Funded by the European Union. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Education and Culture Executive Agency (EACEA). Neither the European Union nor EACEA can be held responsible for them.
Found 5 applets
| Name of the applet | Description |
|---|---|
| UoP OptLearn Algorithm | The algorithm developed uses the k-means clustering algorithm to categorize the questions by difficulty levels and customize the path according to the student’s needs and abilities. |
| HAW OptLearn Algorithm | This OptLearn algorithm is an improved and advanced version of the UMA implementation, which integrates a recommender system using the Surprise module. This version allows loading and consolidating data from various sources, such as a file with the details of the questions clustering (provided by IPB in this case), into structured DataFrames, and the generation of a rating dataset for analysis and recommendation, with a focus on optimizing the learning experience through personalized question recommendations, while also keeping extensibility in mind. The code has been improved and the main recommender system algorithms have been upgraded to improve performance. |
| UniGenova OptLearn Algorithm | The strategy for question selection is twofold: for the initial question, a random selection is made since there is no available information on past test answers to inform the predictive model. Subsequently, from the second to the last question, a Random Forest model comes into play. This model predicts the probability of the user correctly answering each potential question, and the question with the highest predicted probability becomes the next in line. The underlying goal of this approach is to bolster student motivation by providing questions deemed more manageable, fostering a positive and encouraging learning experience. |
| UniMalaga OptLearn Algorithm | The algorithm aims to develop a recommendation system for the MathE e-learning platform. It employs two main algorithms: SVD++ and KNN with Means. These algorithms estimate the ratings of the questions by the students and recommend the most suitable ones. |
| IPB OptLearn Algorithm | The algorithm proposed aims to make the learning experience adaptive and personalized, considering the individual characteristics and needs of the student and their trajectory on the MathE platform. Thence, the methodology developed encompasses clustering methods to categorize the questions by difficulty levels and graph theory methods to customize the path according to the student’s needs and abilities. |
| UniLjubljana OptLearn Algorithm | This recommender system 1. Takes a list of questions with keywords as an input (.csv or .xlsx file), and generates a concept map. At this stage of development, the concept map is fixed and independent of student history; 2. Takes concept map and student history (may be empty) and sequentially generates recommended questions. During phase 2 (recommender), the algorithm estimates selected learning indicators and maximizes selected ones. The student history is updated for every next question generation. |