Data Cleaning Algorithm Based on Variance Optimization Weighted K-Nearest Neighbor Algorithm

Conventional data cleaning algorithms mainly use the FSC (feasible search circle) sliding quartile feasible search circle to calculate the efficiency of multiple time series, which is vulnerable to the impact of complete sequence feature scanning, resulting in poor performance evaluation indicators of the algorithm. Therefore, a data cleaning algorithm based on variance and optimized weighted <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$K$</tex> nearest neighbor algorithm is proposed. That is, the variance optimization weighted <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$K$</tex> nearest neighbor algorithm is used to build the data repeated cleaning model, and the key extraction sliding window detection is carried out, thus completing the data cleaning algorithm design. The experimental results show that the performance evaluation indexes of the designed variance optimization weighted <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{K}$</tex> nearest neighbor data cleaning algorithm are good, reliable, and have certain application value, which has made certain contributions to promoting the intelligent development of big data.