NLP For Disaster Tweets

Data cleaning: Used multiple python libraries (pandas, neatext, nltk) to normalize and clean text, fix contractions, and remove: stopwords, emoji's and links. Lemmatization and accounted for duplicate and retweets.

Visualizations: Used multiple python libraries (Matplotlib, Seaborn, plotly) to create bar plots and histograms during our exploratory data analysis stage.

Natural Language Processing: 

• Sentiment Analysis: Implemented sentiment analysis using  (nltk) to assess and classify the emotional tone expressed in the text as positive, negative, or neutral.

• Vectorization: Employed vectorization techniques from (sklearn) to convert text data into a format suitable for machine learning algorithms. This step is crucial for transforming textual information into numerical features that can be utilized by machine learning models.

• Machine Learning : Leveraged machine learning algorithms from the (sklearn) library to develop models for various tasks related to the project. This may include classification, regression, or other relevant tasks based on project requirements.

• Deep Learning: Explored deep learning methodologies using state-of-the-art tools (BERT, Keras, and TensorFlow). Implemented deep learning models to capture intricate patterns and nuances within the textual data, enhancing the project's overall predictive capabilities.