Application Of Artificial Intelligence In Criminal Investigation

There has been a steady increase in the crime rate all over the world. This reflects in a large number of crimes yet to be solved. Crime investigators are overwhelmed by the magnitude of these crimes and often suffer from stress, depression, anxiety, and burnouts when faced with intricate cases. Artificial Intelligence employed at various stages of the Crime Solving Process will accelerate the process of criminal investigation and reduce the pressure faced by Crime Investigators.

Introduction

Crime is an action or omission of a penal law which is punishable by fine, imprisonment or death. In the event of a criminal offence, it is essential for law enforcement agenciesto apprehend and charge the offender. Criminal Investigation is the study of facts that are employed to aid in criminal trials.An entire criminal investigation includes crime scene investigation, searching, interviews, interrogations, evidence collection and preservation and various methods of investigation. Evidence encompasses all the available facts, information and tangible objects that are present in support of an assertion. Judgments and conclusions can be founded upon it.

Evidence can be of the following types:

1. Associative Evidence links a suspect with a criminal offence.
2. Circumstantial evidence is evidence that helps investigators draw inference but is not adequate for conviction.
3. Competent Evidence is evidence which properly collected, named, filed, and continuously secured.
4. Corpus Delicti Evidence establishes that a crime took place.
5. Direct Evidence is complete evidence capable of asserting the truth without any inference and leading to a conviction.
6. Material evidence is evidence relevant to a specific case and forming a substantive part of the lawsuit or having a legitimate influence on the case’s outcome.
7. Trace Evidence is very small evidence such as hair or fibres.
8. Physical Evidence is something tangible that helps to establish the facts of a case.
9. Probative evidence is the grounds necessary to investigate or prosecute a case; proves (or tends) to prove guilt or innocence.

Forensics are the scientific methods utilized to solve a crime. Forensic investigation is the assemblage and analysis of all crime-related physical evidence to conclude a suspect. Investigators examine blood, fluid, or fingerprints, residue, hard drives, computers, or other technology to interpret how a crime took place.

Forensic Science is the applied science that practices techniques and technologies of physics, chemistry, and biology to examine physical evidence left at a crime scene. Chemical tests can be drawn with hair, blood, dirt, cloth fibres, etc. Tools include DNA testing, microscopy, and spectroscopic analysis. A forensic scientist may study:

1. Sound and video recordings which are analysed and enhanced as needed.
2. Examination of crime scenes
3. DNA profiles used in the comparison of biological material and with those of victims and suspects.
4. Physical examination of documents (e.g. handwriting, typewriting) and chemical examination of documents (e.g. inks and papers).
5. Analysis of drugs and poisons in human tissue and body fluids
6. Testing and comparison of evidence such as bullets, casings, fibres, glass, gunshot residues, metals, oils, paints, petrol, plastics, soils, etc.
7. Examine the scenes to establish cause and origin of fires and detonations.
8. Forensic ballistics conduct tests with firearms and ammunition to match a bullet to the gun from which it was discharged.
9. Detect, enhance, record, preserve, and identify the latent information, including fingerprints, footprints, shoe prints, tire marks, and tool marks

Crime Scene Investigation

1. Semantic Scene Classification: In pattern recognition, scene classification is a very common task in which the system scans a picture and analyzes the various elements in it and uses them to categorize the image to a certain category or group. In this process often such a situation may occur when the divisions are not mutually exclusive by definition. AI can identify the character of crime using Semantic Scene Classification.
2. Crime Scene Modelling: Crime Scene Modelling refers to the conception of a multidimensional model of the crime scene by accounting, all the important aspects of the crime scene. Artificial Intelligence can be applied to create a virtual 3D model of the crime scene using the images captured at the crime scene which can enable investigators to understand hidden facets of the crime.
3. Crime Scene Simulation: A simulation is a virtual representation of the behaviour of a system to simulate various outcomes associated with it.AI can develop a simulation of the crime sceneby using the model. AI will thus create potential scenarios of how the crime took place by reviewing factors such as blood splatter patterns, the position of entry and exit wounds etc.

Forensic Investigation

1. Facial Recognition: A facial recognition system is a technology that uses biometrics to identify a person from a photograph or video. It compares the information with a database of known faces to find a match.AI can use facial recognition to identify victims and suspects by comparing the images with existing databases.
2. Fingerprint Identification: Forensic scientists use fingerprints in criminal investigations as a means of identification. Fingerprint identification is one of the most significant criminal investigation tools due to two features: their persistence and their uniqueness. A person’s fingerprints do not change over time. AI can identify fingerprints by comparing them with the victim’s, suspects or known offenders. AI can also provide an analysis of a person’s personality based on the fingerprints.
3. Handwriting Analysis: AI performs handwriting analysis on written statements, letters or suicide notes. When analyzing the entire document using AI, the objective is to identify key handwriting features. These features are uniform throughout the whole document and the system functions by identifying fundamental character similarities. This helps to identify the original writer of the paper.This is accomplished by identifying the unique features of the handwriting in the entire document. Likewise, each has a unique way of shaping certain letters, and this is consistent throughout the whole document. The software can pick significant features of handwriting as well as unique characters. AI can also identify personality traits by analyzing handwriting and determine whether a person is capable of committing a particular crime or not.
4. Information Inspection: AI can examine the enormous amounts of data collected from the crime scene and the data available in the databases effortlessly thus freeing up personnel for other tasks. AI can perform tests on the data to check whether it was tampered with or edited and also aid in retrieval of lost data.
5. Forensic Photograph Analysis: Forensic photography, as well-referred to as crime scene photography, is an activity that records the initial appearance of the crime scene and physical evidence, to provide a permanent record for the courts. AI analysis of photographs reveals important information and connections that could have been overlooked by humans. AI can recognise objects in the image and discover similarities in multiple images. For instance: AI can map a murder weapon to one in the background of a social media post.

Criminal Investigation

1. Criminal Profiling: Criminal Profiling is an investigative strategy used by law enforcement authorities to identify likely suspects. It has been employed by investigators to link cases that may have been committed by the same perpetrator. AI can be used to generate a profile of the culprit by analyzing attributes of the victim, weapon and the method of the crime.
2. Suspect Prioritization: A suspect is someone who is under suspicion for a particular crime. AI can create a list of potential suspects based on the profile. Each of these suspects is further ordered based on means, motive and opportunity.
3. Surveillance of Investigative Interviews and Interrogations: An ‘investigative interview’ is an interview conducted to elicit evidence or information from a person (i.e., witness, victim, complainant or suspect) during the process of an investigation. Investigative interviews conducted by police can vary greatly in purpose, scope and content.Interrogation is, in criminal law, the process of questions asked by an investigator to a person arrested or suspected to seek answers to a crime.AI can review the Investigative Interviews and interrogations to note anomalies in body language that may indicate deceit, agitation or resentment.
4. Network Analysis: Network Analysis is the process of creating a network of links by plotting points on a map or a board. Network Analysis can be used to retrace a person’s steps on a map. AI performed Network Analysis can reveal links between suspects and victims.

Result

Engaging AI in Criminal Investigation will substantially speed up the rate of Crime Solving and quick apprehension of the perpetrator will prevent them from committing more crimes.AI will reduce the caseload of the law enforcement department and create a healthy working environment. Facial recognition has a success rate of only 61.4 percent. On the other hand, fingerprint identification is 98.6 percent accurate. Nevertheless, AI is only as accurate as the information it is provided with in the form of past cases and convictions. Likely, AI will also duplicate the mistakes committed in the past instead of learning from them. This is problematic because machine learning models are only as dependable as the data they’re trained on. If the underlying data is biased in any form, there is a risk that structural inequalities and unfair biases are not just replicated, but also amplified. In this regard, AI engineers must be especially wary of their blind spots and implicit assumptions; it is not just the choice of machine learning techniques that matters, but also all the small decisions about finding, organising and labelling training data for AI models.

Conclusions

The applications of AI in Criminal Investigation are multiplying exponentially with the evolution of new potential. Incorporating AI in Criminal Investigation will revolutionize crime-solving practices. It will accelerate the speed at which justice is served and also reduce the offence rate. The precision of facial recognition system as a biometric technology is lower than iris recognition and fingerprint recognition. The number of false positives is also high. AI can also use fingerprints to interpret the overall personality of an individual. While, AI can also use handwriting for personality assessment, it depicts the person’s state of mind while writing. However, Artificial Intelligence should be employed with utmost care in the Criminal Justice Systems. Courts are using AI to sentence criminals which often leads to a wrongful conviction. AI can be extremely helpful, but it should not be permitted to overtake human judgement. The solutions produced by AI should be reviewed by humans before taking any decisions.