NARCO ANALYSIS AND DEEP LEARNING

Page 1

International Research Journal of Engineering and Technology (IRJET)

e-ISSN: 2395-0056

Volume: 09 Issue: 10 | Oct 2022

p-ISSN: 2395-0072

www.irjet.net

NARCO ANALYSIS AND DEEP LEARNING Prof.Roopa1, Sunil Kumar H S2 1Assistant Professor

(MCA), Dept of MCA, Vidya Vikas Institute of Engineering and Technology, Mysore, Karnataka, India 2 UG Scholar (), Dept of Electric and Electronic communication(ECE), Maharaja Institute Technology, Mysore, Karnataka, India ---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - The complexity of criminal investigations has

1.1Objective

been brought on by the development of technology together with increasing individual freedom, a decline in the influence of society, church, and family, and an increase in crime intensity. Always keep the law in step with societal trends. In order to deal with the evolving nature of crimes and offenders' tactics, the criminal justice system needs also be developed. Therefore, investigatory organisations are adopting a variety of contemporary technologies to get the truth from the offender and any eyewitnesses engaged in the crime through the introduction of technological innovation. The most widely utilised tests all across the world, including India, are the Lie Detector, BEAP, and Narco-analysis tests. The use of narco-analysis tests in criminal investigations is one of the most hotly contested topics right now. Mathematical functions are used in Deep Learning, a subset of Machine Learning, to map the input to the output. In order to establish a connection between the input and the output, these functions can extract non-redundant data or patterns from the data.

The Narco-analysis test's goal is to recover a person's usage of drugs through their imagination, but because the test will neutralize their imagination because they will access their subconscious minds, it is assumed that whatever they say is spontaneous and accurate. According to the experts' findings, the accused's comments are captured on audio and video tapes.

1.2 Scope Criminal investigations are no longer immune to the impacts of current technological advancements in all aspects of life, which made it necessary to develop scientific instruments to increase the effectiveness of investigative techniques for identifying crimes. As a result, we have seen a surge in the employment of contemporary scientific methods, such as narco-analysis tests, in criminal investigations. This is how modern-day criminals utilise science and technology to carry out their illicit operations.

Key Words: Neural Networks, CNN, RNN, LSTM

2. EXISTING SYSTEM

1. INTRODUCTION

The admissibility of scientific evidence, such as that from narco-analysis, is not explicitly addressed in any law. It is known that 20% of those who undergo narco-analysis are ultimately proven to be innocent; as a result, these procedures not only assist to quickly identify the innocent but also the genuine culprit, motive, and conspiracies, among other things.

"A man's true possession is his memory; without it, he is neither rich nor poor. Crime wave deceives, criminal mind perceives, and criminal soul conceives. —Alexander Smith It is only new that science has been used to look at criminal instances. In the past, courts would rely on papers and other non-scientific evidence, such as testimony from eyewitnesses, but this evidence cannot be relied upon because its veracity cannot be verified. Narco-analysis, a recent innovation in the field of forensic science inquiry, has considerably improved the capacities of forensic science laboratories, "cold cases," and other investigative techniques. These developments in criminal investigation have given new life to cases that had been labelled as unresolved or dead. A larger family of machine learning techniques built on artificial neural networks and representation learning includes deep learning, often referred to as deep structured learning. The three types of learning are supervised, semi-supervised, and unsupervised

© 2022, IRJET

|

Impact Factor value: 7.529

Disadvantages 

If a chemical is administered in the improper amount, it might put a patient in a coma or cause their death.

If the individual is drug dependent, the method is not very successful.

3. PROPOSED SYSTEM The investigative agency uses this scientific test to gather concealed evidence and establish the accused's guilt or innocence. The outcome of such a test can serve as a hint throughout the investigative process since it is a valuable

|

ISO 9001:2008 Certified Journal

|

Page 55


Turn static files into dynamic content formats.

Create a flipbook