International Journal of Mechanical Engineering and Technology (IJMET) Volume 10, Issue 01, January 2019, pp. 1540–1546, Article ID: IJMET_10_01_156 Available online at http://www.iaeme.com/ijmet/issues.asp?JType=IJMET&VType=10&IType=1 ISSN Print: 0976-6340 and ISSN Online: 0976-6359 © IAEME Publication Scopus Indexed USING BIG DATA ANALYTICS TO PREDICT AND REDUCE CYBER CRIMES Dr. Fatma Mohamed Abdullah College of law, Prince Sultan University, Saudi Arabia Fellowship from Higher Education Academy, UK Fabdullah@psu.edu.sa ABSTRACT The challenges of crime prevention in today's world are increasingly complex and require new technologies that can handle the huge amount of confidential data generated through different sources. Cybercrime is an important issue for research as it affects many mainstream sectors such as defences, social media, government, industry, private, military and scientific sectors. etc. Internet criminals use distorted or hacked data to capture their actions. The development of the unpredictability of digital interventions requires efficiency specialized and judicial examination. The cybercrime rate is increasing and challenging the investigative personnel. Crimerelated data is increasingly being generated nowadays and is often digital in nature. At present, data generated cannot be efficiently handled using traditional analysis techniques. Instead of using traditional data analysis techniques it would be useful to use big data analyses for this massive data. This paper will examine the analyse of such data that were generated largely using big data analyses to provide their analysis groups to the artificial neural network which in turn produces a crime prediction pattern. So The police department can use the prediction pattern to allocate its resources in order to reduce the crime rate. Keywords: Cybercrimes, Big data, Crime Predictive Model, Cybercriminals, attacks, artificial neural networks Cite this Article: Dr. Fatma Mohamed Abdullah, Using Big Data Analytics to Predict and Reduce Cyber Crimes, International Journal of Mechanical Engineering and Technology 10(1), 2019, pp. 1540–1546. http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=10&IType=1 1. INTRODUCTION In today's digital age, most people are heavily dependent on technology for everyday tasks. With the rise of technological advances, cyber-attacks also increased. Over the past few years, there have been numerous security infractions. When the penetration of sensitive data, both enterprises and consumers are affected. http://www.iaeme.com/IJMET/index.asp 1540 editor@iaeme.com Using Big Data Analytics to Predict and Reduce Cyber Crimes Increasing crime day by day is the main issue facing human society. The crime occurs when the personal space or the workspace of the criminal and the target intersects at one point. The target may be one person or group of people or it can be a territory. The crime may be accidental or planned. The accidental crime is regrettable and unexpected. An unintended crime occurs in many places. The group of people fights with others because of a small issue that may harm people who have nothing to do with it. Planned crime is a crime committed intentionally. The person who intends to commit the offense, primarily research the target or target area and accordingly study them for implementation. Segregated places have greater chances of crime, with police patrols less than. [1] Earlier, data on crime are mostly police complaints, news reports and articles available in handwritten or printed form, but with technological development, crime data is available in hard copy as well as electronic version format. The previous scenarios differ with the low crime rate, and the data generated on criminal activities were also low. That the amount of less traditional data analysis techniques are effective data to analyse and predict crime. Previous data on criminal activities play a vital role in mapping crime and predicting where crime can occur. Analysing those previously available data was a very tedious and timeconsuming task by traditional data extraction techniques although data were much lower. Data generation nowadays is vast due to the increase in the crime rate, which cannot be addressed by conventional data analysis techniques. These large generated data are large data that can be easily processed with Big Data Analytics [2]. Digital data may be organized, semi-structured or unorganized. The digital data analysed so far has been a systematic type of data to predict crime. Structured data can be considered as ordered data in tabular format with the help of appropriate rows and columns. Previous data is useful for predicting volatile places or saying hot spots. After applying some data extraction techniques such as aggregation, classification and other techniques, locations where there were more opportunities for crime were identified, and police capacity could be allocated there. At present, Internet usage is increasing rapidly. The use of the Internet is also responsible for providing communication between criminals to complete their targeted mission. Thus, data generation is large and often consists of semi-structured or unstructured data and can be analysed using large data collection [3]. To analyse such a huge amount of data either in a semi-structured or unorganized form, traditional data mining techniques are not capable of doing so. For this purpose, large data analysis data is used. Crime mapping is used to analyse, map and visualize crime events or crime patterns to be an idea of predicting crime. Thus mapping security crime as well as the police will help to absorb their resources accordingly to prevent crime. Earlier crime maps could be made by a few people who owned special tools. At present, both scientists and practitioners have the ability to map the crime using available spatial data and with the help of advanced technology. Crime mapping is therefore carried out mainly to reduce crime from society by identifying hotspots places where crime can occur at a higher rate. [4] 2. THEORY OF CRIME PREDICTION Many researchers tried to explain why crime in areas or are there any style can be concluded from past events. One of these theories that answering these questions is the theory of crime prevention [5]. According to the crime did not happen at random, either planned or opportunism. Provides that any criminal activity occurs when there is an intersection in the workspace with the goal and the aggressor. People’s place consists of places visited by the daily routine, like the workplace, educational institutions and shopping centers etc. These are called selected sites of the offender or the victim, called nodes too. The paths that people take each day are connected to many nodes that create an area of personal space. This personal http://www.iaeme.com/IJMET/index.asp 1541 editor@iaeme.com Dr. Fatma Mohamed Abdullah area is also a person's awareness space. Thus, the crime pattern theory states that a crime involving two persons can occur only when the personal space intersects at one point or another. Thus we can say that crimes are not entirely random, but can be studied and analysed to provide favourable expectations. May not be as accurate as those appearing in the film Minority but nevertheless some expectations can be expected. [5] Simply put, if the area provides an opportunity for the offender to commit the crime, and within the victim's personal awareness, then the crime will occur. Thus, isolated areas that do not have any proper patrols provide a greater chance of crime. Although places like shopping and entertainment are places where the culprit and the victim are likely to meet. The reason for this is that there are a large number of people visiting places there and criminals can easily place their potential victims. The study of human behaviour is beyond our scope of study, as we are only interested in finding a pattern that can prevent further crimes. Such as one example of identifying places where many people are victims of a kidnapping chain or pocket selection. This is concentrated mainly in certain areas only [1]. Thus, the crime pattern theory provides an organized way to proceed in predicting the exploration and analysis of crime patterns. [5] 3. CLASSIFICATIONS AND TYPES OF CYBERCRIME The concept of cybercrime is very different from conventional crime. Also due to the growth of Internet technology, this crime has gained serious and unrestricted attention compared to traditional crime. It is therefore necessary to examine the peculiarities of cybercrime. Gordon and Ford tried to create a conceptual framework that is the law Manufacturers can be used when compiling meaningful legal definitions from a technical and social perspective. Under their scheme, electronic crimes are classified into two types: 1. The first type has the following characteristics: It is usually an individual event, or separate, from the point of view victim. This is often facilitated by the introduction of crime programs. Such as keystrokes, viruses, rootkits, or Trojans on a computer used in the system. Introductions (but not necessarily) of weaknesses can be facilitated. At the opposite finish of the spectrum is that the second sort of crime, which Includes, however, isn't restricted to, activities like cyber stalking and harassment, Extortion and manipulation of the stock market, corporate spy complex Planning or carrying out terrorist activities on the Internet. Properties of this Type as follows: Generally facilitated by programs that do not fit the classification Of crime. For example, conversations may occur using IM , and clients or files may be transferred using the FTP protocol. There are frequent contacts or events from a point of view the user. [6] As we mentioned earlier that cybercrime is different from Traditional crime. The same as the traditional cybercrime. Crime also forms of many kinds. Some of the types Cybercrime as described in Figure bellow as cybercrime Evolve with the invention of a new technique itself. http://www.iaeme.com/IJMET/index.asp 1542 editor@iaeme.com Using Big Data Analytics to Predict and Reduce Cyber Crimes 4. REASONS BEHIND CYBERCRIME There are many reasons why cybercriminals do cybercrime; For quick money. To fight an issue that one believes he believes in. Low marginal cost of online activity due to global access. Catching by law enforcement agency is less effective and more expensive. New opportunity to do legal work using technical engineering. Official investigation and criminal prosecution are rare. There is no concrete regulatory measure. Lack of reports and criteria Difficult to identify Limited media coverage. Internet companies are committed collectively and not by individuals [8, 7]. http://www.iaeme.com/IJMET/index.asp 1543 editor@iaeme.com Dr. Fatma Mohamed Abdullah 5. THE CHALLENGES OF CYBERCRIME An endless discussion there regarding the pros and cons Cyber Crime. There are many challenges ahead Combating cybercrime. Some of them are here Annotated below: Lack of trained and qualified manpower Implementation of countermeasures. There is no policy for a special email account to defend Troops, police and security agency personnel. The cyber-attacks did not occur only by terrorists Also from neighbouring countries unlike our National interests. Lack of awareness and cybersecurity culture, Both individually and organizationally. Minimum eligibility to join the police Does not include any knowledge of the computer sector. They are almost illiterate for cybercrime. Speed changes Internet technology always beats Provide government. Sector so that they cannot Identify the origin of these cybercrime [8]. Promotion of research and development in ICT is not Follow the mark Security forces and law enforcement personnel are Not equipped to handle high-tech crimes. The current protocols are not self-sufficient, which Establishes investigative responsibility for crimes This extends internationally. Budgets for security purposes by the government Especially for law enforcement training, Security officers and ICT investigators are less Compared to other crimes. [10]. 6. BIG DATA MECHANISM IN ANALYTICS CYBERCRIME The crime analysis tool must be able to accurately and efficiently identify crime patterns for future prediction and the pattern of micro-crime. The use of hotspot groups was identified as the basis for the predictive algorithm. Thus, the Hotspot concept helps predict crime. Depending on the obsolete events, they are used to identify areas where crime incidents are rising so that appropriate police resources can be deployed at designated locations. Analysis of crime data can be performed using data extraction techniques using tools such as the Weka tool, Quick Tool, R Tool, KNIME, ORANGE, Tanagra, q-radar etc. The crime data analysis is performed using the k-means assembly technique to extract the data. Because of technological development, criminals use their technological equipment to carry out the crime. These digital data are used to analyse crime. The crime analysed will be useful in predicting hot spots. Again, the data used in the analysis and for the purpose of predicting the use of data mining are structured data when there is unorganized or semi-structured data, and data mining techniques consume a lot of time at that moment. Obtained criminal data, preparing that data for a simple quick tool and performing k-clustering on that data for groups. After obtaining the clusters, analyse these clusters to predict the crime [3] Other data extraction techniques can also be applied to the analysis of crime data and can be predicted to identify hotspots. The other technology mainly includes classification, the Kassembly algorithm means the expectation algorithm, etc. After applying the k algorithm, it may provide improved results and what we get after applying k-means clustering only. The KMeans algorithm can be applied to Big Data Analytics. These are some of the traditional and time-consuming methods of crime mapping because they require more structured data.[11,12] The distribution of data related to the crime geographically is also a daunting task but can now be implemented using tools such as the R tool. With some spatial geospatial packages that need to be installed and run with the R tool will significantly affect the data to be http://www.iaeme.com/IJMET/index.asp 1544 editor@iaeme.com Using Big Data Analytics to Predict and Reduce Cyber Crimes distributed to geographic regions. These criminal data can be compiled using appropriate technology. In Big Data Analytics, the GA-based assembly (genetic algorithm) can also be performed to analyse or perform the assembly. The R tool is used to distribute the data geographically. This tool is capable of generating geographic representation of geographically distributed data. Different packages are available with this tool that need to be installed for the data distribution procedure. Data analysis as well as different types of distributed data can be obtained from this tool. [13] Artificial neural network is a group of different nerve cells or nodes (treatment elements) that give predictions based on available data or data collected. The prediction accuracy of the artificial neural network is often very high compared to other systems such as Fuzzy Logic Series or Bayesian Network. The main drawback of the artificial neural network is that it takes some time to learn the implementation of the artificial neural network. [14] Finally, we can have summarized the procedures to: the data collected will be distributed primarily on the geographical location and on the basis of group creation. In the second stage, groups created using Big Data Analytics are analysed. Finally, clusters are analysed into the artificial neural network that will produce the prediction pattern. This pattern of predictions can be used by the security authorities to allocate resources that help to reduce crime. 7. CONCLUSIONS Through the proposed system, it is presented as to how big data in the field of cybercrimes recognition can be accommodated Often it says about how to manage things and become easy when part analysis Become robust while analysing complex data sets and a variety of data. It usually becomes Compelling improved techniques that can be included to avoid or prevent cyber-attacks and cybercrime as well. Also, the analysis of the data by the previously mentioned techniques, there can be many threats without specific problems before eliminating by the use of Weka tool, Quick Tool, R Tool, KNIME, ORANGE, Tanagra and q-radar mechanism. Data can also be analysed for Knowledge and Implementation result in various applications. This process adds an effective approach that can eliminate cybercrime. REFERENCES [1] Saoumya, Anurag Singh Baghel: A Predictive Model for Mapping Crime Using Big Data Analytics, International Journal of Research in Engineering and Technology, IJRET, Volume: 04 Issue: 04 | Apr-2015 [2] Lenin Mookiah, William Eberle, AmbareenSiraj, Survey of Crime Analysis and Prediction, Proceedings of the twenty-Eighth International Florida Artificial Intelligence Research Society Conference, 2015 [3] RenukaNagpal, RajniSehgal, Crime Analysis using K-Means Clustering, International Journal of Computer Applications (0975 – 8887) Volume 83 – No4, December 2013 [4] Vikas Grover, Richard Adderley, Max Bramer, Review of Current Crime Prediction Techniques [5] Balkin, S. D., & Ord, J. K. Automatic neural network modelling for univariate time series. International Journal of Forecasting, 2000,16, P.509–515. [6] Gordon, S., & Ford, R.On the definition and classification of cybercrime. Journal in Computer Virology, 2(1), 13–20.2006 http://www.iaeme.com/IJMET/index.asp 1545 editor@iaeme.com Dr. Fatma Mohamed Abdullah [7] Govil, J., Ramifications of Cyber Crime and Suggestive Preventive Measures, in International Conference on Electro/Information Technology, 2007 IEEE. 2007: Chicago, IL. p. 610-615. [8] Jones, A., Technology: illegal, immoral, or fattening? in Proceedings of the 32nd annual ACM SIGUCCS fall conference. 2004, ACM: Baltimore, MD, USA. p. 305-309. [9] Roshan, N., What is cyber Crime. Asian School of Cyber Law, 2008: Access at http://www.http://www.asclonline.com/index.php?titl e=Rohas_Nagpal, [10] Majesty, H., Cyber Crime Strategy, S.o.S.f.t.H. Department, Editor. 2010, The Stationery Office Limited: UK. p. 42. [11] KeshavSanse, Meena Sharma, Clustering methods for Big data analysis, IJARCET, Volume 4, Issue 3, March 2015. [12] Leong, K., and Chan, S. C. A content analysis of web-based crime mapping in the world’s top 100 highest gdp cities. Crime Prevention & Community Safety 15(1):1–22.2013 [13] Nivranshu Hans, Sana Mahajan, SN Omkar, Big Data Clustering Using Genetic Algorithm On Hadoop Mapreduce, International Journal of Scientific & Technology Research Volume 4, Issue 4, April 2015. [14] Setu Kumar Chaturvedi, Nikhil Dubey, A Survey Paper on Crime Prediction Technique Using Data Mining, Int. Journal of Engineering Research and Applications, Vol. 4, Issue 3(version 1), March 2014 http://www.iaeme.com/IJMET/index.asp 1546 editor@iaeme.com