Financial trading decisions based on deep fuzzy self-organizing map

The volatility features of financial data would considerably change in different periods, that is one the main factors affecting applications machine learning quantitative trading. Therefore, to effectively distinguish fluctuation patterns markets can provide meaningful information for trading decision. In this article, a novel intelligent system based on deep fuzzy self-organizing map (DFSOM) companied with GRU networks proposed, where DFSOM utilized clustering acquire multiple an unsupervised way. Firstly, order capture trend and evade effect high noises data, images extended candlestick charts instead raw are processed obtained applied following learning, produced both price volume information. Secondly, by using features, two-layer constructed carry out clustering, models time scales improve processing time-dependent Thirdly, used implement prediction task, which model constructed. feasibility effectiveness proposed method verified various real datasets.