Enhanced User-driven Ranking System with Splay Tree

Abstract

E-learning is one of the information and communication technology products used for teaching and learning process [35]. An efficient and effective way to construct trust relationship among peer users in e-learning environment is ranking. User-driven ranking systems are based only on the feedback or rating provided by the users. In [46-48] the authors provide a variety of trust and reputation methods. Certified Belief in Strength (CBS) [45] is a novel trust measurement method based on reputation and strength. In [38] author presents a recommendation system based on the relevant feedback review to predict the user's interests, that are ranked based on the recommendations history they provide previously. Users with higher rating obtain high reputation compared to less scored users. In question answering websites like StackOverflow, new or low scored users are ignored by the community. This discourage them and their involvement with the community reduces further down, as power law states, alleged low users are pushed to the bottom of the ranking list. Avoid this condition by encouraging less reputed users and prevent them from moving further down in ranking level. Thus, low reputed users are provided with few more chances to participate actively in the e-learning environments. A splay tree is a Binary Search Tree with self-balancing skill. The splay tree brings the recently accessed item to the top of the tree, thus active users are always on the top of the tree. A splay tree is used to represent user's ranks, and to semi-splay low ranked users again in the tree thus preventing them from further drowning in the ranking list. The focus of this research work is to find and enhance low reputed users in reputation system by providing few more chances to take part actively in the e-learning environment using the splay tree. Normalized discounted cumulative gain (NDCG) acts as a decision part for identifying drowning users.