Unsupervised Information Extraction and Retrieval

Clustering

Clustering is classification of objects into different groups. For the classification the distances between objects and the all characteristics of them are used.

The distances more used are:

• Euclidean, the more usual.

• Manhattan.

• Hamming.

Objects are grouped using distance calculation, the objects nearer of each other belong the same group.

Clustering is often used in information retrieval and information extraction for making groups of near concepts, so the result of any process performed over the group should be similar to the result obtained from one member. Existing commercial applications are:

• iResearch Report. A system of information extraction and document summing-up.

• News Feed Researcher. An extension for the previous systems for news and feeds.

This site is developed for the assignement Information and Retrieval Organization of the subject Information Retrieval and Access over Computer Engineering of Carlos III University of Madrid.

Available versions:

Download the DOC article

Download the PDF article