The Evolution of 'More Like This'

# Summary "More Like This" (MLT) is a search function that finds documents similar to an already-selected one, traditionally using lexical matching of important words through full-text search mechanisms like TF-IDF and BM25. Modern implementations increasingly use embeddings—numerical vector representations of documents—which enable better matching of synonyms, paraphrases, and conceptually similar content, though lexical approaches remain superior for exact matches like error codes and product identifiers. The article outlines how MLT has evolved from simple word-matching to vector-based similarity search while exploring the strengths and limitations of each approach.