neer-match-utilities 1.0.1

Extended funcationality for NEural-symbolic Entity Reasoning and Matching

Neer Match Utilities

neermatch is a framework for entity matching that supports deep learning, symbolic learning, and hybrid approaches, using neural networks, gradient boosting, logit, or probit as classifiers. It is tailored to the needs of economic historians and quantitative economists, but generalizes to any domain requiring entity matching.

The neermatch framework encompasses three packages:

1. py-neer-match: The Python implementation of the basic functionalities (ANNs only). Learn more
2. py-neer-utilities: A Python package that provides additional models and functionalities to streamline and support the entity matching workflow. (this project)
3. r-neer-match: The R implementation of the basic functionalites. Learn more

The project is financially supported by the Deutsche Forschungsgemeinschaft (DFG) under Grant 539465691 as part of the Infrastructure Priority Programme New Data Spaces for the Social Sciences (SPP 2431). Reading the article Karapanagiotis and Liebald (2023) helps to understand the theoretical foundation and design of neermatch (note that the article refers to an earlier version of the framework, previously labeled as mlmatch).

The documentation provides examples of how neermatch may be used. The data used in these examples are available in this folder of the GitHub repository.

Contributors

Marius Liebald (maintainer)

Pantelis Karapanagiotis (contributor)

Erik Hellberg (contributor)

Installation

pip install neer-match
pip install neer-match-utilities

Official Documentation

The documentation is hosted under https://www.marius-liebald.com/py-neer-utilities/index.html

License

The package is distributed under the MIT license.

References

Gram, Dennis, Pantelis Karapanagiotis, Marius Liebald, and Uwe Walz. 2022. “Design and Implementation of a Historical German Firm-Level Financial Database.” ACM Journal of Data and Information Quality (JDIQ) 14 (3): 1–22. https://doi.org/10.1145/3531533.

Karapanagiotis, Pantelis, and Marius Liebald. 2023. “Entity Matching with Similarity Encoding: A Supervised Learning Recommendation Framework for Linking (Big) Data.” http://dx.doi.org/10.2139/ssrn.4541376.

———. 2024a. “NEural-symbolic Entity Reasoning and Matching (Python Neer Match).” https://github.com/pi-kappa-devel/py-neer-match.

———. 2024b. “NEural-symbolic Entity Reasoning and Matching (R Neer Match).” https://github.com/pi-kappa-devel/r-neer-match.

Lin, Tsung-Yi, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Dollár. 2017. “Focal Loss for Dense Object Detection.” In Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2980–88. IEEE. https://doi.org/10.1109/ICCV.2017.324.

Lipton, Zachary C., Charles Elkan, and Balakrishnan Naryanaswamy. 2014. “Optimal Thresholding of Classifiers to Maximize F1 Measure.” In ECML PKDD 2014.