Keyword search example for small dictionaries. Accepts data from external storage and converts to in-memory matrices using for cosine similarity calculation. Needs a library to create text embeddings. Contains several methods:
- Similarity between vocabulary-based embeddings. Beware of typos ❗
- Fuzzy search with a levenshtein distance it it;
- n-grams comparation, where n = [2, max_dictionary_sentence].