Classifying the news articles into 4 categories namely Health, Business, Entertainment, Technology using the following ML models:

1. Logistic regression
2. Support Vector Machine
3. Naive Bayes
4. Random forest
5. K-NN

Summarize the news articles using Extractive text summarization ( selecting top sentences from the article)

1. Text rank algorithm (variation of page rank)
2. K-means clustering
3. Latent semantic analysis

Scrapped news articles from urls provided by UCI Machine Learning repository link
For scrapping the news articles, Newspaper3k library built in Python was used. The library contains nlp() method using which keywords and summary of the news article can be extracted.
Article's content and summary have been scrapped to create the data for the project. Code

The following libraries of Python have to be installed: pandas, sklearn, ntlk, newspaper3k
Run the following command to install pip install -r requirements.txt

Raw text has unwanted characters (\n,\t,$ etc) and contains stop words (a, an, the) which has to removed before generating the vector representation. The following text preprocessing techniques have been used:

1. Converting to lower case
2. Removal of stop words
3. Tokenize
4. Removing contractions (does'nt -> does not)
5. Stemming/Lemmatization