Researchers from Tokyo University of Science have conducted a study to explore the use of Machine Learning in detecting Deception. Machine learning is a subset of artificial intelligence that teaches computers to perform tasks by learning from data and patterns, rather than relying on pre-defined rules.

Detecting deception is crucial in various situations, such as interviewing crime victims or suspects and assessing patients with mental health issues. Human interviewers may struggle to ask the right questions or accurately identify deception.

To address this challenge, the researchers aimed to develop an automated Deception Detection system using machine learning. The goal was to create a fair and accurate system that assists interviewees in telling the truth while identifying genuine suspects without wrongly accusing innocent individuals.

In their study, the researchers collected data from four male graduate students using a natural approach. The subjects were shown random images and asked to freely talk about them while making deceptive statements. Facial expressions and pulse rates were captured using a web camera and smartwatch, respectively.

The team employed a machine learning technique called Random Forest (RF) to build the deception detection model. They utilized the collected data, including facial expressions and pulse rates, to train the model.

The model’s performance was assessed using a 10-fold cross-validation technique. The results showed promising performance, with accuracy and F1 score ranging from 75% to 80%, and the highest accuracy reaching around 87%. Key cues for deception detection included changes in pulse rate, gaze movements, and specific facial areas around the eyes and mouth.

While the researchers believe their approach could be a valuable tool in detecting deception, they acknowledge the limitations of their study due to a small sample size and limited resources. They emphasize the need for a larger and more diverse dataset to develop a more robust deception detection system.

This study lays the groundwork for future research on machine learning-based deception detection and highlights the potential of facial expressions and pulse rates as cues for identifying deception in human interactions.

Tokyo University of Science: https://www.tus.ac.jp/en/

The post Machine Learning Helps Detect Deception, Tokyo University Study Finds appeared first on TS2 SPACE.