Application of Machine Learning In Forensic Science

Application of Machine Learning In Forensic Science

Mohammad Haroon (Integral University, India), Manish Madhava Tripathi (Integral University, India) and Faiyaz Ahmad (Integral University, India)
Copyright: © 2020 |Pages: 12
DOI: 10.4018/978-1-7998-1558-7.ch013

Abstract

In this chapter, the authors explore the use of machine learning methodology for cyber forensics as machine learning has proven its importance and efficiency. For classification and identification purposes in forensic science, pattern recognition algorithms can be very helpful.
Chapter Preview
Top

Machine Learning

Machine learning is the application of artificial intelligence. Arthur Samuel describes machine learning as a field where the computer learns without any explicit program. A more formal definition of machine learning can be given as: computer trying to learn from its input and output pattern. In general, the machine automatically designs an algorithm with the help of input and output data. Traditionally we design the algorithm then we submit the input data then after we get the result, but in machine learning, first, we trained the machine by submitting input as well as output then machine design the algorithm. Tom Michel provides a more modern definition of machine learning, a computer program is said to learn from own experience during the solving of any task, and the performance of the computer program must be improved with respect to experience like in checker playing game.

  • E = the experience of playing such game

  • T= the task of playing game

  • P= the probability that the program will win the game.

The machine-learning broadly categorized as into main domain

  • 1: supervised learning

  • 2: unsupervised learning

Top

Supervised Learning

In these methods, the learning process will be complete under the supervision of any instructor, means all the output data are already given along with input data set. The system learns on the idea that there is a relationship between input and output data sets.

Supervised learning is further categorized into two subdomains, first is regression and second is classification. In regression Problem, we try to predict the result with the help of continuous function. Meaning that, we're trying to map input variable to some continuous function, in classification problem we trying to map input variable to some discrete function (S. H. B. S. R. Bulkley, 1999).

Key Terms in this Chapter

KNN: KNN is the distance-based algorithm K-nearest neighbor.

Logistic Regression: Logistic regression is used when the response variable is categorical in nature.

Digital forensics: Computer forensics, also known as digital forensics, on the other hand is a much more specific discipline, which involves the analysis of computers and other electronic devices in order to produce legal evidence of a crime or unauthorized action.

Unsupervised Learning: Unsupervised learning is a machine learning technique, where you do not need to supervise the model.

Machine Learning: Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.

Human Learning: Human learning generates knowledge, residing in the brain.

Supervised Learning: In supervised learning, you train the machine using data which is well “labeled.” Supervised learning allows you to collect data or produce a data output from the previous experience.

Complete Chapter List

Search this Book:
Reset