Thanapon Noraset

Computer science graduate student at Northwestern University.

I am a computer science PhD student in Department of Electrical Engineering and Computer Science at Northwestern University. My advisor is Prof. Doug Downey, and our lab is WebSAIL. My research interests include natural language processing and machine learning. I have been working on information extraction in Wikipedia and scholar articles.

Currently I am focusing on statistical language modelling. I am exploring how data from dictionaries and encyclopedias can be used to effectively improve natural language capabiliy of language models.


Please refer to my profile on Google Scholar.


Dictionary Models (active)

This project aims to build generative models of definitions in dictionaries and encyclopedias. The project is part of my current research. I am exploring deep learning algorithms and other language models.

Key Phrase Extraction

We build a system that can extract key phrases from scholar articles. We collaborate with Allen Institute for Artificial Intelligence to extract and deploy the key phrases in (check it out).

WebSAIL Wikifier

WebSAIL Wikifier is an Entity Linking system that identifies and links phrases to a Wikipedia page. We participated in TAC 2013 and ERD 2014. We also attempted to "wikify" English Wikipedia to increase stuctured information on Wikipedia, check out 3W for information and resources.

WebSAIL Wikifier: TabEL

My colleague, Chandra, has extened WebSAIL Wikifier to extract enities from tables on websites. The algorithm here is considered as an improved version of the original project, though I have not yet applied on normal articles. For more information and resources, see TabEL project page.


TextJoiner is a system that allows users to interactively query and perform joins on facts expressed in Wikipedia using text patterns like "cities such as $x". Just language models and word embedding.

The backend might be unavailable. Other than discussion, I mainly implemented UI in this project.


This project extracted Wikipedia tables and used machine learning to enable table exploration via "search" and "join". WikiTables allows user to find and view columns from different table side by side, and potentially discover an interesting corelation. Demo: Join and Search.

Again, I mainly implemented UI in this project.

Not useful (for curious cats)

There are lot of boxes in this page. By the way, this website uses Bootstrap.
I am from Faculty of Information and Communication Technology at Mahidol University, Thailand. Currently on educational leave.
Right now, I am using Torch for neural network framework.
Sometimes my code contains Solaire of Astora (Sunbro) which indicates critical code that magically works.

I do not know much about art and design, but my motto to go about building UI is "even if it does not work, it still looks nice".
My photo was taken by Jidapa Kraisangka.