Project Work‎ > ‎

Data Mining and Information Retrieval

Trying to decipher data on the web

I took the following courses (among others) at The University of Illinois at Chicago (UIC):

Data Mining and Text Mining and
Query Processing in Database Systems

Projects:

  • In Analysis of News Articles using Metasearch Engines we find sentences that are unique to Conservative and Liberal News Providers. Sometimes, it is interesting to see how a controversial topic (such as 'Stem Cell Research', 'War on a country') receives opposite opinions in different newspapers.

  • Search Engine Evaluation was a term-long project - an attempt to determine which of the top search engines - Google, Yahoo! and Live - would give 'satisfactory' results for a given query. It involved all students in class running queries on the three search engines and evaluating the search engines. I wrote a program that learns the data obtained from the evaluation and guess, for a given query, how the three search engines would perform.

  • MS-GSP Algorithm mines for sequential patterns using multiple minimum supports. It is a level-wise search algorithm, based on the GSP and the Apriori Algorithms. An implementation of this algorithm using C# can be found here.