see the original item page
in the repository's web site and access all digital files if the item*



Near Real-Time Cryptocurrency Sentiment Analysis (EN)

Boutsoukis, Anastasios (EL)

Berberidis, Christos (EL)
Peristeras, Vassilios (EN)
Magnisalis, Ioannis (EN)

masterThesis

2023-04-11
2023-04-11T11:51:40Z
2023-01-07


Sentiment Analysis is a field of Natural Language Processing that addresses the problem of extracting sentiment or, more generally, opinion from the text. Obtaining deeper knowledge on this topic can be beneficial for a wide range of scientific fields. In this approach, we will use different tools and develop a web-based framework that enables and magnifies the analyzing process, producing at the same time information about the cryptocurrency field in dashboards and charts. This helps researchers and investors to classify, compare, and evaluate their studies about trading. The analysis is conducted on cryptocurrency title news data, scraped from Reddit and RSS feeds. Moreover, posts from Twitter were also obtained to detect and measure the sentiment of specific coins in real time. By filtering and analyzing the data using some Natural Language Processing tools and lexicons, their sentiment is determined by the emotion found in the data collected from Reddit, RSS feeds, and Twitter. Following the implementation of the pre-processing and normalization of the dataset gathered, using the open-source library, Dash and Plotly library, we will create a web-based interactive dashboard-framework that indicates cryptocurrencies’ prices and sentiment in different figures, from the data collected from these sources in real-time. To achieve this, the iteratively collected data in real-time will be discarded immediately after their process. (EL)


Sentiment analysis (EL)
Natural language processing (EL)
Cryptocurrency (EL)
RSS feeds (EL)

English

School of Science and Technology, MSc in Data Science
IHU (EL)

Default License




*Institutions are responsible for keeping their URLs functional (digital file, item page in repository site)