BioMedical IR System

• Building Inverted Positional Index and Answering Specialized Wildcard Queries

• Parsing HTML files and preparing a structured corpus using the extracted text

• Preparing Inverted Index on the tokens in the corpus and answering WildCard queries

• Ranked Retrieval for Free Text Queries

• Construction of tf-idf vectors, Champion Lists

• Computing the top 10 results for each query based on tf-idf scores, champion list scores and cluster pruning scheme.

• Text Classification of text documents

• Implementation & Classification of text docs using the following classifiers

  • Naive Bayes (Multinomial & Bernoulli)
  • Rocchio Classifier