Robust Missing Value Estimation: A Comparative Study of Closet Fit Algorithm and Traditional Methods
Keywords:
missing value estimation, Closet Fit Algorithm, moving average, data quality, robustness.Abstract
Missing data poses significant challenges in data analysis, compromising accuracy and reliability. This study investigates the performance of three missing value estimation algorithms: Simple Moving Average, Moving Average with Range, and Closet Fit
Algorithm (CFA). A comprehensive evaluation using real-world datasets reveals CFA's superiority in accuracy, scalability, and robustness. CFA's iterative refinement approach effectively handles non-linear relationships and diverse data distributions,
outperforming traditional methods. The findings highlight CFA's potential in enhancing data quality and reliability, contributing to the development of more accurate missing value estimation methods.
References
Gaur, S., & Dulawat, M. S. (2011). Improved closest fit techniques to handle missing attribute values. Journal of Computer and Mathematical Sciences, 2(2), 384-390.
Bhavsar Nidhi S., Khushbu Yadav, and Darshanaben Dipakkumar Pandya. "A New Clustering Approach for Anomaly Intrusion Detection." International Journal of Scientific & Research Studies, vol. 5, no. 23, 2023, pp. 10667.
Pandya, D. D., Modi, B. K., & Bhavsar Nidhi S. (2022). Closest fit approach for atypical value revealing and deciles range anomaly detection method for recovering misplaced value in data mining. International Journal of Scientific Research in Computer Science, Engineering, and Information Technology, 7(4), Article 25.
