https://github.com/anmolsonthalia/bitcoin_value_predictor
Creating a Bitcoin value predictor by correlating social media sentiment with market trends using sentiment analysis and machine learning algorithms.
https://github.com/anmolsonthalia/bitcoin_value_predictor
bitcoin-payment bitcoin-prediction data-extraction data-preprocessing linear-regression lstm-model time-series
Last synced: 10 months ago
JSON representation
Creating a Bitcoin value predictor by correlating social media sentiment with market trends using sentiment analysis and machine learning algorithms.
- Host: GitHub
- URL: https://github.com/anmolsonthalia/bitcoin_value_predictor
- Owner: anmolsonthalia
- License: mit
- Created: 2024-03-24T05:17:40.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2024-03-24T05:25:21.000Z (almost 2 years ago)
- Last Synced: 2024-11-15T04:07:17.647Z (about 1 year ago)
- Topics: bitcoin-payment, bitcoin-prediction, data-extraction, data-preprocessing, linear-regression, lstm-model, time-series
- Language: Jupyter Notebook
- Homepage:
- Size: 257 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# BitCoin-Value-Predictor
## Abstract:
The project attempts to predict the future value of Bitcoins by identifying the correlation between social media sentiment and market sentiment. We will achieve this by collecting user feeds from social media such as twitter, facebook and linkedin. Once we have our corpus we will map their associated sentiments using IBM Watson’s Natural Language Understanding API. While mapping sentiments to our corpus we attempt to capture granular level categories namely joy, anger, happiness, etc. We use these as feature vectors to our ML/DL algorithms. Then we compare the results of the different algorithms and choose the one with the best accuracy score.
## Technologies:
* Programming Languages: Python, Java
* Big data technologies: Spark ML, Spark-SQL, Hadoop Mapreduce
* Libraries: Pandas, Matplotlib, Scikit-learn, TensorFlow , Keras
## Data Sources:
1. Twitter Api to get the tweets about BitCoins/Cryptocurrencies.
2. LinkedIn Api to get the corpus on blogs.
3. Web Scraping to get data from News articles.
## Dropbox link with data:
https://www.dropbox.com/s/oy5zcf4aiorr0dr/Archive.zip?dl=0
## References:
* J. Bollen and H. Mao. Twitter mood as a stock market predictor. IEEE Computer, 44(10):91–94.
* http://dataconomy.com/2014/07/bitcoin-big-data-can-predict-future-value-virtual-currency
* Mittall et Goel. Stock Prediction Using Twitter Sentiment Analysis
* http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.375.4517&rep=rep1&type=pdf
* https://arxiv.org/pdf/1610.09225.pdf