Comparative Analysis of Results Obtained by Utilizing Expert Systems and Neural Networks in Financial Analysis

Artificial intelligence techniques in finance and financial markets: A survey of the literature

“…However, NNs cannot explain the results, while expert systems can explain each step's results. If new rules have to be added, expert systems can take these accounts, but for NNs, the additional data has to be gathered from the initial stage (Kapczynski & Zurek, 2001).…”

“…However, NNs cannot explain the results, while expert systems can explain each step's results. If new rules have to be added, expert systems can take these accounts, but for NNs, the additional data has to be gathered from the initial stage(Kapczynski & Zurek, 2001).Since NNs suffer from the slow learning period and are unable to explain their reasoning, researchers have experimented with adding other techniques to improve the results(Se-Hak & Kim, 2004). In one case, it was found that using rough set components could reduce the input variables from 28 ratios to 18 ratios and improve the predict-Similarly, the error backpropagation algorithm was used to train NNs for predicting the effect of currency movements on stock prices…”