practicum-1

Practicum 1

Learning objectives

• Ranking documents using the Vector Space Model
• Building an inverted index
• Term-at-a-time and document-at-a-time scoring

Exercises

1. Term weighting and vector space retrieval: Score a toy-sized document collection against a query using the vector space model (i.e., TFIDF term weighting and cosine similarity).

2. Building an inverted index: Build an inverted index from a collection of 1000 documents, with the term frequencies stored.

3. Term-at-a-time scoring: Implement term-at-a-time scoring using vector space retrieval with TFIDF term weighting.

4. Document-at-a-time scoring: Implement document-at-a-time scoring using vector space retrieval with TFIDF term weighting.