Analysis of fuzzy logic methods for forecasting customer churn

The object of research is the process predicting churn customers telecommunications companies based on fuzzy logic and neural networks. carried out application an approach that implemented through combined use main assumption study hypothesis a network formed basis algorithms can improve accuracy customer relative to available solutions. This result can’t be achieved neglecting existing resource constraints requirements, which must determined separately for each case research. relevance problem forecasting with large number users considered. A model proposed feature this test sample normalized data used at networks, are processed form parameters membership functions correspond inference system, is, conclusions made apparatus. Also, find function, used. Such systems previously known information, learn, gain new knowledge, predict time series, perform image classification, besides, they quite visual user. methods considered, make it possible obtain in inference. expediency choosing these explained by fact were automatic control showed sufficiently high quality results. prospects using outflow shown, results software implementation presented.