Search

Over the past decade, computer games and other interactive technologies have shown great potential when used in innovative ways to enhance learning. It is now known that learning is associated not only with cognitive ability but also with affect. The incorporation of affective embodied pedagogical agents (EPAs) in computer programs for learning can significantly influence learner beliefs and efficacy. However, there have been a number of criticisms and contradictory empirical findings calling into question earlier results and theoretical claims on the effectiveness of using EPAs. Therefore, this paper reviewed studies that evaluated the effectiveness of affect in EPAs in a computer-based learning environment (CLE). The statistical meta-analytic review (k = 30, n = 2,150) found that the use of affect in EPAs has a significant and moderate impact (r = 0?.35) on students' learning motivation and a relatively smaller impact on knowledge retention (r =?0.29) and knowledge transfer (r?= 0?.26). The effect sizes of the three outcome variables in this review are stronger compared with previous meta-analysis on the impact of the embodiment level of EPAs, indicating that the use of affect is more desirable and effective in CLEs. Finally, implications for both designers and researchers of using affective EPAs in CLEs are discussed.

When solving many regression problems, there exist a large number of input features. However, not all features are relevant for current regression, and sometimes, including irrelevant features may deteriorate the learning performance. Therefore, it is essential to select the most relevant features, especially for high‐dimensional regression. Feature selection is an effective way to solve this problem. It tries to represent original data by extracting relevant features that contain useful information. In this paper, aiming to effectively select useful features in least squares support vector regression (LSSVR), we propose a novel sparse LSSVR based on Lp‐norm (SLSSVR), 0