Fachgebiet Datenbanken und InformationssystemeAbschlussarbeiten
Mining social media to discover the factors that affect bitcoin price [BSc]



In recent year, bitcoin has become the most popular cryptocurrency, which gains the attention of many investors and data scientists. Predicting the price of bitcoin is a significant and challenging data analysis task, which would facilitate investment decisions. The difficulty in this task is understanding the multiple factors that affect the bitcoin price trend. Mining the enormous social media text data on the internet to discover these factors is a potential direction. Most of the existing work is limited to sentiment analysis and does not conduct in-depth exploration of the text content [1][2]. Recent work made primary attempts to find the relevance between keywords in social media text and the bitcoin price change [3].

Problem / Task:  

Our task is mining the social media text to discover the factors that affect bitcoin price. First, we want to pre-process the web data to generate a high-quality social media text corpus. Then we want to use existing natural language processing methods for semantic analysis to extract relevant information from the text. This information will be further classified with machine learning techniques. Finally, we can analyze and summarize the possible factors that affect bitcoin price. If possible, we can also attempt to predict the price change based on these discovered factors.


  • Programming skills
  • Interest in big data mining
  • Interest in natural language processing & machine learning 

Related Work:

[1] Germán Cheuque Cerda, Juan L. Reutter: Bitcoin Price Prediction Through Opinion Mining. WWW (Companion Volume) 2019: 755-762

[2] Giulia Serafini, Ping Yi, Qingquan Zhang, Marco Brambilla, Jiayue Wang, Yiwei Hu, Beibei Li: Sentiment-Driven Price Prediction of the Bitcoin based on Statistical and Deep Learning Approaches. IJCNN 2020: 1-8

[3] Andrew Burnie, Emine Yilmaz: An Analysis of the Change in Discussions on Social Media with Bitcoin Price. SIGIR 2019: 889-892 

Advisor and Contact:

Binger Chen <chen@tu-berlin.de> (TU Berlin)

Prof. Dr. Ziawasch Abedjan <abedjan@dbs.uni-hannover.de> (LUH)