STEMM Institute Press
Science, Technology, Engineering, Management and Medicine
Research on an AI Interview Evaluation System Integrating Multi-Agent Systems and Virtual Digital Humans
DOI: https://doi.org/10.62517/jbdc.202501408
Author(s)
Jiayi Wu1, Jiaqi Zhang1, Liuyang Gao1, Jialiang Feng1, Bo Meng1, Yifan Wu1, Mingming Gong2
Affiliation(s)
1SArtificial Intelligence and Software Engineering, Henan University of Technology, Zhengzhou, Henan, China 2iFLYTEK COLTD, Hefei, Anhui, China
Abstract
Current AI interview systems face challenges in achieving natural interaction, conducting multidimensional assessments, and ensuring interpretability. This paper proposes and validates an intelligent evaluation framework integrating multi-agent collaboration with 3D virtual digital humans (VDH). The system enables complex question-answering and multidimensional evaluation through the coordinated operation of four functional agents: resume analysis, interview skills, written test training, and job recommendation. Leveraging a 3D Virtual Digital Human interviewer, the system supports multimodal fusion interaction encompassing voice, text, and facial expressions. It automatically generates interview questions, processes multimodal data, and produces multidimensional scoring alongside interpretable feedback reports. Experiments demonstrate that this research significantly enhances the immersion and naturalness of human-machine interaction while strengthening the objectivity, professionalism, and explainability of evaluations. It provides an innovative solution and theoretical foundation for intelligent talent assessment.
Keywords
Multi-agent; Virtual Digital Human; AI Interview; Multimodal Interaction; Comprehensive Evaluation; Explainability
References
[1]Altınbaş G G,Yorulmaz O .Enhancing Professional Skills: Assessing the Impact of a Computer-Based Training Program on Interview Skills, Self-Efficacy, and Anxiety Levels in Psychology Students. Journal of Rational-Emotive & Cognitive-Behavior Therapy, 2025, 43(4): 51-51. DOI:10.1007/S10942-025-00615-Z. [2]Fosci P, Psaila G .Evolving J-CO-QL+ with fuzzy evaluators for flexible queryisng of JSON data sets. Neurocomputing, 2025, 633129621-129621. DOI:10.1016/J.NEUCOM.2025.129621. [3]Gonçalo A, Filipe M ,Rogério C , et al. JSON Schemas with Semantic Annotations Supporting Data Translation. Applied Sciences, 2021, 11(24): 11978-11978. DOI:10.3390/APP112411978. [4]Kinouchi T, Ogawa A, Wakabayashi Y, et al.Domain adaptation using non-parallel target domain corpus for self-supervised learning-based automatic speech recognition. Speech Communication, 2025, 174103303-103303. DOI:10.1016/J.SPECOM.2025.103303. [5]Tao Y, Liu J, Lu C, et al. CMDF-TTS: Text-to-speech method with limited target speaker corpus. Neural networks: the official journal of the International Neural Network Society, 2025, 188107432. DOI:10.1016/J.NEUNET.2025.107432. [6]Chen C, Liu K, Jing J, et al. Wide-Frequency vibration positioning of asymmetric interferometers based on dual endpoint detection and iterative VMD. Optics and Laser Technology, 2025, 187112882-112882. DOI:10.1016/J.OPTLASTEC.2025.112882. [7]Astrid V, Laura M, Michele B, et al. Automatic detection of charcoal kilns on Very High Resolution images with a computer vision approach in Somalia. International Journal of Applied Earth Observation and Geoinformation, 2023, 125. DOI:10.1016/J.JAG.2023.103524. [8]Technology-Communication Technology; Researchers' Work from Henan Agricultural University Focuses on Communication Technology (Analysis of Socket Communication Technology Based On Machine Learning Algorithms Under Tcp/ip Protocol In Network Virtual Laboratory System). Internet Weekly News, 2019. 1023. [9]Dong-Jo K, Hyun-Ju P. A design and implementation of transmit/receive model to speed up the transmission of large string-data sets in TCP/IP socket communication. Journal of the Korea Institute of Information and Communication Engineering, 2013, 17(4): 885-892. [10]Castillo O, Dinçer H, Yüksel S, et al. Publisher Correction: Assessing carbon–neutral supercapacitors in renewable energy systems with self-improving agent-based molecular fuzzy intelligent algorithms. Scientific Reports, 2025, 15(1): 33712-33712. DOI:10.1038/S41598-025-22245-2. [11]Christoph T ,Ann-Christin B ,Felix M , et al.Multi-modal framework to model wet milling through numerical simulations and artificial intelligence (part 1).Chemical Engineering Journal, 2022, 449. DOI:10.1016/J.CEJ.2022.137794.
Copyright @ 2020-2035 STEMM Institute Press All Rights Reserved