; Type B with only the significant factors (obtained from regression analysis) as input data. The Type B model exhibited a greater accuracy than Type A, by 3.26 and 9.47 % higher for theft and sex crimes, respectively.

The Relationship between Residential Distribution of Immigrants and Crime in South Korea

  • Park, Yoonhwan
    • 유통과학연구
    • /
    • 제16권7호
    • /
    • pp.47-56
    • /
    • 2018
  • Purpose - This study aims to not only investigate spatial pattern of immigrants' residence and crime occurrences in South Korea, but shed light on how geographic distribution of immigrants and immigrant segregation affect crime rates. Research design, data, and methodology - Th unit of analysis is Si-Gun-Gu municipal level entities of South Korea. The crime data was obtained by Korea National Police Agency and two major types(violence and property) of crime were measured. Most demographic, social, and economic variables were derived from Korean Census Data in 2015. In order to examine spatial patterns of immigrants' distribution and crime rates in South Korea, the present study utilized GIS mapping technique and Exploratory Spatial Data Analysis(ESDA) tools. The causal linkage was investigated by a series of regression models using STATA. Results - Spatial inequality between urban metropolitan vs rural areas was visualized by mapping. Assuming large Moran's I value, spatial autocorrelation appeared to be quite strong. Several neighborhood characteristics such as residential stability and economic prosperity were found to be important factors leading to crime rate change. Residential distribution and segregation for immigrants were negatively significant in the regression models. Conclusions - Unlike the traditional arguments of social disorganization theory, immigrant segregation appeared to reduce violent crime rate and the high proportion of immigrants also turned out to be a crime prevention factor.

Crime Mapping Based on Experts' and Residents' Assessments of Neighborhood Environment

  • Kim, Jaecheol
    • 한국측량학회지
    • /
    • 제35권4호
    • /
    • pp.213-220
    • /
    • 2017
  • This study examines the limitations of existing crime mapping that relies mainly on reported crime data, suggests a crime mapping method based on experts' and users' assessments of a neighborhood environment as an alternative approach, and conducts a case study on a real-world site by applying the suggested approach. According to the results of the case analysis, while the areas adjoining arterial roads with heavy pedestrian traffic were shown as high crime risk areas in the crime map based on actual reported crime data, the areas adjoining local roads with low pedestrian traffic were high-risk areas in the crime risk area map based on experts' and residents' evaluations. This study makes a contribution to the field in that it demonstrates the detailed application process of crime risk area mapping according experts' and residents' evaluations, compares the results with those of an existing crime map, and finally shows that the former can function as a complement to the latter.

Designing a Crime-Prevention System by Converging Big Data and IoT

  • Jeon, Jin-ho;Jeong, Seung-Ryul
    • 인터넷정보학회논문지
    • /
    • 제17권3호
    • /
    • pp.115-128
    • /
    • 2016
  • Recently, converging Big Data and IoT(Internet of Things)has become mainstream, and public sector is no exception. In particular, this combinationis applicable to crime prevention in Korea. Crime prevention has evolved from CPTED (Crime Prevention through Environmental Design) to ubiquitous crime prevention;however, such a physical engineering method has the limitation, for instance, unexpected exposureby CCTV installed on the street, and doesn't have the function that automatically alarms passengers who pass through a criminal zone.To overcome that, this paper offers a crime prevention method using Big Data from public organizations along with IoT. We expect this work will help construct an intelligent crime-prevention system to protect the weak in our society.