EFFECTS OF PERSONALIZED ALGORITHMIC SORTING ON INFORMATION SEEKING SUCCESS AMONG E-LEARNERS WITH DIVERSE PRIOR KNOWLEDGE

Abstract

This research examines the influence of individualized algorithmic sorting on the information-seeking effectiveness of e-learners with various levels of prior knowledge in Indian institutions of higher learning. The study also examinedthe effectiveness of adaptive ranking in facilitating retrieval, and also look at whether prior knowledge buffers this relationship. The quasiexperimental design took place in a group of 180 university students who were divided into personalized and non-personalized search interface groups. A structured pre-test was used to divide the participants into low, medium, and high rior knowledge. The success of Information seeking was assessed by the accuracy of retrieval, relevance score, time of completing a task, and satisfaction of the user. The analysis involved the use of twoway ANOVA and moderation regression analysis. Findings have shown that individual algorithmic sorting contributes greatly to the success of general information search (F = 12.84, p < 0.01). A significant main effect is also present in the case of prior knowledge (F = 18.67, p < 0.01). The ANOVA (F= 6.42, p < 0.05) interaction effect is statistically significant with the aid of moderation regression coefficients (β= 0.29, p < 0.05) and proves that prior knowledge mediates the relation between the two variables under consideration, i. e., algorithmic sorting and search results. Personalization especially helps low prior knowledge learners achieve a better relevance identification, and also reduced search time, whereas high prior knowledge learners are good in either interface condition. The results are favorable to the implementation of adaptive information retrieval systems in digital libraries and online learning platforms to guarantee equal and effective access to Information in the Indian higher education setting. The study provides empirical evidence to the Information Behavior and Information Retrieval research by showing the interaction of system-level personalization with cognitive characteristics. It also includes useful ideas to use when constructing inclusive digital classrooms, which can use personalized algorithmic ranking without restricting the exploratory behavior of more knowledgeable learners.