Research Information System
International Conference on Control and Optimization with Industrial Applications , Baku, Azerbaijan, 24 - 26 August 2022, vol.2, pp.375-377, (Full Text)
Risk analysis of investment projects is critical because of the significant economic losses that can be generated by risk factors. It is necessary to take into account numerous factors and techniques while doing risk analysis on investment projects. An unpredictably unclear future, involving numerous kinds of risk and uncertainty, frequently affects the expected outcomes for feasibility as well as profitability. In most cases, investment appraisals make use of tools like cost-benefit analysis and risk analysis. Net Present Value (NPV) and Interest Rate Return (IRR) are the most commonly used metrics in cost-benefit analysis (excluding the process of risk analysis). Sensitivity analysis, probability analysis, break-even analysis, decision trees, uncertainty analysis, and the Monte-Carlo simulation technique are well-known and commonly used approaches [2]. According to a review of the literature and evaluation reports, it is typical to find that the assessment of such values does not provide enough information for a meaningful decision, particularly for energy investment projects such as the production of oxygen. Deep learning algorithms have been extensively studied in the realm of risk or investing. To improve investors’ investment strategy and rate of return, Xie and Liu examined and discussed the use of deep learning to financial data processing and investment models, as well as investment plan establishment.[4, 5] Convolutional Neural Network (CNN) was shown to be the most effective model for financial forecasting by Sohangir et al. Another research found that the accuracy of financial risk predictions and the optimization of investment portfolios were critical, so he compared the performance of LSTM and Recurrent Neural Network (RNN) in classification; the results indicated that the LSTM had a better performance, which was of great significance for maximizing investment profits [1, 3]. Based on the existing studies, the paper propose to use LSTM which is a specialized RNN technique and CNN as deep learning techniques to build a risk assessment model for investment projects and analyze investment risks scientifically and effectively. Using these deep learning techniques, we will assess the risk analysis of the investment projects that businesses may identify and mitigate investment risks.