DEV Community

Cover image for Similarity Search: Measuring Meaning From Jaccard to Bert
James Briggs
James Briggs

Posted on • Updated on • Originally published at pinecone.io

Similarity Search: Measuring Meaning From Jaccard to Bert

#machinelearning #python #deeplearning #datascience

Semantic Search: Measuring Meaning From Jaccard to Bert

I put together this article covering six of the most popular similarity search techniques. Three traditional, and three vector-based. Which are:

Traditional:

  • Jaccard Similarity

  • w-shingling

  • Levenshtein distance

and Vector-based:

  • TF-IDF

  • BM25

  • Sentence-BERT (SBERT)

Top comments (0)

Read next

geazi_anc profile image

Data Engineering with Scala: Mastering Real-Time Data Processing with Apache Flink and Google Pub/Sub

Geazi Anc -

vivekyadav200988 profile image

Understanding the ETL Process with Real-Time Data: Extraction, Transformation, Loading, and Visualization

Vivek Yadav -

mikeyoung44 profile image

Energy-Efficient Language Models: Addition is All You Need

Mike Young -

jonahklayer profile image

Django AI Assistant for VS Code

Jonah Katz -