3 ms·
There a pyparsing+elasticsearch tutorial on how this was built... Part 1: http://blog.close.io/post/48209152055/sales-data-search-writing-a-query-parser-ast-us
by philfreo 13y ago
There a pyparsing+elasticsearch tutorial on how this was built...
Part 1: http://blog.close.io/post/48209152055/sales-data-search-writing-a-query-parser-ast-using http://blog.close.io/post/48209152055/sales-data-search-writ...
Part 2: http://blog.close.io/post/49711234004/sales-data-search-writing-a-query-parser-ast-using http://blog.close.io/post/49711234004/sales-data-search-writ...
- thom 13y agoAre these the natural language queries the PandoDaily article talks about, or is that a new thing launching today?
- philfreo 13y agoOur search is not yet "natural language" in the technical sense. We've got an extensive number of search "keywords" as described in these tech posts that align fairly well with how you might speak/write a query in natural language, but still rely on a specific syntax rather than pure English phrases. Looking forward to seeing what delver.io is up to :) The full-sentence example questions in the pando article are very straightforward to get an answer to in Close.io, but they can't be written as-is.