Data mining (Essay Sample)

448490991712Data Mining in Crime Analysis Nahid Jabeen and Parul Agarwal Abstract Crimes are a worldwide problem that can harm the nation under both social and economic conditions. Crime control is an inescapable step. It is compul- sory for the welfare and sustainable development of a nation. We know very well that in the digital world, it is not an easy task to expose the criminals and the vulner- able areas that are continuously getting affected by their wrongdoings. The police departments of every nation are also continuously working in a paced manner to overcome the crimes, criminals and their techniques. The difficulty in investigating a large amount of data regarding crimes and criminals has become a major challenge for police department officials. An approach is needed that can classify, systemati- cally investigate and forecast the crimes that help to reduce the crime rate. There are various methodologies and paradigms which will help police officials to discover and eliminate crimes from society. Data mining empowers us with several practical and convenient ways to assess large and distinct sets of information. It helps to uncover hidden information from the large database of criminal records for investigating, controlling and preventing crime for organizations and users. Various researchers and data analysts gave their valuable time and knowledge to the field of data mining. The paper focuses primarily on presenting a short and snappy overview of various research papers focused on the techniques of data mining that have been applied in crime analysis. Keywords Data mining · Crime analysis · Criminal investigation · Weka · ARIMA 680427300310 N. Jabeen · P. Agarwal (✉) Department of Computer Science and Engineering, School of Engineering Sciences and Technology, Jamia Hamdard, New Delhi, India e-mail: pagarwal@jamiahamdard.ac.in N. Jabeen e-mail: nahid531996@gmail.com © The Author(s), und97 1 Introduction A human being is not criminal by his\her birth. It is the state of affairs of humans’ social life that make them involved in crime. Crime has the evilest influence on communities which could destroy society. Crimes create problems for a country’s economy by burdening them financially with the additional need for multiple security forces [1]. So it is always needed to analyze the knowledge of crime and keep up with them effectively. The analyzed result will not be completely accurate, but to some degree, it will minimize the crime rate. Crime analysis can be defined as a structured way to identify, locate and predict incidents. It becomes a vital requirement of an approach to clearly understand the future possible crime patterns, such that if a crime cannot be prevented from occurring, at least a preparation can be made with them. The traditional methods needed a large amount of paperwork, manpower and time to fetch patterns of data to predict the possible future crimes [2]. An approach between informatics and criminal justice can be used to establish a data mining technique that can help to solve crime more quickly. Data mining is a technique for extracting hidden data from a generally huge dataset and turning it into useful information for further use [3]. The essence of data mining has become a quickly growing field between criminal investigators and crime analysts. A large number of police department- related records have revealed the need for more than ever to use a systematic and intellectual approach to crime investigation. Depending upon the modern type of criminal investigation, many modern scientific techniques are employed collectively. Data is a valuable asset that is used to link and analyze crime scenes. The crime analysis process involves observation of crime reports and identifying the correlated data such as patterns, series, statistics, maps, etc. These correlated data sets can be used in correlating crimes and fetching patterns that can be identified for making the prediction. For example, when two crimes have happened in a particular place, then they are linked. If the crimes are already in the past, then you can predict potentially associated crimes. In this paper, various techniques and methodologies are described that can enhance the current systems for the efficient investigation process, minimize errors and reduce time complexity. 2 Literature Review In the field of crime data mining and criminal investigation, data mining many scholars had done their research work. The focus of researchers is always in reducing, preventing and enhancing the quality of criminal investigation. A few notable research papers in the related area as follows. According to [4], authors described the crime factors using the relationship between computer science and criminal justice to analyze and predict the crime. Some challenges are also mentioned that are facing by law enforcement officials in identifying crime patterns and trends effectively. The clustering strategy is preferred. The classification algorithm gives the result of the existing and solved crimes. The five steps of crime analysis with some data mining approach, i.e., to analyze the crimes are mentioned in Table 1. According to [5], authors have applied a general data mining system based on the familiarity obtained from the COPLINK project with the support of scholars at the University of Arizona with the assistance of the police departments Tuscon and Phoenix since 1997. Table 2 introduces several new methods of data mining to distinguish structured and unstructured data as described in the paper. According to [6], authors describe the following data mining components in this paper. This paper introduces the crime analysis process into two main components under the two components of COPLINK project, i.e., COPLINK CONNECT and COPLINK DETECT. The techniques and categories of the data mining components are in Table 3. In [7], authors focus on the existing Indian e-governance. This paper describes briefly an interactive query-based interface that is used by the National Crime Record Bureau (NCRB) as a tool for criminal analysis. This paper also gives a reasonable summary of the Indian police system and key characteristics of the new Crime Crim- inal Information System (CCIS) and Common Integrated Police Application (CIPA), along with current status and shortcomings. In [8], authors describe various crime prevention data mining methods using some strategies in the following steps—data extraction, data collection, pre-processing, clustering, classification, pattern predic- tion, and visualization. In this paper, using Google’s marker clustering (GMAPI) and crime-prone locations on the Indian map, the WEKA tool discussed in [9] was used to achieve visualization. To check the selected dataset, random forest and cross-validation are used. Table 4 provides some uses of data mining techniques: According to [2], authors applied clustering techniques to make investigators able to predict and prevent criminal activity. They applied the K-means algorithm in a clustering technique to store the data and forecast the possible result. The result of this paper aimed to make government officials predict crime and criminals based on previous data and location. They also compare some formerly built systems along with the disadvantage of the algorithm being used. In [10], authors worked Table 1 Steps in crime analysis and their approaches S. no. Steps followed in crime analysis Approaches used 1. Data collection MongoDB is used to collect unstructured data 2. Classification Advantages of Naive Bayes Classifier: * It is easy and fast to converge * In the case of probability, it solves zero-frequency problems 3. Pattern identification Apriori Algorithm is used to determine frequently occurring crime in a particular place with the help of association rules 4. Prediction Decision tree 5. Visualization The heat map is used to indicate activity level Table 2 Data mining techniques and their feature S. no Techniques Purposes Applications Limitations 1. Entity extraction This detects specific It is used in police Its output depends on trends from data such narrative reports to the large amounts of as text, images, or automatically classify clean input data audio individuals, being available addresses, vehicles, and personal characteristics 2. Clustering It groups objects It is used to classify High computation technique from data into classes criminals who are intensity is required with similar perpetrating crime in characteristics similar ways or to differentiate between groups of different gangs 3. Association rule This identifies a It is used in the It needs mining commonly occurring detection of network high-structured and dataset and presents intrusion to determine rich data the pattern as the history of contact guidelines between users and to predict possible network attacks 5. Classification It identifies mutual It is used to identify To predict accurately, properties between the source of it requires complete altered entities spamming via e-mail training and testing involved in crime and and to forecast data systematizes them criminal patterns in into predefined less time groups Table 3 Components and techniques of data mining S. no Components Techniques Category 1. Crime entity extraction Named-entity extraction M...