A search engine written in C for historical documents of the Amazon region, stored locally.
StackC
Why I built it
This was an information retrieval course project. It builds the parts of a search engine by hand: a vocabulary with a hash table, an inverted list, and a ranking of documents by how similar they are to the query.
Goals
- Build a vocabulary of all terms in the documents
- Build an inverted list that maps each term to its documents
- Weight terms with TF and IDF
- Rank documents by similarity to the query
- Run on a local collection of 35 historical texts
Screenshots
