https://pizza.rs/tags/similarity/Recent content in Similarity on INFINI PizzaHugoenhttps://pizza.rs/docs/references/search/specialized/more_like_this/Mon, 01 Jan 0001 00:00:00 +0000https://pizza.rs/docs/references/search/specialized/more_like_this/<h1 id="more-like-this-query"> More like this query <a class="anchor" href="#more-like-this-query">#</a> </h1> <p><code>more_like_this</code> finds documents similar to given text or documents. The engine extracts the input&rsquo;s terms, keeps the significant ones, and runs the resulting top terms as a disjunctive query — a &ldquo;find more of this&rdquo; query without any embedding model.</p> <p><code>fields</code> is required and limited to text fields.</p> <h2 id="how-it-works"> How it works <a class="anchor" href="#how-it-works">#</a> </h2> <p>The query is rewritten at the search entry point — it is a client of the analysis and term pipeline, not an index structure. Per field: gather the <code>like</code> text (free text is analyzed with the first field&rsquo;s analyzer; a referenced document contributes the text of every query field), tokenize it, drop terms below <code>min_term_freq</code>, outside the word-length bounds, or listed in <code>stop_words</code>, and remove every term the <code>unlike</code> inputs produced. The surviving terms from all fields are pooled, ranked by source term frequency (ties broken alphabetically for determinism), and the top <code>max_query_terms</code> become a <code>bool.should</code> of <code>term</code> queries governed by <code>minimum_should_match</code>. If nothing survives, the query matches nothing. When <code>include</code> is <code>false</code> (the default) and the <code>like</code> input named documents, their ids are added as a <code>must_not</code> — the seed documents trivially contain every selected term.</p>