Investigating risk assessment in post-pandemic household cryptocurrency investments: an explainable machine learning approach

Abstract

This study provides an applicable methodological approach applying artificial intelligence (AI)-based supervised machine learning (ML) algorithms in risk assessment of post-pandemic household cryptocurrency investments and identifies the best performed ML algorithm and the most important risk assessment determinants. The empirical findings from analyzing 13 determinants from 1,000 dataset collected from major cryptocurrency communities online suggest that the logistic regression (LR) algorithm outperforms the remaining six ML algorithms by using performance metrics, lift chart, and ROC chart. Moreover, to make the ML algorithm results explainable and tackle the “black box” issue, the top five most important determinants are discovered, which are the interaction between investment amount and investment duration, investment amount, perception of traditional investments, cryptocurrency literacy, and perception of cryptocurrency volatility. The present study contributes to the literature on risk assessment, especially on the household cryptocurrency investments in the post-pandemic era and the body of knowledge on explainable supervised ML algorithms.

A New Framework for Fraud Detection in Bitcoin Transactions Through Ensemble Stacking Model in Smart Cities

“…Li et al in [25] aimed to develop an efficient risk assessment technique for households investing in cryptocurrencies. The authors proposed an explainable ML approach that combined different ML algorithms with transparent and interpretable decision-making.…”

“…Proposed SINN-RD [24] Accuracy, Recall and Precision Existing techniques do not tackle Blackbox issue Used LR, KNN, RF, and NN [25] ROC, F1score, Specificity and Recall Existing techniques are not efficient for handling big data Used SVM [26] Precision, Recall, F1-score and AUC Binary classification issue in cryptocurrency Used ML techniques (LR, KNN, NB, RF and XGBoost) [27] Accuracy, F1-score, Precision and Recall Limited accuracy and performance of individual classifiers Used SNN and optimizeable DT [29] Precision, Recall, F1-score, and AUC. Existing research do not concentrate on fraud detection Used RF and XGBoost [30] Accuracy, F1-score and AUC-ROC curve Issue in Virtual Private Network (VPN) tunneling Used SVM, KNN, NB and RF [32] MSE, TP and FP illicit activities in [31].…”