THE STUDY AIMS TO PREDICT THE HUMAN THOUGHTS USING ARTIFICIAL INTELLIGENCE ALGORITHMS

DOI: https://doie.org/10.0618/Jbse.2024298087

K.Jamberi, R.Muthumeenakshi

Keywords:

Artificial Intelligence, Convolutional neural networks, Recurrent neural networks, Computer brain.

Abstract:

Guessing what people think and feel with artificial intelligence (AI) programs represents a leap forward for both neuroscience and AI. In this work our method explores the application of advanced machine learning techniques known as convolutional brain networks (CNNs) and recurrent neural networks (RNNs) for decoding brain signals and predicting cognitive states. This novel technique prepares and processes fMRI and electroencephalogram (EEG) data to detect signature patterns of brain activity that correspond to given ideas and states of mind. The proposed method explicitly guides our CNN-RNN model to localize these patterns, lever- aging the spatial feature extraction of CNNs and the temporal sequence learning of RNNs. The technique evaluates the performance of the model based on metrics like accuracy, precision, recall and F1-score and shows how well it can predict human thinking. These findings improve our understanding of brain-computer interfaces (BCIs) and pave the way for applications of neuro prosthetics, mental health diagnostics, and human-computer interaction. Further work will focus on enhancing the accuracy of the model, increasing the range of cognitive states that can be reliably identified, and matters of ethical issues around mind-reading technology.

PDF