My research interests lie in the fields of Bioinformatics, Natural Language Processing (NLP), Text Mining, and Machine Learning (ML).
On-going Projects
Text mining for gene-disease relationship types in biomedical literature
Collaboration with University of Cambridge - DTAL
This project explores automated methods for extracting and classifying gene-disease relationships (GDR) from biomedical literature. We utilise natural language processing (NLP) techniques, including transformer-based models, to improve the identification of different relationship types. This work contributes to more accurate and granular biomedical knowledge extraction, supporting research into disease mechanisms and therapeutic targets.
Methodology: Transformer-based NLP models for Named Entity Recognition (NER) and relation extraction
Applications: Construction of biomedical knowledge graphs, precision medicine
2. Clustering-based negative sampling for pathogen-host protein interaction
Presented at the CIBB conference, Benevento, Italy (2024)
Negative sampling is one of the major challenges in training machine learning models for protein-protein interactions (PPI). In this project, a clustering-based approach will be introduced to generate more biologically meaningful negative samples and thus improve the training of models for predicting pathogen-host interactions.
Methodology: Clustering techniques combined with machine learning for robust negative samples
Applications: Infectious disease research, vaccine discovery and drug targeting
New Projects
Assessing earthquake damage and determining the need for aid via social media channels
Submitted - Tubitak 1001
This project will utilize NLP and social media data mining to classify and assess earthquake-related messages, determining the extent of damage and identifying aid requirements. It includes both sequence classification models and entity recognition for structured information extraction.
Methodology: NLP models for damage assessment and named entity recognition for aid requests
Applications: Disaster response optimization, crisis informatics