School of Computer Science, Hubei University of Technology, Wuhan 430068, China.
Fujian Provincial Key Laboratory of Data Intensive Computing, Quanzhou 362000, China.
Sensors (Basel). 2022 Jul 28;22(15):5645. doi: 10.3390/s22155645.
Data are a strategic resource for industrial production, and an efficient data-mining process will increase productivity. However, there exist many missing values in data collected in real life due to various problems. Because the missing data may reduce productivity, missing value imputation is an important research topic in data mining. At present, most studies mainly focus on imputation methods for continuous missing data, while a few concentrate on discrete missing data. In this paper, a discrete missing value imputation method based on a multilayer perceptron (MLP) is proposed, which employs a momentum gradient descent algorithm, and some prefilling strategies are utilized to improve the convergence speed of the MLP. To verify the effectiveness of the method, experiments are conducted to compare the classification accuracy with eight common imputation methods, such as the mode, random, hot-deck, KNN, autoencoder, and MLP, under different missing mechanisms and missing proportions. Experimental results verify that the improved MLP model (IMLP) can effectively impute discrete missing values in most situations under three missing patterns.
数据是工业生产的战略资源，高效的数据挖掘过程将提高生产力。然而，由于各种问题，在实际收集的数据中存在许多缺失值。由于缺失数据可能会降低生产力，因此缺失值插补是数据挖掘中的一个重要研究课题。目前，大多数研究主要集中在连续缺失数据的插补方法上，而少数研究则集中在离散缺失数据上。本文提出了一种基于多层感知器（MLP）的离散缺失值插补方法，该方法采用动量梯度下降算法，并利用一些预填充策略来提高 MLP 的收敛速度。为了验证该方法的有效性，实验比较了在不同缺失机制和缺失比例下，该方法与模式、随机、热插补、KNN、自动编码器和 MLP 等八种常见插补方法的分类精度。实验结果验证了在三种缺失模式下，改进的 MLP 模型（IMLP）在大多数情况下都能有效地插补离散缺失值。
