Clustering-Based Hybrid Approach for Multivariate Missing Data Imputation

In the era of big data, a significant amount of data is produced in many applications areas. However due to various reasons including sensor failures, communication failures, environmental disruptions, and human errors, missing values are found frequently These missing data in the observed data make a challenge for other data mining approaches, requiring the missed data to be handled at the preprocessing stage of data mining. Several approaches for handling the missing data have been proposed in the past. These approaches consider the whole dataset for making a prediction, making the whole imputation approach to be cumbersome. This paper proposes the procedure which makes use of the local similarity structure of the dataset for making an Imputation. The K-means clustering technique along with the weighted KNN makes efficient imputation of the missed value. The results are compared against imputations by mean substitution and Fuzzy C Means (FCM). The proposed imputation technique shows that it performs better than other imputation procedures.

@article{Dubey2020,
  title     = {Clustering-Based Hybrid Approach for Multivariate Missing Data Imputation},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {11},
  number    = {11},
  year      = {2020},
  publisher = {The Science and Information Organization},
  author    = {Aditya Dubey and Akhtar Rasool},
  doi       = {10.14569/IJACSA.2020.0111186},
  url       = {https://doi.org/10.14569/IJACSA.2020.0111186}
}