Introduction. The effectiveness of investment decisions depends directly on the accuracy of the assessment of the investment attractiveness of an organisation performed by the investor. One of the ways to assess the investment attractiveness of a company is by using the indicators of enterprise value (EV). However, they do not guarantee accuracy when assessing the investment attractiveness of specific companies. As a result, such companies can be either overvalued or undervalued. This happens because EV does not take into account the specifics of the industry, the specifics of the smaller segments within the industry, and the specifics of economic activities in a particular region. Purpose. To determine the EV indicators which can be used for the accurate assessment of the investment attractiveness of companies operating in a particular industry. Methodology. In our study, we used a Classification and Regression Trees (CART) machine learning method. Results. The article presents the results of a comparative analysis of EV indicators used to assess the investment attractiveness of companies that can be found in scientific literature and bank references. We suggest that the EV indicators used to assess the investment attractiveness of companies should be calculated based on the analysis of the industry’s statistics using the CART algorithm. EV indicators should be determined by maximising the reduction of heteroscedasticity with regard to the investment potential. To test the suggested method, we used it to calculate the required current liquidity ratio, quick liquidity, the leverage ratio, and the long-term debt financing for pig farming enterprises in the Voronezh, Belgorod, Kursk, Lipetsk, and Tambov Regions. Conclusions. The study demonstrated that the classification trees method is the most accurate in calculating the EV indicators for assessing the investment attractiveness of companies operating in a particular industry. The suggested method proved to be the most accurate in predicting all the financial coefficients for pig farming enterprises. Thus, the leverage ratio (0.35) and the concentration of borrowed capital (0.65) proved that pig farming enterprises in the Voronezh Region should be financed by borrowed capital. The long-term debt financing ratio was 0.49. The assessment of the short-term investment potential of pig farming enterprises should be more precise, since the current liquidity and quick liquidity ratios are high: 2.73 and 1.44 respectively.