Group Of Natural Language Processing Research

Natural Language Processing (NLP) is a subfield of Artificial Intelligence and Linguistics. However, significant developments in the field of informatics have popularized research areas in NLP, making it a standalone discipline today.

Since language ability is a uniquely human characteristic, linguistics holds an important place within Cognitive Sciences. NLP deals with the problems of computer generation and comprehension of natural languages. If a model of language can be built in a computer environment, a highly useful tool for communication is obtained.

The main task of NLP is to design and implement computer systems whose primary goal is to analyze, understand, interpret, and generate a natural language. Research areas of NLP include gaining a better understanding of the structure of natural languages, using natural language as an interface between computers and humans, and performing computer-assisted language translation. All studies in this field have primarily been developed for English, which is a prevalent language in science and business. Due to morphological differences between languages, directly translating the rules and algorithms used for English into Turkish is not quite possible. Therefore, studies to be conducted for Turkish can be carried out by Turkish linguists or domestic/foreign computer engineers who master Turkish exceptionally well.

As application areas of NLP, we can list word processors, querying databases using natural language instead of SQL and converting it into SQL queries by analyzing it to return results to the user, semi-automatic text translation between languages, accessing monolingual or multilingual dictionaries, generating natural language sentences and texts, text summarization, speech recognition, and speech synthesis.

Our Natural Language Research group aims to produce solutions to natural language problems using approaches such as Machine Learning,  Deep Learning, Large Language Models, etc. In addition to applying existing algorithms to current problems, it designs and develops academic and industrial projects by developing new algorithms and approaches.

Our Natural Language Processing Research Group is in its establishment phase and welcomes researchers and students who wish to conduct studies on natural language processing and data science. Ongoing and planned studies will be shared here soon.

*Information regarding completed work and research carried out by the research group is presented below. In addition to currently ongoing master's and doctoral theses, scientific articles, conference papers, and project studies are actively being conducted. They will be announced on the research group page as they are completed.

For information:  metinbilgin@uludag.edu.tr

Group Coordinator

Assoc. Prof. Dr. Metin Bilgin 

Our Team Members

  Assist. Prof. Dr. Ali Duran (Linguistics Specialist)

  Lecturer Dr. Murat Eser

  Dr. Mehmet Bozdemir

  Süleyman Var (M.Sc. Student in Artificial Intelligence)

  Ferhat Kızıl (M.Sc. Student in Computer Engineering) 

  Halil Geleş (M.Sc. Student in Computer Engineering)

  Anar Gurbanov (M.Sc. Student in Computer Engineering) 

  Mustafa Sungur Polater (M.Sc. Student in Computer Engineering) 

  Berre Yeşilyurt (M.Sc. Student in Computer Engineering) 

  Res. Assist. Büşra Çantacıoğlu (M.Sc. Student in Computer Engineering)

  Mert Kentsü (M.Sc. Student in Computer Engineering)  
  Muhammed Salih Aydoğan (M.Sc. Student in Computer Engineering)  

  Yunus Emre Yavuz (M.Sc. Student in Artificial Intelligence)  

  Lecturer Bahadır Özen (Ph.D. Student in Computer Engineering)

  Lecturer Engin Demir (Ph.D. Student in Computer Engineering)

  Res. Assist. Ahsen Battal (Ph.D. Student in Computer Engineering)

Completed Theses

1. Generation of SQL Queries from Turkish Verbal Expressions and Performance Analysis via Retrieval-Augmented Fine-Tuning Method - Mehmet Bozdemir - 2026

2. Detection of Expressions Containing Irony and Sarcasm in Turkish Texts Using Transformer-Based Collective Learning Methods - Murat Eser - 2026

3. Development of a Turkish Vision Language Model with Fine-Tuning Supported by Synthetic and Domain-Specific Data - Master's Thesis - Tufan Yiğit Çiftçioğlu - 2026

4. Expectation-based sentiment analysis with deep learning - Master's Thesis - Melek Turan - 2023

5. Analysis of CPU and GPU Performances of Clustering Algorithms - Master's Thesis - Melek Güler - 2022

6. Generating Python syntax from Turkish verbal expressions via machine translation - Master's Thesis - Mehmet Bozdemir - 2022

7. Investigation of the effect of grading training examples in deep learning algorithms - Master's Thesis - Kaan Karaköse - 2022

8. Performance analysis of contextual vectors in visual question answering problem - Master's Thesis - Özlem Hakdağlı - 2022

9. Compatibility-themed solution of the vehicle routing problem on a heterogeneous fleet with a hybrid approach - Master's Thesis - Nisanur Bulut - 2020

10. Deep learning-based transparent object recognition - Master's Thesis - Korhan Mutludoğan - 2020

Publications

• Schema Retrieval with Embeddings and Vector Stores Using Retrieval-Augmented Generation and LLM-Based SQL Query Generation, Mehmet Bozdemir, Metin Bilgin, Appl. Sci. 2026, 16(2), 586

• DPATransLLM: Detection of Pronominal Anaphora in Turkish Sentences Using Transformer-based, Large Language Models and Hybrid Ensemble Approach,M Bilgin, E Demir,APPLIED SCIENCES 15 (23), 1-17

• Irony and Sarcasm Detection in Turkish Texts: A Comparative Study of Transformer-Based Models and Ensemble Learning Score, M Eser, M., BilginAPPLIED SCIENCES 15 (23), 1-25

• Gated transformer network based EEG emotion recognition, M Bilgin, A Mert, Signal, Image and Video Processing, 2024.

• Research on modeling the thixotropic properties of cementitious systems using regression methods in machine learning, M Bilgin, Ö Biricik, M Eser, A Mardani, HG Şahin, Construction and Building Materials 411, 1-14, 2024.

• Aspect-Based Sentiment Analysis in Turkish Texts with Pre-Trained Models, M Bilgin, M Turan, THE SEVENTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL MATHEMATICS AND ENGINEERING SCIENCES, 37-42, 2023.

• Sentiment Analysis from Turkish News Texts with BERT-Based Language Models and Machine Learning Algorithms, M Bilgin, E Demir, 8th International Conference on Computer Science and Engineering, 147-150, 2023.

• Text Classification With Zero Shot Learning in Turkish Text, M Bilgin, M Bozdemir, 5th International Conference on Natural and Applied Science and Engineering, 84, 2023.

• A Novel Syntactic-Based Approach to Calculate Similarities Among Languages, M Bilgin, Süleyman Demirel University Journal of Natural and Applied Sciences, 27(1), 125-136.

• Research on Behavior of Two New Random Entity Mobility Models in 3-D Space, M Bilgin, M Eser, Arabian Journal for Science and Engineering 47 (2), 1159-1171, 2022.

• Compatibility Themed Solution of the Vehicle Routing Problem on the Heterogeneous Fleet, M Bilgin, N Bulut, International Arab Journal of Information Technology 19 (5), 774-784, 2022. 

• Detecting transparency of glasses with capsule networks based on deep learning, M Bilgin, K Mutludoğan, 6th International Conference on Computer Science and Engineering (UBMK), 2021.

• A New Approach to Automatically Find and Fix Erroneous Labels in Dependency Parsing Treebanks, M Bilgin, International Arab Journal of Information Technology 18 (3), 356-364, 2021. 

• Route Determination Method, M Bilgin, TR Patent 2019/14,532, 2021. 

• A new statistics-based approach to improve Word2Vec's sentiment classification success, M Bilgin, Selcuk University Journal of Engineering Sciences 20 (3), 63-72, 2021.

• A New Approach Based on Simulation of Annealing to Solution of Heterogeneous Fleet Vehicle Routing Problem, M Bilgin, N Bulut, International Conference on Artificial Intelligence and Applied Mathematics in Engineering, 2021.

• A new strategy for curriculum learning using model distillation, K Karaköse, M Bilgin, Global Journal of Computer Sciences: Theory and Research 10 (2), 57-65, 2020.

• Classsification of Turkish Tweets by Document Vectors and Investigation of the effects of parameter changes on classification success, M Bilgin, Sigma Journal of Engineering and Natural Sciences 38 (3), 1581-1592, 2020.

• A Study on Message Delay Time in Mobile Crowd Sensing, M Bilgin, Dokuz Eylül University Faculty of Engineering Journal of Science and Engineering, 22 (65), 393-400, 2020.

• Novel random models of entity mobility models and performance analysis of random entity mobility models, M Bilgin, TURKISH JOURNAL OF ELECTRICAL ENGINEERING & COMPUTER SCIENCES 28 (2), 708-726, 2020.

• Sentiment analysis of tweets based on document vectors using supervised learning and semi-supervised learning, M Bilgin, İF Şentürk, Journal of Balıkesir University Institute of Science and Technology 21 (2), 822-839, 2020.

• American sign language character recognition with capsule networks, M Bilgin, K Mutludoğan, 3rd International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT), 2019.

• Sentiment analysis of tweets based on document vectors using supervised learning and semi-supervised learning, M Bilgin, İF Şenturk, Journal of Balıkesir University Institute of Science and Technology, 21(2), 822-839, 2019.

• Sentiment Analysis with Term Weighting and Word Vectors, M Bilgin, H Köktaş, International Arab Journal of Information Technology 16 (5), 953-959, 2019.

• A Study on Named Entity Recognition with OpenNLP at English Texts, M Bilgin, Journal of Applied Intelligent System 4 (1), 1-8, 2019.

• Comparison of Model Construction Times of Word Vector Methods, M BİLGİN, Journal of Information Technologies 12 (2), 141-146, 2019.

• A New Method Proposal to Improve Classification Success of Turkish Texts, M BİLGİN, Uludağ University Journal of The Faculty of Engineering 24 (1), 125-136, 2019.

• Object Oriented Programming with C#, M Bilgin, M Eser, Kodlab, 2018.

• Machine Learning Theory and Algorithms, M Bilgin, Papatya Bilim, 2018.

• Identification of Vehicle Design and Transition of Traffic Signs with Image Processing Method, M BILGIN, Z ZEYBEK, International Journal of Electronics, Mechanical and Mechatronics Engineering (IJEMME), 8(2), 1555-1569, 2018.

• Sentiment analysis on Twitter data with semi-supervised Doc2Vec, M Bilgin, İF Şentürk, 2017 international conference on computer science and engineering (UBMK), 661-666, 2017.

• Performance Analysis of Classical Machine Learning Methods on Real Datasets, M Bilgin, Academic Informatics Conference, 2017.

• Word2Vec Based Sentiment Analysis for Turkish Texts, M Bilgin, H Köktaş, International Conference on Engineering Technologies, Konya, Turkey, 106-109, 2017.

• Dependency parsing with stacked conditional random fields for Turkish, M Bilgin, M Amasyalı, Journal of the Faculty of Engineering and Architecture of Gazi University 32 (2), 2017.

• Semantic role labeling with relative clauses, M BİLGİN, MF AMASYALI, International Journal of Electronics Mechanical and Mechatronics Engineering, 6(2), 1165-1175, 2016.

• Comparison of Machine Learning Methods for the Sequence Labelling Applications, MF AMASYALI, M BİLGİN, 23nd Signal Processing and Communications Applications Conference, 503-506, 2015.

• Automatic Labeling of Subordinate Clauses in Compound Sentences, M BİLGİN, MF AMASYALI, Academic Informatics, 2015.

Our Projects

• Turkish Personalized Learning Assistant Developed with Retrieval-Augmented Generation and Multi-Agent Architecture: Bilgebot, M. Bilgin (Principal Investigator), TÜBİTAK 1001, 2026-2028

• CODEMASTER GPT - Edge Computing-Based Generative AI-Supported Decision Support Assistant for Detection and Resolution of Production Anomalies in Industrial Facilities, M. Bilgin (Principal Investigator), TÜBİTAK 1507, 2026-2027

• Large Language Model-Based Enterprise Document Consultant, M. Bilgin (Principal Investigator), TÜBİTAK 1505, 2025-2026

• SMARTBPM: AI-Powered Automated Process Management, M. Bilgin (Consultant), TÜBİTAK 1501, 2025-2026

• Development of a Transformer-Based Segmentation Model Analyzing User Behaviors in Sales and Marketing, M. Bilgin (Consultant),  TÜBİTAK 1501, 2025-2026

• AI-Powered Recruitment and Candidate Evaluation System, M. Bilgin (Consultant), TÜBİTAK 1507, 2025-2026

• AI-Powered Dynamic Reporting Tool ManageSQL, M. Bilgin (Consultant), TÜBİTAK 1501, 2025-2026

• Large Language Model-Based Smart Assistant for Metal Cutting Machines, M. Bilgin (Consultant), TÜBİTAK 1711, 2025-2026

• Converting Natural Language Expressions into Structured Query Language (SQL), M. Bilgin (Consultant), M. Bozdemir, BAP, 2025-Ongoing

• A Multilingual Repository of Phraseme Constructions in Central and Eastern European Languages (PhraConRep), M.Bilgin (Working Group Member)  EU COST Project, 2023 - 2027

• Randomised Optimisation Algorithms Research Network (ROAR-NET), M.Bilgin (Working Group Member)  EU COST Project, 2023 - 2027

• Universality, diversity and idiosyncrasy in language technology (UniDive), M. Bilgin (Working Group Member)  EU COST Project, 2022 - 2026

• Smart Parking Management System for EV Charging Stations, M.Bilgin (Consultant), Technopark Project, 2024-2025

• Veriket: Classification and Tagging of Data, M. Bilgin (Consultant), TÜBİTAK 1507, 2024-2025

• Development of Domain-Specific Turkish Large Language Model and Offline Information Query Infrastructure, M. Bilgin (Consultant), TÜBİTAK 1507, 2024-2025

• AI-Powered Occupation and Skill Classification Project, M. Bilgin (Consultant), TÜBİTAK 1007, 2024-2025

• Development of User Experience-Oriented Voice Command System Compliant with Cybersecurity Requirements for Control of Superstructure Components in Buses, M. Bilgin (Consultant), TÜBİTAK 1501, 2024-2025

• Development of Deep Learning-Based Named Entity Recognition (NER) Tool, M Bilgin (Consultant), TÜBİTAK 2209

• AI-Based Turkish Text Normalization Tool, M Bilgin (Consultant), TÜBİTAK 2209

• Deep Learning-Based Turkish Text Summarization with Supporting Visual Aid, M Bilgin (Consultant), TÜBİTAK 2209

• Automatic generation of syntactic expressions in python language from Turkish verbal expressions, M. Bilgin (Consultant), M. Bozdemir, BAP, 2022

• Wearable Robots for Augmentation, Assistance or Substitution of Human Motor Functions, M. Bilgin (Management Committee Member)  EU COST Project, 2018- 2022

• Design and Implementation of Natural Language and Image Processing Based Assistant for Hearing and Speech Impaired Individuals, M.Bilgin (Consultant), Technopark Project, 2020-2021

• Motion Management and Control for Mobile Data Carriers, M Bilgin (Researcher), TÜBİTAK 3501, 2018-2020.

Developed Tools

• To Be Published Here Soon...

Former Members

• Kaan Karaköse

• Korhan Mutludoğan 

• Nisanur Bulut 

• Özlem Hakdağlı 

• Melek Güler 

• Melek Turan 

• Tufan Yiğit Çiftçioğlu