Smart Sort: Design and Analysis of a Fast, Efficient and Robust Comparison Based Internal Sort Algorithm

TL;DR

Smart Sort is a novel comparison-based internal sorting algorithm that combines heap construction with partitioning, offering robustness and efficiency validated through empirical and theoretical analysis across various probability distributions.

Contribution

The paper introduces Smart Sort, a new robust sorting algorithm that fuses heap construction with quicksort partitioning, and provides comprehensive empirical and theoretical performance analysis.

Findings

Exhibits robustness across diverse probability distributions.

Achieves excellent average-case performance.

Potential for further improvement with dynamic partition ratios.

Abstract

Smart Sort algorithm is a "smart" fusion of heap construction procedures (of Heap sort algorithm) into the conventional "Partition" function (of Quick sort algorithm) resulting in a robust version of Quick sort algorithm. We have also performed empirical analysis of average case behavior of our proposed algorithm along with the necessary theoretical analysis for best and worst cases. Its performance was checked against some standard probability distributions, both uniform and non-uniform, like Binomial, Poisson, Discrete & Continuous Uniform, Exponential, and Standard Normal. The analysis exhibited the desired robustness coupled with excellent performance of our algorithm. Although this paper assumes the static partition ratios, its dynamic version is expected to yield still better results.