Editing Imputation (statistics) (section)

===Harmonizing Spline regression and Tensor Decomposition algorithms for missing data imputation===

Where Matrix/Tensor factorization or decomposition algorithms predominantly uses global structure for imputing data, algorithms like piece-wise linear interpolation and spline regression use time-localized trends for estimating missing information in time series. Where former is more effective for estimating larger missing gaps, the latter works well only for small-length missing gaps. SPRINT (Spline-powered Informed Tensor Decomposition) algorithm is proposed in literature which capitalizes the strengths of the two and combine them in an iterative framework for enhanced estimation of missing information, especially effective for datasets, which have both long and short-length missing gaps.<ref>{{cite journal |last1=Sharma |first1=Shubham |last2=Nayak |first2=Richi |last3=Bhaskar |first3=Ashish |title=Harmonizing recurring patterns and non-recurring trends in traffic datasets for enhanced estimation of missing information |journal=Transportation Research Part C: Emerging Technologies |date=2025 |volume=174 |doi=10.1016/j.trc.2025.105083}}</ref>