Work

Supplementary data for a systematic review of multimodal versus unimodal affect recognition

Year: 2026

Type: supplementary-materials

Abstract: This dataset accompanies a systematic review comparing multimodal and unimodal approaches for recognising emotional and psychological distress from audio, video and text. The review followed PRISMA 2020 guidelines and included 30 studies identified through Scopus, IEEE Xplore, Web of Science and PubMed. The workbook contains the full search strategy for all four databases, the matched comparisons between multimodal and unimodal systems, the extracted model results, and the complete PROBAST+AI risk-of-bias and applicability assessment for every study. A README sheet explains each sheet and gives the exact filters needed to reproduce the main results. Three studies were assessed but not included in the quantitative analysis, and they are flagged in the data. The included papers are identified by DOI. The data are shared under a CC BY 4.0 licence.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.23997697

Authors: Vinay, Varenya, Wangchuk, Tandin, Gonsalves, Tad, S, Ruban

Institutions: St. Aloysius (Deemed to Be University), College of Science and Technology, Royal University of Bhutan, Sophia University

Language: English

Topic: Emotion and Mood Recognition

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

SDG: Good health and well-being

Open Access status: green

OpenAlex ID: W7225409552

Cites:

Cited by:

Related to: 10