Skip to main content

Fast and simple probabilistic data matching package

Project description

healmatcher

  • healmatcher is a simple but fast probabilistic data matching package developed by NYULH HEAL Lab.
  • The package is best optimized for matching healthcare database (e.g. EHR) as it has designed to link Medicaid and Client Database System data.
  • Splink package is extensively being used to run core linkage processes.
  • Currently, the model supports 4 variables (sex, date of birth, last 4 digits of ssn, and first 2 letters of last name) to run the linkage process.

How to install

pip install healmatcher

How to use (example)

# Install package
!pip install healmatcher

# Load package
from healmatcher import hm

# create example dataset
testa = pd.DataFrame({
    'sex':[1,2,1,2,1,2,1,2,1,2],
    'dob':['2012-1-1','2011-12-1','1999-1-1','1998-11-1','2012-11-1','1984-1-1','1982-1-1','1975-1-1','1967-1-1','1954-1-1'],
    'ssn':[1111,2222,3333,4444,5555,6666,7777,8888,9999,1010],
    'ln':["as",'ss','zz','rr','ww','wa','tr','tt','hh','gq'],
    'PROVIDER_NUMBER':[2,1,1,1,1,1,1,1,2,1]
})
testb = pd.DataFrame({
    'sex':[2,2,1,1,1,2,1,2,1,1],
    'dob':['2012-1-1','2001-12-1','1999-1-1','1998-11-1','2012-11-1','1984-1-1','1982-1-1','1975-1-1','1967-1-1','1954-1-1'],
    'ssn':[1111,2222,3333,4444,5555,6666,7777,8888,9999,1010],
    'ln':["as",'ls','zz','rr','wb','wa','tr','tt','ha','gq'],
    'PROVIDER_NUMBER':[2,1,1,1,1,1,1,1,2,1]

# Run matching
hm(
    df_a = testa,
    df_b = testb,
    col_a=['sex','dob','ssn','ln'],
    col_b=['sex','dob','ssn','ln'],
    match_prob_threshold = 0.001,
    iteration = 20,
    model2 = True,
    blocking_rule_for_training_input = 'PROVIDER_NUMBER',
    onetoone = True,
    match_summary = True
)

Updates

  • use_save_model=True : Load pre-trained model to run matching
  • save_model_path = PATH : add path to load a model (json format)
  • export_model=True : argument to save current model
  • export_model_path=PATH : add path to save current model

Follow up

  • Please visit our repo if you have any questions.

Webpage

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

healmatcher-0.0.27.tar.gz (5.7 kB view details)

Uploaded Source

Built Distribution

healmatcher-0.0.27-py3-none-any.whl (5.9 kB view details)

Uploaded Python 3

File details

Details for the file healmatcher-0.0.27.tar.gz.

File metadata

  • Download URL: healmatcher-0.0.27.tar.gz
  • Upload date:
  • Size: 5.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/5.0.0 CPython/3.8.0

File hashes

Hashes for healmatcher-0.0.27.tar.gz
Algorithm Hash digest
SHA256 49fcd42db3ab42b4b27ef396b6669a126eaa433c7a40022ac057f453d46c1f48
MD5 a697ad32b33c39c7fc08654c375e71c9
BLAKE2b-256 b1c9b4be1919a9d347c9f4e34250aa3252f93aafe4e0241936c68eb1f5e484a5

See more details on using hashes here.

File details

Details for the file healmatcher-0.0.27-py3-none-any.whl.

File metadata

File hashes

Hashes for healmatcher-0.0.27-py3-none-any.whl
Algorithm Hash digest
SHA256 1b234b127ac1412f34724faf86f446518d0770b14aebeedda2840cfaa071ece8
MD5 e5955e72bff17c4b3f188f63c3b20326
BLAKE2b-256 d11eea8135e6ad604ebec63091ac9526f59d5cebda0176e66d2d93b10068cb73

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