Knowledge Discovery with Genetic Programming for Providing Feedback to Courseware Authors

Data mining in course management systems: Moodle case study and tutorial

“…And some examples of public domain mining tools are Weka (Weka, 2007) and Keel (Keel, 2007). There are also some specific educational data mining tools such as the Mining tool (Zaïane & Luo, 2001) for association and pattern mining, MultiStar (Silva & Vieira, 2002) for association and classification, Tool (Chang, Hung, & Shih, 2003) for performing a quantitative analysis based on students' learning performance, EPRules (Romero, Ventura, & Bra, 2004) for association, KAON (Tane, Schmitz, & Stumme, 2004) for clustering and text mining, Synergo/ColAT (Avouris, Komis, Fiotakis, Margaritis, & Voyiatzaki, 2005) for statistics and visualization, GISMO (Mazza & Milani, 2005) for visualization, Listen tool (Mostow et al, 2005) for visualization and browsing, TADAEd (Merceron & Yacef, 2005) for visualizing and mining, O3R (Becker, Vanzin, & Ruiz, 2005) for sequential pattern mining, MINEL (Bellaachia, Vommina, & Berrada, 2006) for mining learning paths, CIECoF (García, Romero, Ventura, & Castro, 2006) for association rule mining, Simulog (Bravo & Ortigosa, 2006) for looking for unexpected behavioural patterns, and Sequential Mining tool (Romero et al, in press) for pattern mining.…”

“…The support of the rule is the percentage of transactions that contains both antecedent and consequence in all transactions in the database. Association rule mining has been applied to web-based education systems for: building recommender agents that could recommend on-line learning activities or shortcuts (Zaïane, 2002); diagnosing student learning problems and offer students advice (Hwang, Hsiao, & Tseng, 2003); guiding the learner's activities automatically and recommending learning materials (Lu, 2004); determining which learning materials are the most suitable to be recommended to the user (Markellou, Mousourouli, Spiros, & Tsakalidis, 2005); identifying attributes characterizing patterns of performance disparity between various groups of students (Minaei-Bidgoli, Tan, & Punch, 2004); discovering interesting relationships from student's usage information in order to provide feedback to course author (Romero et al, 2004); finding out relationships in learners' behaviour patterns (Yu, Own, & Lin, 2001); finding students' mistakes that often accompany each other (Merceron & Yacef, 2004); guiding the search for best fitting transfer models of student learning (Freyberger, Heffernan, & Ruiz, 2004); and optimizing the content of the elearning portal by determining what most interests the user (Ramli, 2005).…”

Adaptive Web-Based Educational Hypermedia

“…People are beginning to get used to (and accept) such unstable link anchors, because of "top stories", "joke of the day" and other links that routinely lead to information that changes frequently. In a learning context, in [12,13,14], the use of an adaptive course (with adaptive tests) is described in which the outcome of a test would automatically lead to the start of a chapter at a beginner's, intermediate or advanced level. It is almost like the automatic progression to higher levels in shootout or adventure games.…”

“…However, in order to perform that adaptation AHA! stores and updates a lot of information in the user model and optionally also maintains a complete log of all the user's actions, thus providing ample opportunity for data mining applications to analyze the users' behavior and detect potential issues or problems that are experienced by a significant number of learners [13,12]. Figure 3 shows the overall architecture of AHA!.…”

Blended Learning Evaluation In Higher Education Courses

“…There are studies regarding these features such as: to identify interesting and unexpected learning patterns, which in turn may provide decision lines, enabling teachers to more efficiently organize their teaching structure [33]; to provide feedback to the course instructor about how to improve courseware [25]; to analyze the user's access log in improve e-e-learning and to support the analysis of trends [1]; to help the teacher to discover beneficial or detrimental relationships between the use of webbased educational resources and student's learning [6]; to reveal information about university students' enrolment [28]; learning decomposition and logistic regression to compare the impact of different educational interventions on learning [3]; and usage data analysis to improve the effectiveness of the learning process in e-learning systems [19].…”