Optimizing Self-organizing Lists-on-Lists Using Pursuit-Oriented Enhanced Object Partitioning

Abstract

Central to the field of Computer Science is the issue of storing, maintaining and retrieving data, and the “list” structure, maintained as a Singly-Linked-List (SLL), leads to one such Adaptive Data Structure (ADS). Recently, researchers have proposed the concept of hierarchical Singly-Linked-Lists on Singly-Linked-Lists (SLLs-on-SLLs), where the primitive elements of the primary list are, in and of themselves, sub- lists. The question of knowing which elements should be in the respective sub-lists is far from trivial, and is achieved using Learning Automata (LA). This paper demonstrates how we can incorporate one such LA, namely the Pursuit-Enhanced Object Migration Automaton (PEOMA) to improve the performance of the ADS operating in Non-stationary Environments (NSEs). The ADS is designed as a hierarchical SLL-on- SLL. The hierarchical concept employs a sub-context data structure to group together the elements that have a probabilistic dependence in the “unknown” distribution of the Environment. In this paper, we propose that the PEOMA reinforcement scheme can be powerful in learning the probabilistic distribution of the Environment to capture dependent elements within the sub-groups. The PEOMA improves on its predecessor algorithm by incorporating the Pursuit technique, which increases the likelihood of selecting superior actions with higher reward estimates by “pursuing” the currently known “best” action. The research shows that the PEOMA-enhanced SLLs-on-SLLs provide results that are an order of magnitude superior to the “de-facto” MTF and TR schemes used in such Environments with so-called “locality of reference”. Also, the results surpass the performances of EOMA-enhanced hierarchical SLLs-on-SLLs schemes in NSEs including when the data-structure has some knowledge of the change in the dependence distribution of the Environment.