# More Like This and shingles / phrases **URL:** \ **Category:** Elasticsearch\ **Created:** [September 16, 2017, 8:49pm UTC](https://discuss.elastic.co/t/more-like-this-and-shingles-phrases/100775 "2017-09-16T20:49:20Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![james.daily](https://avatars.discourse-cdn.com/v4/letter/j/5daacb/32.png) [@james.daily](https://discuss.elastic.co/u/james.daily)\ **Post date:** [September 16, 2017, 8:49pm UTC](https://discuss.elastic.co/t/more-like-this-and-shingles-phrases/100775/1 "2017-09-16T20:49:20Z")
Is it possible to combine shingles with a More Like This query? I added a custom shingles analyzer to my index, but MLT queries do not seem to use the shingles field. Being able to combine these is _extremely_ important to our use case (comparing large numbers of large text documents). The default Bag of Words approach leads to too many false positives from out of context words. In particular, setting the `per_field_analyzer` to `my_shingles_analyzer` or `custom` does not seem to work. ---
**Author:** ![james.daily](https://avatars.discourse-cdn.com/v4/letter/j/5daacb/32.png) [@james.daily](https://discuss.elastic.co/u/james.daily)\ **Post date:** [September 17, 2017, 3:34am UTC](https://discuss.elastic.co/t/more-like-this-and-shingles-phrases/100775/2 "2017-09-17T03:34:01Z")
I got it to work by combining multiple More Like This queries each with their own `analyzer` instead of trying to use `per_field_analyzer`. That worked out better anyway, allowing me to have separate settings (e.g. `stop_words`, `min_word_length`) for unigrams vs bigrams. ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [October 15, 2017, 3:34am UTC](https://discuss.elastic.co/t/more-like-this-and-shingles-phrases/100775/3 "2017-10-15T03:34:46Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.