NLP-With-Python

Natural Language Processing using Python (January 2025) - Lab Activities

Day 1:

1. Use Stop Words
2. POS tagging
3. Generate Antonyms
4. Generate Synonyms

Day 2:

1. Calculating frequency of top N words
2. Using NLTK and RE module to tokenize text, remove stop words, special characters
3. Using wordcloud library to generate word cloud from cleaned text

Day 3: Apply CountVectorization to the input data & answer the following:

1. Count Vectorizer Matrix
2. Vocabulary (unique words in corpus)
3. Calculating frequency of top N words (top 1 word = most frequent term)
4. Finding out words that appear in all/maximum sentences/documents
5. (Doubts):
   1. What would happen if we set stop_words='english' in the CountVectorizer?
   2. Should we manually explain the tone of 'happy' or should we write a code (Like for pick out term that occurs in all documents)

Day 4: Multinomial Naive Bayes model on 'Movie Review' dataset:

1. Pre processing steps - Using NLTK and RE module to:
   1. Tokenize text
   2. Remove stop words
   3. Remove special characters
2. Word Cloud for the data
3. MultiNB Model

Capstone Project: Major concepts used:

1. Words Tokenization
2. Stop Words
3. Lemmatization