Multi-objective evolutionary feature selection for online sales forecasting

Sales forecasting uses historical sales figures, in association with products characteristics and peculiarities, to predict short-term or long-term future performance in a business, and it can be used to derive sound financial and business plans. By using publicly available data, we build an accurate regression model for online sales forecasting obtained via a novel feature selection methodology composed by the application of the multi-objective evolutionary algorithm ENORA (Evolutionary NOn-dominated Radial slots based Algorithm) as search strategy in a wrapper method driven by the well-known regression model learner Random Forest. Our proposal integrates feature selection for regression, model evaluation, and decision making, in order to choose the most satisfactory model according to an <em>a posteriori</em> process in a multi-objective context. We test and compare the performances of ENORA as multi-objective evolutionary search strategy against a standard multi-objective evolutionary search strategy such as NSGA-II (Non-dominated Sorted Genetic Algorithm), against a classical backward search strategy such as RFE (Recursive Feature Elimination), and against the original data set.

销售预测借助历史销售数据，结合产品特性与独有属性，对企业的短期或长期经营业绩进行预判，可用于制定科学合理的财务与业务规划。本研究依托公开可得数据集，构建了高精度的在线销售预测回归模型，该模型采用的新型特征选择方法，以知名回归模型学习器随机森林（Random Forest）驱动的包装式特征选择框架为载体，将多目标进化算法ENORA（Evolutionary NOn-dominated Radial slots based Algorithm）作为搜索策略。本研究方案整合了回归特征选择、模型评估与决策环节，旨在多目标场景下通过后验（a posteriori）流程筛选出最优模型。我们对以ENORA作为多目标进化搜索策略的模型性能进行测试，并与标准多目标进化搜索策略（如NSGA-II（Non-dominated Sorted Genetic Algorithm，非支配排序遗传算法））、经典递归后向搜索策略（如RFE（Recursive Feature Elimination，递归特征消除））以及原始数据集的性能进行对比。