Classification of Crypto-Currency Data Using Data Mining Techniques

Author(s): Atish Tangawade and Aniket Muley

Abstract:

Crypto-currencies became popular with the emergence of Bitcoin and Litcoin have shown an unprecedented growth over the last few years. After more than a decade of existence, crypto-currencies may now be considered an important class of assets presenting some unique appealing characteristics but also sharing some features with real financial assets. Crypto-currencies price behaviour is still largely unexplored, presenting new opportunities for researchers to highlight similarities and differences with standard financial prices. Consequently, the cryptocurrency market can be a conducive arena for investors, as it offers many opportunities. However, it is difficult to understand. The current article focused on the Bitcoin and Litcoin Cryptocurrency informational collection with classification perspective. Our idea is to recognize the Artificial Intelligence (AI) device that, classify Cryptocurrency brings about more proficient way. To perform arrangement expectation, we have applied different data mining methods viz., Decision Tree (DT), Random Forest (RF), Extreme Boost model, Support Vector Machine (SVM) Model, Linear and Generalized Linear Models, brain organization or neural network (NN) model, beneficiary administrator trademark (ROC) bend. We found that, general linear model is more liked and it gives a lot of proficient outcomes. In this paper, various algorithms of data mining are demonstrated and outlined comprehensively for the Cryptocurrency data. Overall, dealing with the large scale dataset, the results obtained through the various machine learning techniques we found that, low price, high price, volume of the Crypto currency parameters plays significant role in classification and in future this study will be useful for prediction perspective.

Keywords: Classification, Cryptocurrency, Data Mining, Optimization.

Pages: 24-31      |       368 View       |       23 Download