Traditional time series prediction methods are unable to handle the complex nonlinear relationship of a large data set. Most of the existing techniques are unable to manage multiple dimensions of a data set, due to which the computational complexity escalates with the increasing size of a data set. Many machine learning (ML) methods are unable to handle known unknown predictions. This paper presents a new forecasting method in the neural network structure based on the induced ordered weighted average (IOWA) weighted average (WA) and fuzzy time series. The proposed model is more efficient than existing complexity handling fuzzy time series prediction methods and other traditional time series prediction methods. The proposed model can accommodate the IOWA operator, weighted average, and relevance degree of each concept in a particular problem for a fuzzy nonlinear prediction. The contribution of this paper is twofold. First, it contributes to theory by proposing a new IOWAWA layer in the neural network to handle complex nonlinear prediction for a large data set. The second contribution is the application of the approach to predict nonlinear stock market data. The robustness of the approach is tested using Australian Securities Exchange (ASX) stock data by considering a case study of the housing and property sector. We further compare the prediction accuracy of the approach with sixteen existing methods. The experimental results demonstrate that the proposed model outperforms existing methods.

中文翻译：

---

使用 ANFIS 诱导的 OWAWA 进行复杂股市预测的预测智能

传统的时间序列预测方法无法处理大型数据集的复杂非线性关系。大多数现有技术无法管理数据集的多个维度，因此计算复杂性随着数据集大小的增加而升级。许多机器学习 (ML) 方法无法处理已知未知预测。本文提出了一种基于诱导有序加权平均（IOWA）加权平均（WA）和模糊时间序列的神经网络结构预测方法。所提出的模型比现有的复杂性处理模糊时间序列预测方法和其他传统的时间序列预测方法更有效。所提出的模型可以适应特定问题中每个概念的 IOWA 算子、加权平均和相关度，用于模糊非线性预测。本文的贡献是双重的。首先，它通过在神经网络中提出一个新的 IOWAWA 层来处理大型数据集的复杂非线性预测，从而为理论做出贡献。第二个贡献是该方法在预测非线性股票市场数据中的应用。通过考虑住房和房地产行业的案例研究，使用澳大利亚证券交易所 (ASX) 的股票数据测试了该方法的稳健性。我们进一步将该方法的预测准确性与 16 种现有方法进行比较。实验结果表明，所提出的模型优于现有方法。