#community-help

Clarifying Fields Inside `text_match_info`

TLDR Matthew asks for clarification on text_match_info fields. Jason explains no documentation exists but offers help. This leads to Kishore Nallan explaining specifics about best_field_weight and best_field_score.

1mo

Join the chat

Oct 18, 2023 (1 month ago)

Photo of md5-ac3436827bebcd62308d83f3823df584

Matthew

04:51 PM

I'm having trouble finding any documentation defining the fields inside text_match_info. Does it exist?

04:55

Matthew

04:55 PM

It would be nice to have detail on how to interpret each field.

04:56

Matthew

04:56 PM

"text_match_info": {
    "best_field_score": "1108091338752",
    "best_field_weight": 15,
    "fields_matched": 1,
    "score": "578730123365187705",
    "tokens_matched": 1
}

Photo of md5-8813087cccc512313602b6d9f9ece19f

Jason

07:38 PM

We haven't documented this yet...

But any particular behavior I can help explain?

Oct 19, 2023 (1 month ago)

Matthew

01:32 PM

What is best_field_weight and how is it reflected in the score?

01:42

Matthew

01:42 PM

Also best_field_score

02:47

Matthew

02:47 PM

It also appears that the text_match_info is the same for each result?

Oct 20, 2023 (1 month ago)

Photo of md5-4e872368b2b2668460205b409e95c2ea

Kishore Nallan

04:15 AM

best_field_weight is the weight of the field with the best matching text when scored against the query tokens. If a user provided set of field weights are not given, the fields in the query are assigned descending order of weights. best_field_score is the actual score of the best matching field.

We use these values to compute an aggregated text match score per record.

> It also appears that the text_match_info is the same for each result?
This can happen if all records being returned exactly contain all the query tokens.

Typesense

Lightning-fast, open source search engine for everyone | Knowledge Base powered by Struct.AI

Indexed 3015 threads (79% resolved)

Join Our Community

Similar Threads

Issues With `text_match` Scoring for Search Queries in Typesense

Colin encountered issues with the `text_match` scoring on Typesense v0.23.1. Jason and Kishore Nallan identified a potential issue with numeric overflow in the text match score and applied an unverified patch. The final resolution is unclear.

17mo

Sorting Search Results Based on Weights

Pantelis wanted to sort search results by weight. Kishore Nallan suggested using "prioritize_exact_match: false" and "text_match_type: 'max_weight'" to achieve the desired result.

5mo

Query on "weighted_score" & Issue with Synonym Highlighting

Stefan asked about "weighted_score" field and reported a possible synonym highlighting issue. Kishore Nallan clarified the use of "weighted_score". The possible synonym issue is still being investigated.

19mo

Phrase Search Relevancy and Weights Fix

Jan reported an issue with phrase search relevancy using Typesense Instantsearch Adapter. The problem occurred when searching phrases with double quotes. The team identified the issue to be related to weights and implemented a fix, improving the search results.

111

8mo

Identifying Best Match Field in Document Search

Jack wanted to display the best match field in a document search. Despite Kishore Nallan's suggestion of looking at field weights, Jack decided to sort highlighted fields by number of matched tokens.

2mo