Skip to content

arso-project/tantivy-ir-evaluation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

42 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Tantivy Information Retrieval Evaluation

As far as we know this is the first Information Retrieval evaluation of Tantivy, a "full text search engine library" written in rust.

Datasets

  • Movieset

Evaluation

Metrics

Tasks

For the dataset MOVIE search on the title, body and a combination of both fields (fulltext) is evaluated.

Lucene

As a reference we used lucene.

Task MP@3 MP@R MAP
MOVIE (Title) 0.134 0.065 0.039
MOVIE (Body) 0.434 0.241 0.250
MOVIE (Fulltext) 0.5 0.264 0.237

Tantivy

with limit = 100

Task MP@3 MP@R MAP
MOVIE (Title) 0.1 0.066 0.039
MOVIE (Body) 0.367 0.282 0.291
MOVIE (Fulltext) 0.5 0.294 0.283

Conclusion

For MOVIE Lucene and Tantivy return compareable results. Tantivy is even slightly better, although this could be a result of the limit to 1000 documents.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published