Information Retrieval and Computational Linguistics Systems

Zhiping Zheng
Computertational Linguistics Department
Saarland University
zheng@coli.uni-sb.de
Following is a partial list of online systems I developed. Most of them are related to information retrieval and computational linguistics. AnswerBus Question Answering System
An open-domain QA system. It accepts natural language questions and extracts possible answers from the Web. You can use English, French, German, Spanish, Italian and Portuguese to ask questions.

AnswerBus News Engine
This system indexed over 700,000 news stories from CNN owned Web sites with an embedded search engine.

Seven Tones: Specialized Search Engine in Linguistics and Languages (More about this system)
A search engine only searches information about linguistics and languages.

International News Connection (More about this system)
A real-time news channel that connects you to current world news from up to fourteen news sources. The news is automatically classified and constantly updated.

Content-Based Image Search and Retrieval
These search engines search images by color, shape and texture.

Automatic Sentence Segmentation
An HTML/plain speech text parser. It will parse the content of a Web page into sentences. AnswerBus system uses this tool.

Related Word Finder
The program uses distributional similarity to look for "related words" in TREC AP news corpus (about 240,000 news stories).

Specialized Automatic HTML Document Summarizer
It can summrize web pages related to different topics. The summarization will be more accurate for web documents in a specific domain - linguistics and languages.

On-line Literati Solution - A Fun Anagram
Using this program, You can find solutions for Yahoo!'s online literati games.