Using Deep Learning and Transformer Models to Identify Inconsistencies between Interview Responses and Resume Claims

Haque, Jareen Sayma (2025) Using Deep Learning and Transformer Models to Identify Inconsistencies between Interview Responses and Resume Claims. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (2MB) | Preview

Abstract

In modern recruitment, a large ”trust gap” exists between claims made on resumes and statements given in interviews, an issue cannot be addressed with mere human checks or manual verification alone. The aim of this work is to fill the gap by developing, applying, and validating a novel deep learning framework to automatically detect semantic inconsistencies between these two sources. The core objective was to determine the extent to which transformer-based models could accurately perform this task, comparing three distinct architectures which are BiEncoder, Cross-Encoder, and a T5 based Natural Language Inference (NLI) model on a custom-built dataset of real-world humane and synthetically generated resume-interview pairs. The findings indicates that utilizing both contexts simultaneously is far superior, with the fine-tuned Cross-Encoder model significantly outperformed the other architectures, achieving 73.1% accuracy and a balanced F1-score of 72.7%. This paper establishes the first empirical benchmark for the novel work of resume to-interview consistency detection presenting a proven blueprint for a new class of smart HR tools designed to support evidence-based hiring. This work contributes to fairer, transparent recruitment practices by developing scalable verification mechanism to aid human judgment rather than replacing it.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Yaqoob, Abid UNSPECIFIED
Uncontrolled Keywords:	Deception Detection; Transformer Models; Sentiment Analysis; Recruitment Efficiency; NLI; Transformers; Bias Mitigation; HR
Subjects:	P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing H Social Sciences > HD Industries. Land use. Labor > HD28 Management. Industrial Management > Human Resource Management Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning H Social Sciences > HD Industries. Land use. Labor > HD28 Management. Industrial Management > Human Resource Management > Recruitment
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Ciara O'Brien
Date Deposited:	01 Jul 2026 10:54
Last Modified:	01 Jul 2026 10:54
URI:	https://norma.ncirl.ie/id/eprint/9429

Actions (login required)

View Item