/ Natural Language Processing

Text vectorization tool to outperform TFIDF for classification tasks

Text vectorization tool to outperform TFIDF for classification tasks

textvec

Textvec is a text vectorization tool, with the aim to implement all the "classic" text vectorization NLP methods in Python. The main idea of this project is to show alternatives for an excellent TFIDF method which is highly overused for supervised tasks. All interfaces are similar to scikit-learn so you should be able to test the performance of this supervised methods just with a few changes.

Textvec is compatible with: Python 2.7-3.7.


WHY: Comparison with TFIDF

As you can read in the different articles1,2 almost on every dataset supervised methods outperform unsupervised.
But most text classification examples on the internet ignores that fact.

IMDB_bin RT_bin Airlines Sentiment_bin Airlines Sentiment_multiclass 20news_multiclass
TFOR 0.9088 0.7820 0.9173 NA NA
TFICF 0.8992 0.7661 0.9220 0.8067 0.8552
TFBINICF 0.8978 0.7628 0.9238 NA NA
TFRF 0.8977 0.7609 0.9207 NA NA
TFIDF 0.8923 0.7539 0.8939 0.7763 0.8335
TFPF 0.8949 0.7464 0.9164 NA NA
TF 0.8786 0.7286 0.9017 0.7865 0.7796
TFIR 0.8361 0.7159 0.9017 NA NA
TFCHI2 0.8734 0.6990 0.8900 NA NA
TFGR 0.8581 0.6793 0.8883 NA NA

Here is a comparison for binary classification on imdb sentiment data set. Labels sorted by accuracy score and the heatmap shows the correlation between different approaches. As you can see some methods are good for to ensemble models or perform features selection.

imdb_bin

For more dataset benchmarks (rotten tomatoes, airline sentiment) see Binary classification quality comparison


Install:

Usage:

pip install textvec

Source code:

git clone https://github.com/zveryansky/textvec
cd textvec
pip install .

HOW: Examples

The usage is similar to scikit-learn:

from sklearn.feature_extraction.text import CountVectorizer
from textvec.vectorizers import TfBinIcfVectorizer

cvec = CountVectorizer().fit(train_data.text)

tficf_vec = TfBinIcfVectorizer(sublinear_tf=True)
tficf_vec.fit(cvec.transform(text), y)

For more detailed examples see Basic example and other notebooks in Examples

Currently impletented methods:

  • TfIcfVectorizer
  • TforVectorizer
  • TfgrVectorizer
  • TfigVectorizer
  • Tfchi2Vectorizer
  • TfrfVectorizer
  • TfrrfVectorizer
  • TfBinIcfVectorizer
  • TfpfVectorizer

Most of the vectorization techniques you can find in articles1,2. If you see any method with wrong name or reference pls commit!


TODO

  • [ ] Add TFBNS
  • [ ] Remove dependence of sklearn
  • [ ] Tests
  • [ ] Docs
  • [ ] GridSearch for benchmark

GitHub