Crime Analytics

Increasing crime rates play a crucial role in hindering the growth of a nation as a unit, especially in the case of developing countries like India. This particular development in the crime behavior always has a greater impact on foreign investments as well as international image of the country in the global environment. There is a strong need to monitor and analyze the crimes occurring across the country in addition to regulating them. Though the task of crime regulation is solely under the hands of the government, analysis of the crimes should be carried out in order to derive conclusions and suggest necessary actions and measures extracted from analytical analysis of the past crime database. This method can play a significant role in helping the government in taking better actions for maintaining law and order in the nation and also effectively utilizing their valuable resources.

Crime analytics can be viewed as branch of analytics that deals with methods and applications of statistical tools and techniques to analyze the crime data that will help the police to resolve the crimes at faster rate and get easy insights from the data. It includes the analysis of crimes and criminals, crime victims, disorder, quality of life issues, traffic issues, and internal police operations. And output of crime analysis can be used for criminal investigation and prosecution, patrol activities, crime prevention and reduction strategies, problem solving, and the evaluation of police efforts. Rising crime rate across the globe, increase in different types of crimes and increase in pressure to reduce law enforcement officer have made crime analytics very essential in today’s world.

Thanks to the technological advancements over the past decade, we are able to generate and store terabytes of digital information in the form of emails, videos, chat files, text files, CCTV files, GPS information, etc. and effective analysis of these data can provide evidence central to solving crime and achieving prosecutions. Expecting law enforcing agencies or officials to understand different type of data formats and dig through this volume of data and with less time with insights needed for them is close to impossible. A right tool is very critical to unearth truth or view the required insights from this mountain of data

The International Association of Crime Analysts (IACA) recognizes four categories of crime analysis, ordered from specific to general:

1) Crime intelligence analysis

2) Tactical crime analysis

3) Strategic crime analysis

4) Administrative crime analysis

Crime Intelligence analysis – Crime intelligence analysis is the analysis of data about people involved in crimes, particularly repeat offenders, repeat victims, and criminal organizations and networks.

Tactical Crime Analysis – Tactical crime analysis is the analysis of police data directed towards the short-term development of patrol and investigative priorities and deployment of resources.

Strategic Crime Analysis – Strategic crime analysis is the analysis of data directed towards development and evaluation of long-term strategies, policies, and prevention techniques.

Administrative Crime Analysis – Administrative crime analysis is analysis directed towards the administrative needs of the police agency, its government, and its community.

In India the application of crime analytics is fairly less when compared to western countries. The Chicago Police Department, for instance, has applied a predictive computing model to generate a ‘heat list’ of approximately 400 people most likely to be involved in crime, as victims or offenders. Elsewhere, in San Mateo County, California, police are putting their data to work to predict potential crime scenes. In the city of Memphis only, serious crimes fell to 30 percent and while other violent crimes went down to 15 percent after implementing predictive analytics. In the same lines you can see crime analytics getting implemented at European countries as well.

In India, National Crime Records Bureau (NCRB) is a separate agency which collects crime related data for the last 28 years. It functions as clearing house information on crime and criminals including those operating at National and International levels. NCRB is now using GIS based analytics services to predict and curb the crime rates. It has its very own fingerprints, sketch and all other sort of database.

Key Phases of Crime Analytics

Analytics that runs on the crime data can be viewed under three different phases:

• Pre Crime Analysis

1. Predictive analytics to identify the location where next crime might happen (burglaries and auto theft tend to be clustered in terms of time and location and the individuals committing these crimes tend to have predictable patterns)
2. Analyzing social media data to identify hot zones where riot / assault might happen
3. Preparing a “heat list” of top offenders and probable victims or probable areas can help law enforcement agency plan their monitoring routines or activities
4. Aggregated human behavioral data captured from the mobile network infrastructure, in combination with basic demographic information, can be used to predict crime
5. Analytics software can predict if a prisoner is granted parole will he commit any crime or violate parole conditions

• Run Time Analysis:

This is dynamic and current data analytics and it changing analytics with respect to time. Statistical analysis can be run on every data soon after its completions to arrive at certain numbers like Crime Index. This analysis will helps to get an overview of safety and prevailing conditions as on date.

1. Creating an Area Crime index using various risk factors (bars, clubs, liquor stores, etc.)
2. Social media feeds can be used to identify if any agitation is happening or is going to happen in any particular area

• Post Crime Analysis:

1. Analyzing trends and patterns for various type of crime in different area to provide support accordingly
2. Leads identification after analyzing offenders and modus operandi information.
3. Identifying links between suspects, and their involvement in crimes and criminal activity.

How to predict crime rates

Here the prediction of crime trend is classified into three groups namely,:

1. Prediction of crime trends according to the different types of crimes such as robbery, dacoity, assault of women, child trafficking etc.
2. Prediction of crime trends for the different clusters made according to the different parameters available in the data, such as amount of loss of property, number of lives lost in the crime, number of criminals involved in the crime.
3. Prediction of overall crime trends for any particular area.

Conclusion

New developments in science and technology will help the human race to make their lives simple and happy and on the other hand created lot of new methods of doing different illegal activities. It becomes a huge challenge for any policing organization to effectively control and resolves the crimes that are happening around. Crime analysis is not only bounded in theory, it has many aspects to help security department of any country to take right actions before the crime happens. The current study focused on how the data enabled decisions really makes a different than a traditional approach of solving the crimes.

The analysis helps in three different stages of crime analysis as…

1. Pre Crime analytics will help to get good quality information on different parameters of crime before actually crime happens in the field. The predictive models will help to estimate the different categories of crime so that efficient planning is possible in terms of increasing the preventive measures in the respective areas.
2. Post crime analysis is focused in resolving the crimes rates at a faster rate. Based on the historical patterns of crime, a specific rule based clusters are formed where each cluster is associated with a specific solving principles of crime. Whenever a new crime is characterized as one of the member of the cluster, then it automatically come with a directions to resolve the crime. In other words, we are using historical information of solving the crimes to the current crime resolution.

• There are certain analysis methods we can run in the current time or dynamic time so that instant decisions are made from such analysis. Area crime index is such measure which will created based on current crimes in the respective area and it is dynamic value with respect to time. This Index will help in modifying the current plan with respect to latest changes.

Few recommendation on the current empirical study:

In the current study, we run some models on the crime data pertaining to India and few recommendations were made as follows:

1. This report considers the number of crimes happened in different states in India for the year 2013 to create crime index. Analysis of past 2013 data indicates that Andhra Pradesh is highly crime prone state followed by Kerala and West Bengal.
2. In respect to severity of the crime happened, Uttar Pradesh is showing high index followed by Bihar and Madhya Pradesh. The index number with respect to lower levels of area will help the government agencies to plan the resources in a better way.

With the availability of right amount of data with all other attributes mentioned earlier, it is possible to create a tool for run time analysis of crime that can be a lot of help to take immediate decision to follow. From the historical data, different clusters can be formed according to the similarity of crime types. It can be useful to categorize the crimes and its attributes.

For the further study, we need to collect crime data not only with number of different types of crimes occurred but also with the details such as location (Rural, Urban, Industrial etc.) of crime, the day (weekdays or weekend) and time (day/night) of incidence, details of criminals (where do they stay, age, profession, education, attachment with social media, contact number etc.), distance of crime area from the nearest police station, age and sex of victims etc. These details of the data can be used for further accurate prediction of crime rates in different areas of interest.

Authors: Balasubramanyam Pisupati, Urjaswi Mallik, Prince Rajan

Robert Bosch Engineering and Business Solutions Private Limited

The post Crime Analytics appeared first on Analytics India Magazine.