Initial coin offering, which is called ICO, is rapidly raised its volume as the bypassed crowdfunding. It reached $3.5B in 2017. This is mainly due to five of successful ICOs, Filecoin $257M, Tezos $232M, EOS $185M, Paragon $183M and Bancor $153M. Since these great successes, many startups are jumping into ICO. However, there is no standard criteria to analyze this investment market. Currently, most of token sales rely on their white papers and SNS channels to attract investors, and also we can find its popularity on Google Trends. The goal of this paper is thoughtfully exploring these data sources using big data analysis tools, then building the visualized analytics on this market. Our main approach is based on domain- knowledge to find related criteria. There are several criteria which are widely accepted in this investment markets. We tested the linear correlation between these features and ICO result to validate these criteria. We collected 500 ICO startups to build our database from the ICO-trackers. It consists of founder, white papers, likes and followers in Twitter, ICO results so on. We also collected USD exchange rates of cryptocurrencies to see the price effect of ICO investments. In this data, we considered top 10 ICO startups in 2017 as the successful cases. Using selected criteria, we modeled the prediction of success by the logistic regression method, which may solve our classification problem, successful or not. Our final result is presented as a visual marketing analytics application which is designed on Elasticsearch and Kibana. Kibana is web-based visualization system which connected to the full-search NoSQL database, Elasticsearch. This system analyzes current ICOs status in real-time then visualize whole analysis on the Kibana dashboard. This dashboard provides summarized criteria analysis and the prediction result to monitor ICOs.
Jin, Seungmin; Ali, Rashid; and Vlasov, A. V., "Cryptoeconomics: Data Application for Token Sales Analysis" (2017). International Conference Information Systems 2017 Special Interest Group on Big Data Proceedings. 1.