langchangetrack 0.1.0

Package for statistically significant language change.

Package for statistically significant language change.

• Free software: BSD license

• Documentation: https://langchangetrack.readthedocs.org.

Features

• This package provides tools to detect linguistic change in a temporal corpora.

• We do this by capturing linguistic change through different methods.

Usage

cd examples/distributional/ ./ngrams_pipeline.sh ../data/temporal_corpus/ ./working/ ./output/ 1900 2000 5 locallinear 100 1000 “ngrams” ../data/temporal_corpus/common_vocab.txt 10

cd examples/pos ./pos_pipeline.sh ../data/temporal_corpus/ ./working/ ./output/ 1900 1930 5 ../data/temporal_corpus/common_vocab.txt 10

cd examples/freq ./freq_pipeline.sh ../data/temporal_corpus/ ./working/ ./output/ 1900 2000 5 ../data/temporal_corpus/common_vocab.txt 10

Requirements

• wheel==0.23.0

• argparse>=1.2.1

• numpy>=0.9.1

• scipy>=0.15.1

• more_itertools>=2.2

• joblib>=0.8.3-r1

• gensim==0.10.3

• statsmodels>=0.5.0

• changepoint>=0.1.0

• nltk>=3.0.0

• textblob>=0.9.0

• textblob-aptagger>=0.2.0

Installation

1. cd langchangetrack

2. pip install -r requirements.txt

3. python setup.py install

History

0.1.0 (2015-02-20)

• First release on PyPI.