UNIVERSITY OF HERTFORDSHIRE
		COMPUTER SCIENCE RESEARCH COLLOQUIUM

			     presents


       "Predicting Big Five Personality Traits From Voice
	      Signals via Ensemble Learning Algorithms"

                   Dr. Mohammad-H Tayarani-N
              (University of Hertfordshire)


                11 November 2026 (Wednesday)
			            F309
		   Hatfield, College Lane Campus
			      13:00 - 14:00


		   Everyone is Welcome to Attend

            Refreshments will be available



Abstract:

This article studies how fillers (short utterances like “ehm” and “uhm”), laughter and words uttered by people can be used to predict the Big-Five personality traits. Fillers, laughter and words were extracted from spontaneous conversations conducted by 120 subjects. Various feature extraction methods, including Fourier, Wavelet, Walsh, Hilbert–Huang transforms, acoustic features, and convolutional neural networks, are used to train the inference model. This article uses a wrapper technique for feature selection and analyses them to identify features that are more representative of personality traits. An ensemble learning of a set of base learners is presented, each trained to be specialized for a specific feature type. Experiments are conducted to compare the proposed ensemble learning algorithm with a range of established machine learning and ensemble learning algorithms. The results demonstrate that the proposed method consistently achieves superior classification performance under the adopted experimental framework.

Published in: IEEE Transactions on Computational Social Systems (Early Access)



---------------------------------------------------
Hertfordshire Computer Science Research Colloquium
http://cs-colloq.cs.herts.ac.uk