search

TheRensselaerIDEA / twitter-nlp

Data Analytics on Twitter with Natural Language Processing
MIT License
17 stars 7 forks source link

[Development] Implement text similarity search in Elasticsearch #7

Closed AbrahamSanders closed 4 years ago

AbrahamSanders commented 4 years ago

Following https://www.elastic.co/blog/text-similarity-search-with-vectors-in-elasticsearch.

This allows for similarity based queries rather than keyword based queries.

Example: Find k-nearest neighbors of the embedding vector of "I haven't been able to buy paper towels for three weeks!"

This will allow the clustering and visualization of topic-targeted sets of tweets from the COVID-19 tweet dataset.

AbrahamSanders commented 4 years ago

Done in 05dc73831c161aef60592f4dbdbdfd57a7afe0c3 and 85d92ac5a35f4ba4cbdb2abc921152daae151f9c