Doğal Dil İşleme

İçerik

Dersin Amacı

This course explores the current design and implementation of computing systems that can process, understand, and generate human language, with an emphasis on the transformative impact of large language models (LLMs) and generative AI (GenAI). Building on a foundation of machine learning and deep learning, the course covers both classical and modern approaches to natural language processing (NLP), tracing the evolution from early syntactic analysis to today's generative, few-shot, and instruction-tuned systems. Topics include part-of-speech tagging, dependency and constituency parsing, semantic representations, and advanced applications such as question answering, summarization, translation, and dialogue systems.
In addition to traditional NLP pipelines, the course will introduce multimodal learning—where text interacts with vision or other modalities—and prompt engineering techniques essential for effectively leveraging LLMs in real-world tasks. Students will gain hands-on experience developing and scaling NLP systems, fine-tuning pre-trained models, and designing prompt-based interfaces for both text-only and multimodal generative models. Through coding assignments and a final project, students will deepen their understanding of how to build, adapt, and evaluate intelligent systems capable of human-like language understanding and generation.
Suggested Background: Data Structures and Algorithms, Linear Algebra, Introduction to Artificial Intelligence-Machine Learning

Dersin İçeriği

Week 1: Introduction to NLP, New Trends, GENAI
Week 2: Parsing, Morphological Analysis, Semantics
Week 3: Language models, tf-idf, Bag of Words (BoW), n-grams: LAB#1
Week 4: Large Language Models, Neural Nets, Embeddings
Week 5: GenAI, Multimodality
Week 6: GenAI & Prompt Engineering-1
Week 7: GenAI & Prompt Engineering-2: LAB#2
Week 8: Speech Analysis
Week 9: Information Extraction, NER, Machine translation
Week 10: Q&A Systems, Chatbots: LAB#3
Week 11: Projects

Dersin Öğrenme Çıktıları

Bu dersi başarı ile tamamlayan öğrenciler;
1. Dil modellerine ve gösterimlerine hakimdir
2. GENAI ve LLM projelerini yürütebilir.
3. DDİ gösterimlerini gerçekleştirebilir.
4. Çok modlu DDİ problemlerini yorumlayabilir.

Kaynaklar

1- Speech and Language Processing, D. Jurafsky& J.H. Martin, https://web.stanford.edu/~jurafsky/slp3/ 3rd edition draft
2- Foundation of Statistical Natural Language Processing, C.D. Manning & H. Schütze, MIT Press, 2003
3- Natural Language Processing with Python, Steven Bird, Ewan Klein, and Edward Loper O’Reilly, 2009: http://www.nltk.org/book/
Supplementary Books:
4- Python 3 Text Processing with NLTK 3 Cookbook, Jacob Perkins, Packt Publishing, 2014
5- Applied Text Analysis with Python, Benjamin Bengfort, Tony Ojeda, Rebecca Bilbro, O’Reilly, 2018
6- Turkish Natural Language Processing, Kemal Oflazer, Murat Saraçlar, Springer, 2018
7- Neural Network Methods for Natural Language Processing, Yoav Goldberg, Morgan & Claypool, 2017

Konu Başlıkları

Değerlendirme Sistemi

Yeterlilik İlişkisi

AKTS - İş Yükü Tablosu