Climate-Change-Sentiment-Classifier

Introduction

A climate-based products vendor seeks to gauge public perception of climate change to refine market research efforts and gain insights into consumer sentiment. To address this, we developed and deployed a machine-learning model capable of predicting individuals’ beliefs on climate change based on their tweets. The model classifies tweets into four categories: Pro (supporting man-made climate change), Anti (opposing man-made climate change), Neutral (neither supporting nor opposing), and News (providing factual information about climate change).

Goal of the Project

We aim to build and deploy a classification machine learning project capable of categorizing tweets into Pro, Anti, Neutral, or News. This classification enables our client to comprehend public sentiments on climate change, facilitating informed decision-making aligned with their target audience.

Skills and Tools Used

• Python libraries (Pandas, NumPy, Matplotlib, and Scikit-learn)
• Natural Language Toolkit (NLTK)
• Data visualization techniques
• Version control: GitHub
• Model deployment tools: Streamlit and AWS EC2 instance
• Collaboration and leadership skills
• Presentation abilities

Task Overview (My contribution)

Here is a high-level overview of my contributions:

• Removing URLs, numbers, and special characters to enhance text cleanliness.
• Standardizing text to lowercase for consistency.
• Stemming: Reducing words to their root form for uniformity.
• Removing stopwords: Eliminating common words with little meaning.
• Tokenization: Breaking text into words or sub-word units.
• Utilizing data visualization to identify patterns.
• Vectorization: Converting text data into numerical representations suitable for machine learning models.
• Feature selection for model training.
• Training classification models using machine learning algorithms such as logistic regression, linear SVC, and multinomial Naive Bayes.
• Model evaluation: Assessing model performance using the Mean F1-Score.
• Model deployment of the machine learning app: Deploying the application using Streamlit and AWS EC2 instance.

Ready to unlock the full potential of your data? I’m excited to collaborate and help your business thrive through data-driven insights. Get in touch!