Skip to main content

An implementation of the generalized merged distance to evaluate entity resolution

Project description

entity-resolution-evaluation

A python package to evaluate entity resolution

This package allows to evaluate entity resolution by efficiently computing several state of the art metrics : basic merge distance, precision, recall, variation of information. It's using the slice algorithm from the paper :

Menestrina, David and Whang, Steven Euijong and Garcia-Molina, Hector (2010) Evaluating Entity Resolution Results http://ilpubs.stanford.edu:8090/975/3/ERMetricVLDB.pdf

Getting Started

Installing

pip install entity-resolution-evaluation

Testing

Evaluate your resolution R against the gold standard S using a metric.

Examples

S = [[0, 1], [2, 3, 4], [5]]
R = [[0, 1, 2], [3, 4], [5]]

evaluate(R,S, 'bmd')
# returns 2

To go from R to S, you have to do 1 split and 1 merge.

evaluate(R,S,'precision')
# returns 0.5, 

Half of the pairs of R are incorrect. (0,2) and (1,2) are incorrect. (0,1) and (3,4) are correct

evaluate(R,S,'recall')
# returns 0.5

Half of the pairs of S are present in R. (0,1) and (3,4) are present. (2,3) and (2,4) are absent.

evaluate(R,S,'variation of information')
# returns 0.6365141682948129

Metrics

You can currently compute the following metrics :

metric value if perfect bounds intepretation
'bmd' 0 [0,infinity] basic merge distance : the number of split and merge necessary to go from R to S
'precision' 1 [0,1] proportion of pairs in R present in S
'recall' 1 [0,1] proportion of pairs in S present in R
'f1' 1 [0,1] harmonic mean of precision and recall
'variation_of_information' 0 [0,infinity] amount of information that is lost and added to go from R to S

Credits

Please visit the Hopkins mission page for more information

License

MIT License

Copyright (c) 2018 Ministère de l'Action et des Comptes Publics, Paul Boosz, Benoît Guigal

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

entity_resolution_evaluation-0.0.2.tar.gz (6.0 kB view details)

Uploaded Source

Built Distribution

File details

Details for the file entity_resolution_evaluation-0.0.2.tar.gz.

File metadata

  • Download URL: entity_resolution_evaluation-0.0.2.tar.gz
  • Upload date:
  • Size: 6.0 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.12.1 pkginfo/1.4.2 requests/2.18.4 setuptools/40.4.3 requests-toolbelt/0.8.0 tqdm/4.27.0 CPython/3.6.5

File hashes

Hashes for entity_resolution_evaluation-0.0.2.tar.gz
Algorithm Hash digest
SHA256 77609c039f46e6d56175f9e7e078b266cb8d7abe56ed0a061166a17f1863a22b
MD5 826809dad9199eec7fb176432b442a70
BLAKE2b-256 caf0651ef32b8a6ae6fa1398330d7584cb4d2e8aff15bf55d0f6235d1e3deec4

See more details on using hashes here.

File details

Details for the file entity_resolution_evaluation-0.0.2-py3-none-any.whl.

File metadata

  • Download URL: entity_resolution_evaluation-0.0.2-py3-none-any.whl
  • Upload date:
  • Size: 8.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/1.12.1 pkginfo/1.4.2 requests/2.18.4 setuptools/40.4.3 requests-toolbelt/0.8.0 tqdm/4.27.0 CPython/3.6.5

File hashes

Hashes for entity_resolution_evaluation-0.0.2-py3-none-any.whl
Algorithm Hash digest
SHA256 75a91187fc873d7d041a45679beaf429d56ff09d07e7d0b1b3c99e023d6df17b
MD5 9080435f97862328f287372b5f15dc14
BLAKE2b-256 4cce01297dcfa2e1b941994535a6e0b7544c448b4e35af6c0e8dfe824c6e7422

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page