Elasticsearch-rails results accuracy not as expected

Question

I am using elasticsearch-rails and elasticsearch-model gem for searching words in my rails app.

Here is my model where I want to search to take place

require 'elasticsearch/model'

    class Article < ActiveRecord::Base
      include Elasticsearch::Model
      include Elasticsearch::Model::Callbacks

      settings index: { number_of_shards: 1 } do
        mappings dynamic: 'false' do
          indexes :title, analyzer: 'english', index_options: 'offsets'
          indexes :text, analyzer: 'english'
        end
      end

      def self.search(query)
        __elasticsearch__.search(
          {
            query: {
              multi_match: {
                query: query,
                fields: ['title^10', 'text']
              }
            },
            highlight: {
              pre_tags: ['<em>'],
              post_tags: ['</em>'],
              fields: {
                title: {},
                text: {}
              }
            }
          }
        )
      end
    end

    # Delete the previous articles index in Elasticsearch
    Article.__elasticsearch__.client.indices.delete index: Article.index_name rescue nil

    # Create the new index with the new mapping
    Article.__elasticsearch__.client.indices.create \
      index: Article.index_name,
      body: { settings: Article.settings.to_hash, mappings: Article.mappings.to_hash }

    # Index all article records from the DB to Elasticsearch
    Article.import

    #Article.import force: true

My questions are how do I do search word? tshirt, T-shirt, Tshirts should all match. Any links for further research is also helpful. Thank you

Answer 1

The keyword you are looking for is fuzziness.

See this blog post for example: http://dev.mikamai.com/post/85901585484/elasticsearch-on-rails-a-primer

records = Article.search(query: {match: {_all: {query: 'wold', fuzziness: 2}}}).records
records.first.title #  => "Hello World"

fuzziness represents the maximum allowed Levenshtein distance, it accepts an integer between 0 and 2 (where 2 means the fuzziest search possible) or the string “AUTO” which will generate an edit distance based on the charachers length of the terms in the query.