Cryptocurrency price analysis
With the introduction of Bitcoin back in 2008, the cryptocurrency market has evolved erratically and at unprecedented speed. There are approximately 2054 mineable cryptocurrencies as of October 2019 and the cryptocurrencies industry is estimated with a total market capitalization approximately about...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/78976 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | With the introduction of Bitcoin back in 2008, the cryptocurrency market has evolved erratically and at unprecedented speed. There are approximately 2054 mineable cryptocurrencies as of October 2019 and the cryptocurrencies industry is estimated with a total market capitalization approximately about $219 billion USD in September 2019. Due to the volatility of the cryptocurrency market, it comes with uncertainty for the people who are trying to invest or using them as actual currency. Some of the factors affecting the value of cryptocurrency include the supply and demand of the currency, popularity of the cryptocurrency, political adaptation and restrictions, and speculations of investors.
The primary objective of this project is to perform sentiment analysis on the news to determine whether the sentiment of news and events will affect the price of the cryptocurrencies. The aim of this project is to create a system is developed that performs the three tasks: the collection of price time series for multiple cryptocurrencies and information about events and news related to cryptocurrencies, execution of sentiment analysis on the news to determine if the news is positive or negative with respect to the price action of cryptocurrency, last but not least, analysis of sentiment to predict future price projections.
The project findings show that there exist certain forms of correlation between news and events and the price of cryptocurrencies. It also shows that extremely positive or negative news could be used to project a potential surge or sudden drop in price.
Recommendations for future enhancement work or potential research would be to improve the accuracy of the sentiment analysis value by enhancing the classifier. |
|---|