modelling optimal predicting future cash flows using new data mining methods (a combination of artificial intelligence algorithms)

The purpose of this study was to present an optimal model for predicting future cash flows using a combination of artificial neural network with genetic algorithms (ann+ga) and particle swarm algorithms (ann+pso). due to the nonlinear relationship among accounting information, the study aimed to utilize these artificial intelligence algorithms to forecast future cash flows. the variables considered for prediction were accruals components and operating cash flows. the analysis was conducted using data from 137 companies listed on the tehran stock exchange during the period of 2009-2017. the findings of this study demonstrated that both neural network models, optimized with genetic algorithms and particle swarm algorithms, performed well in predicting future cash flows when all the variables presented in this study (a total of 15 predictor variables) were included. however, the results also indicated that the neural network optimized with particle swarm algorithm (ann+pso) exhibited a lower error coefficient, indicating better efficiency and higher prediction accuracy compared to the neural network optimized with genetic algorithms (ann+ga).