akshara-project
akshara-project copied to clipboard
CodeforNepal
Sangraha is an initiative to promote the usage of native langauge in computing and literature
Once this is checked in, we may need to do something to make Elastic Search pick up the changes
To improve auto-complete queries, we can add suggestions when indexing documents. https://www.elastic.co/guide/en/elasticsearch/reference/current/search-suggesters-completion.html#indexing Start by tokenizing author/title fields and adding them as suggestions. Eg: we should get a match on "Bhairab...
Currently the ES doc id is randomly generated unique string. We want to make it more user friendly and use title_(author) as the id, which also satisfies the unique constraint.
Kavitakosh has changed their format of how they display title for their content. They now use author name and title of text. The crawling script needs to be fixed to...
https://www.elastic.co/blog/elastic-stack-7-0-0-released We are on elastic 6.3.1 currently which should be fine for our use-case. Investigate if 7 offers some major advantage -- we do want to upgrade to 7 at...
It should be easy for anyone in the team to add new documents to elasticsearch, once we have the raw docs. * schema/format to follow (for all our sources: crawled...
Use query boosting (https://www.elastic.co/guide/en/elasticsearch/guide/current/query-time-boosting.html) to give more weight to title, author then content. For example, in following search the author name exact match should be higher up in search results....
Right now for dev use, sangraha frontend [defaults](https://github.com/Code4Nepal/akshara-project/blob/master/parijat-frontend/src/api.js#L6) to using our prod elasticsearch url. Which is good for getting started on frontend dev easily but there are times when we...
Metadata
Owner
Metadata
Sangraha is an initiative to promote the usage of native langauge in computing and literature