Design and development of an android chatbot application for detection of symptoms of anxiety and depression in a student population

Read full text Summarise with VIRAT
Authors: Σπυρίδων Η. Μπαλκάμος
Cited by: 0
References: 43

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This thesis presents the design and implementation of a prototype mobile health (mHealth) system: a conversational agent (chatbot) for Android devices that detects indicators of anxiety and depression in university students.Increasing academic pressures have created a need for digital screening tools that are accessible and culturally appropriate.However, developing such systems faces two main obstacles: limited clinical data availability due to GDPR restrictions, and poor support for Modern Greek in current Natural Language Processing (NLP) models.To address these challenges, this work proposes a synthetic data generation methodology based on statistical distributions (Gaussian Personas).This approach combines psychometric assessment instruments (SCL-90, BDI-II) with the text generation capabilities of Large Language Models (LLMs), enabling the creation of training data without using real personal information.The system uses a hybrid inference architecture with two components: a BERT model fine-tuned on Greek text for sentiment analysis, and a deterministic layer using Regular Expressions (Regex) to identify critical indicators.The thesis presents a functional prototype that serves as a technical proof-of-concept.Future validation with real-world field data is required.

Fields of Study & Subjects

Digital Mental Health Interventions Mental Health via Writing Emotion and Mood Recognition

Publication details

Journal / source
University of Thessaly Institutional Repository (University of Thessaly)
Publication date
1 January 2026
Language
EL
Document type
Thesis