/bookrat

Find similar books based on content

Primary LanguagePythonMIT LicenseMIT

Book Rat

Find similar books based on content. The python application uses postgresql and prolog custom prolog stemmer. We've downloaded the Chitanka's database and analysed the stemmed words with tf-idf algorithm. After that we determine the top 300 words for every book. Then we can recommend books as an answer to a query - the more common words they have in their top words with ours, the higher the score.