Bitcoin Trend Analysis
The Bitcoin Trend Analysis App is a data-driven project designed to visualize and analyze trends in Bitcoin-related tweets and market data. Built with Streamlit, the application combines data preprocessing, advanced visualizations, and machine learning to explore the interplay between social media sentiment and market behavior. Developed as part of a learning initiative, it demonstrates proficiency in data management, visualization, and text mining techniques.
DATA VIZUALISATIONTEXT MININGSTREAMLIT APPSENTIMENT ANALYSISNLPMACHINE LEARNING
6/16/20241 min read
Main Features
Data Management
The project utilizes Jupyter Notebooks and Pandas for data cleaning and preparation, ensuring the datasets are analysis-ready:Bitcoin Market History: Historical Bitcoin market data was analyzed and cleaned for meaningful insights.
Bitcoin Twitter History: Tweet data related to Bitcoin underwent preprocessing for sentiment and trend analysis.
Cross-Referencing: Timeframes in the datasets were aligned to create overlapping windows for correlation analysis.
Data Visualization
Preprocessing and visualization bring out key insights, leveraging:NLTK Tools: Text preprocessing, including stemming and stop-word removal.
Feature Engineering: Creation of sentiment scores using VaderSentiment, time-series aggregations, and word clouds.
Trend Analysis: Statistical insights and visualizations using Numpy, Scipy, Matplotlib, and Seaborn.
Correlation and Predictive Analysis
Advanced analytics highlight the relationships between sentiment and market trends:Correlation Study: Evaluates how Bitcoin tweets correlate with market movements.
Predictive Modeling: Interactive exploration of predictions using a linear regression model built with Scikit-Learn.
Interactive Dashboards: Streamlit widgets and Dash Plotly visualizations provide an immersive experience for users.
Technology Stack
Languages and Tools: Python, Jupyter Notebook.
Libraries: Pandas, Numpy, Scikit-Learn, NLTK, Matplotlib, Seaborn, Dash Plotly.
Applications: Sentiment analysis, trend visualization, predictive modeling, correlation analysis.
User Interface with Streamlit: Interactive and user-friendly dashboard for visualizing and exploring Bitcoin trends.
