Natural language processing (NLP) focuses on machine learning techniques applied
to natural languages in order to build automated systems that can analyse, understand,
and generate natural human languages.
Hacettepe University Natural Language Processing Research Group (HUNLP) is concerned
with the natural language processing (nlp) applications and various fields in computational
linguistics. We focus on mainly the following areas:
- Morphological segmentation: We are interested in morphological segmentation of agglutinating
languages as a learning problem, which involves statistical learning as an unsupervised learning problem.
- Syntax: We are interested in syntax, that involves learning parts-of-speech in a given
text as another learning problem in computational linguistics.
- Example Based Machine Translation: We are interested in example-based machine translation
systems between English and Turkish. We have developed a system which learns translation templates
from bilingual translation examples using machine learning techniques.
- Text Summarisation / Keyphrase extraction: We do research in text summarisation and keyphrase
extraction problems. We have developed text summarisation and keyphrase extraction systems which
extract summaries and keyphrases of English and Turkish documents.
- Information Extraction: We have developed a Name Entity Recognition system.
- Sentiment Analysis: Sentiment analysis on social media documents.
- Question Answering Systems: We have worked on factoid question answering systems.
- Word sense discovery: Graph based word sense discovery from Collaboratively built
Semi-structured Information
- Topic segmentation
- Morphological disambiguation
We would be pleased to have new members in our research group, who are interested
in natural language processing. Please feel free to contact us, if you think that your
interests match with ours.