Developing a Natural Language Processing System

Tanszéki konzulens: 
A munkatárs fényképe
associate professor
Szoba: IE437
Tel.:
+36 1 463-2899
Email: meszaros (*) mit * bme * hu

A kiírás adatai

A téma státusza: 
Aktív (aktuális, lehet rá jelentkezni)
Kiírás éve: 
2016
A kiírás jellege: 
önálló labor, szakdolgozat/diplomaterv
Research group: 
Intelligent Systems

Project proposal

During the project work students develop natural language processing (NLP) systems for various real-world tasks like text classification (e.g. sentiment analysis, authorship attribution, topic identification etc.) and automated text annotation (e.g. named entity recognition, part-of-speech tagging etc.). Application areas include business and medical information systems and digital humanities. Cooperating with our industrial and academic partners is a possibility for skillful and motivated students.

Customize the proposal according to your ideas, e.g.:

  • information extraction from online sources (Web, Facebook, Twitter etc.)
  • annotate text with knowledge (e.g. Link Open Data sources)
  • identify hate speech or fake news
  • identify unknown authors or classify authors based on stylometry analysis
  • ... you name it

Requirements

What do you need to successfully fulfill your task

  • good knowledge of programming languages, especially Python
  • basic knowledge of natural language processing terms and methods (see for e.g. Chapter 23. and 24. in AIMA)
  • very basic understanding of statistical methods like clustering and machine learning tools
  • motivation

Results

What to expect to learn during your work

  • how NLP tools work in practice
  • how to model natural language texts for various purposes and how to build and use these models
  • what kinds of text attributes can be examined and used
  • how to solve different kinds of NLP-tasks

How to apply

  • collect your ideas (what problems you would like to solve, how you would do it etc.)
  • assess your experience (programming knowledge, previous NLP-related work etc.)
  • write a short introductory letter to the supervisor including the above

 

 
© 2010-2022 BME MIT