Collaborative filtering adapted to recommender systems of e-learning

Abstract

In the context of e-learning recommender systems, we propose that the users with greater knowledge (for example, those who have obtained better results in various tests) have greater weight in the calculation of the recommendations than the users with less knowledge. To achieve this objective, we have designed some new equations in the nucleus of the memory-based collaborative filtering, in such a way that the existent equations are extended to collect and process the information relative to the scores obtained by each user in a variable number of level tests.

Introduction

Recommender systems (RS) cover an important field within collaborative services that are developed in the Web 2.0 environment [1], [2], [3] and enable user-generated opinions to be exploited in a sophisticated and powerful way. RS can be considered as social networking tools that provide dynamic and collaborative communication, interaction and knowledge.

RS cover a wide variety of applications [4], [5], [6], although, those related to movie recommendations are the most well-known and most widely-used in the research field [7], [8], [9]. Nevertheless, the collaborative e-learning field is strongly growing [10], [11], converting this area in an important receiver of applications and generating numerous research papers [12], [13] into the computer science field [14], [15] and into different areas [16], [17]. The endeavor to create distributed, federation [18] and grid [19] collaborative e-learning services are particularly interesting.

The RS stage that normally has the greatest influence on the quality of the results obtained is the collaborative filtering (CF) phase [20], [21]. CF is based on making predictions about a user’s preferences or tastes based on the preferences of a group of users that are considered similar to this user. A substantial part of the research in the area of CF centers on how to determine which users are similar to the given one; in order to tackle this task, there are fundamentally 3 approaches: memory-based methods, model-based methods and hybrid approaches.

Memory-based methods [22], [23] use similarity metrics [21] and act directly on the ratio matrix that contains the ratings of all users who have expressed their preferences on the collaborative service; these metrics mathematically express a distance between two users based on each of their ratios. Model-based methods [22] use the ratio matrix to create a model from which the sets of similar users will be established. Among the most widely-used models we have: Bayesian classifiers [24], neural networks [25] and fuzzy systems [26]. Generally, commercial RS use memory-based methods [27], while model-based methods are usually associated with research RS.

Regardless of the method used in the CF stage, the technical objective generally pursued is to minimize the prediction errors, by making the accuracy [28], [29], [30], [31] of the RS as high as possible; nevertheless, there are other objectives that need to be taken into account: avoid overspecialization phenomena, find good items, credibility of recommendations, precision, recall measures, etc.

Memory-based methods work on a table of U users who have rated I items. The prediction of a non-rated item i for a user u is computed as an aggregate of the ratings r of the K most similar users (k-neighborhoods) for the same item i. The most common aggregation approaches are the average and the weighted sum; the similarity approaches usually compute the similarity between two users x and y: sim(x,y) based on their ratings of items that both users have rated. The most popular similarity metrics are Pearson correlation and cosine.