Skip to content

Mental Factor Classification Using Computational Model of Consciousness

R. Ramteke, S. S. Gharde, D. Bage, A. Dongre

2022 Second International Conference on Next Generation Intelligent Systems July 29, 2022 DOI: 10.1109/icngis54955.2022.10079806 (opens in new tab)

Study at a glance

AI-extracted from the abstract
Characteristics Machine learning classification study Peer reviewed
Sample size 445
Population Human respondents who answered three questions
Measures Accuracy, Precision, Specificity, Recall, F1-Score
Key points A convolutional neural network classified mental factors from a 445-sample dataset with 86.92 percent accuracy on the small database and 93.02 percent on the large database. The authors propose that the Abhidhamma model of 52 mental factors can support the development of Artificial Wisdom by integrating human values and constructive behavior into artificial intelligence.

Abstract

Artificial Wisdom is advancement of Artificial Intelligence where wisdom should be recognized with the intelligence. It means the constructive behavior and values of humanity need to be the part of Artificial intelligence by incorporating wisdom. These can be demonstrates by simulating thought process and hence thinking ability of human beings is recognized as the consciousness. Currently researchers are working on thoughts and consciousness. These thoughts are coexisted with the particular mental factor. Abhidhamma model of ancient Indian literature are claimed 52 mental factors which are categorized in basic three classes such as Ethically Variable Factor, Unwholesome Factor and Beautiful Factor. Proposed model demonstrates the classification of the mental states. Dataset consists of 445 samples collected from various respondents by asking three questions. Preprocessing is performed by using the techniques of Natural language processing and Non-axiomatic logic. Convolutional Neural Network Machine learning technique applied to classify the mental factors. Performance of the proposed system is measured by applying statistical measures such as Accuracy, Precision, Specificity, Recall and F1-Score. Accuracy for small and large database is obtained as 86.92 percent and 93.02 percent respectively.

Citations in the library