Text Classification using Reuters Data

1. reutersdata-scratch-classicML.ipynb : simple data processing from raw source and classical machine learning (SVM, Logistic Regression)

2. reutersdata_keras_processing.ipynb : keras way of processing reuters data (data util functions are integrated within keras)

3. reutersdata-keras-modeling_final.ipynb : data loaded from keras and trained with DL models (MLP, transformer)

Acknowledgement and References:

Adapted from various resources, tools, and descriptions as follows -

data: https://archive.ics.uci.edu/ml/machine-learning-databases/reuters21578-mld/ or http://kdd.ics.uci.edu/databases/reuters21578/reuters21578.html

https://martin-thoma.com/nlp-reuters/

https://github.com/MartinThoma/algorithms/tree/master/ML/nlp

https://www.quantstart.com/articles/Supervised-Learning-for-Document-Classification-with-Scikit-Learn/

keras-team/keras#12072

https://github.com/SteffenBauer/KerasTools/tree/master/Reuters_Analysis

https://colab.research.google.com/github/aayushkumarjvs/deep-learning-with-python-notebooks/blob/master/3_6_classifying_newswires.ipynb#scrollTo=AbOwBCfMF0r7

https://phani1995.github.io/artificial/neural/network/2018/11/20/Reuters_newswire_topics_classification_keras_dataset.html

https://www.kaggle.com/drscarlat/reuters-document-classification-with-keras-tf

https://towardsdatascience.com/text-classification-in-keras-part-1-a-simple-reuters-news-classifier-9558d34d01d3

https://keras.io/examples/nlp/