This project seeks to build a Python software package to match entities between two tables using supervised learning. This problem is often referred as entity matching (EM). Given two tables A and B, the goal of EM is to discover the tuple pairs between two tables that refer to the same real-world entities. There are two main steps involved in entity matching: blocking and matching. The blocking step aims to remove obvious non-matching tuple pairs and reduce the set considered for matching. Entity matching in practice involves many steps than just blocking and matching. While performing EM users often execute many steps, e.g. exploring, cleaning, debugging, sampling, estimating accuracy, etc. Current EM systems however do not cover the entire EM pipeline, providing support only for a few steps (e.g., blocking, matching), while ignoring less well-known yet equally critical steps (e.g., debgging, sampling). This package seeks to support all the steps involved in EM pipeline.

The package is free, open-source, and BSD-licensed.

• Project Homepage: https://sites.google.com/site/anhaidgroup/projects/magellan/py_entitymatching
• Code repository: https://github.com/anhaidgroup/py_entitymatching
• Issue Tracker: https://github.com/anhaidgroup/py_entitymatching/issues

The required dependencies to build the packages are:

• pandas (provides data structures to store and manage tables)
• scikit-learn (provides implementations for common machine learning algorithms)
• joblib (provides multiprocessing capabilities)
• pyqt4 (provides tools to build GUIs)
• py_stringsimjoin (provides implementations for string similarity joins)
• py_stringmatching (provides a set of string tokenizers and string similarity functions)
• cloudpickle (provides functions to serialize Python constructs)
• pyprind (library to display progress indicators)
• pyparsing (library to parse strings)
• six (provides functions to write compatible code across Python 2 and 3)

py_entitymatching has been tested on Linux, OS X and Windows.