Abstract
Investment risks in IT project development are heightened by uncertainty, incomplete information, and fluctuating projected cash flows. These challenges are exacerbated by the lack of robust statistical data, leaving stakeholders with limited tools for making informed decisions. This research addresses these issues by proposing a novel methodology for optimizing risk management in investment processes using advanced deep learning techniques. The study aims to develop and validate an algorithm that quantifies and mitigates investment risks through the integration of machine learning models and convolutional neural networks. A key component of this work is the Risk, Investment, and Compliance (RIC) method, which combines multiple financial indicators into a composite scoring system. The methodology was validated using five years of historical financial datasets from reputable sources, and applied to ten companies across diverse industries to analyse financial performance, market behaviour, and consumer sentiment. Key datasets include Kaggle’s Twitter Dataset, encompassing 1.5 million tweets to assess market sentiment, McKinsey’s dataset of 500 million consumer interactions, and daily updates from Yahoo Finance. The findings demonstrate that the RIC methodology effectively distinguishes between high-risk and secure investments. Companies scoring above 60% were identified as strong investment opportunities, while those below 30% were flagged as high-risk ventures. These results provides a robust framework for managing risks in IT investment projects, enabling more reliable decision-making under uncertainty and offering broad applications across industries.