ASSESSING THE INVESTMENT RISK OF VIRTUAL IT COMPANY BASED ON MACHINE LEARNING

Abstract

A module for assessing the investment risks of a virtual IT company has been developed. It enables to reduce the time spent on assessing the investor’s risks of a virtual IT company. A detailed justification of each selected risk parameter that influences on the success of the investment project of the virtual IT Company has done. A developed algorithm for assessing the investment risk of the virtual IT company is based on machine learning and using the expert scoring method (10 experts from 20 implemented projects were involved) by 23 risk parameters. Forecasting of investment risk assessment modeling of the virtual IT company using machine learning is based on eight methods: Support Vector Classifier, Stochastic Gradient Decent Classifier, Random Forest Classifier, Decision Tree Classifier, Gaussian Naive Bayes, K-Neighbors Classifier, Ada Boost Classifier, Logistic Regression. In addition, a module was developed to support decision-making based on three methods with the best forecast, namely: Support Vector Classifier, Random Forest Classifier, K-Neighbors Classifier.