SKNN Algorithm for Filling Missing Oil Data Based on KNN
Open Access
- 1 October 2019
- conference paper
- Published by IOP Publishing in IOP Conference Series: Materials Science and Engineering
- Vol. 612 (3) , 032099
Abstract
Along with the rapid development of science and technology information in contemporary society, massive data has become a common phenomenon in information processing in various industries, and various data quality problems have also followed. Among them, data loss is a common problem. In the process of oilfield production, the dynamic data of production wells is increasing continuously every day, and data missing problems often occur. Aiming at the missing data, this paper proposes an improved data filling algorithm SKNN based on KNN. Based on the KNN algorithm, the algorithm uses the SMOTE algorithm to randomly generate the fill data range, and uses the multi-filling idea to find the weighted average as the fill data. The experiment with the production data of a well area in Daqing oilfield as the sample set verified that the SKNN algorithm could not only fill the data, but also improve the accuracy.Keywords
This publication has 4 references indexed in Scilit:
- SMOTE algorithm applying imbalanced data in higher education2016
- Missing Data Imputation Approach Based on Incomplete Data ClusteringChinese Journal of Computers, 2012
- An Adaptive Rule Based on Unknown Pattern for Improving K-Nearest Neighbor Classifier2010
- SMOTE: Synthetic Minority Over-sampling TechniqueJournal of Artificial Intelligence Research, 2002