This projects aims to: (Given a Google Group, or maybe any other corpus of text, and an email account from which to access to the mails sent by the members) 1. Fetch the emails to get all the documents generated by the group 2. Fetch the author (e-mail) of each e-mail. 3. Clean the docum...

We have implemented the state of the art query expansion techniques combined with a new family of ranking functions which can take into account some semantic structure in the query. This code has been implemented by José Ramón Pérez Agüera and Hugo Zaragoza This implementation is only for researc...

This project contains an implementation of Explicit Semantic Analysis as introduced by Evgeniy Gabrilovich and Shaul Markovitch: * *Evgeniy Gabrilovich and Shaul Markovitch*. _Computing Semantic Relatedness using Wikipedia-based Explicit Semantic Analysis._ In Proceedings of IJCAI, 2007. The...

In this project, we have built several Ajax-based JITIR systems to help report writers. We have experimented with several different ways in which words and phrases from the users document are used to form search engine queries. The idea of Just-in-Time Information Retrieval (JITIR) is to automati...

. Since there are a great number of topics that liberal and conservative parties express various opinions on, and in different topics, the words used by those authors will be wildly divergent. So we want to focus on a specific topic that liberal and conservative parties hold different views about it...

We will construct a directed graph of all the IDs extracted from BBS site. Something interesting may be obtained by examining the structure of the graph, I suppose.If SCC fails, maximal clique will work at least. Also we could recommend new posts to a given user according to the behavior data we g...

Find tabs that are related to each other. Provide multiple indexes used to find related songs based on many different features of a song.

This is my thesis project for creating an office application plugin for using search repositories in order to create documents.

retrieve and crawl valuable information regarding to patent