COMPETITIVE INTELLIGENCE AND DECISION-MAKING ALGORITHM USING MACHINE LEARNING FOR INDUSTRIAL SECURITY

Abstract

The purpose of this scientific article is to show what competitor data analytics can do with machine learning and neural networks. In this study, we analyzed data on potential partners of the Department of Defense Office of Hearings and Appeals (DOHA) of the USA and obtained a trained algorithm that can help in making decisions based on keywords, which can minimize reputational risks. The published dataset of the Department of Defense Office of Hearings and Appeals (DOHA) of the USA was selected for analysis of the initial data, which displayed the results of the screening of potential partners along with a text justification. This is the reason why we used Recurrent Neural Network (RNN) instead of Convolutional Neural Network (CNN). Neural networks are a very important part of machine learning. As a result, we have developed a trained machine learning model for recommending the best partners, that is, more proven partners, both professional and reputable. In addition, the developed machine learning model does not allow working with an organization of bad partners who could act in bad faith and carry reputational risks.