Corpus ID: 9248221

Collaborative Filtering Applied to Educational Data Mining

  title={Collaborative Filtering Applied to Educational Data Mining},
  author={Michael Jahrer},
We present our overall third ranking solution for the KDD Cup 2010 on educational data mining. The goal of the competition was to predict a student’s ability to answer questions correctly, based on historic results. In our approach we use an ensemble of collaborative filtering techniques, as used in the field of recommender systems and adopt them to fit the needs of the competition. The ensemble of predictions is finally blended, using a neural network. 

Figures and Tables from this paper

Combining Different Classifiers in Educational Data Mining
This paper uses different classification algorithms, such as KNN, SVD and logistic regression for all the data to generate different results, and then combines these to obtain the final result. Expand
The 2010 KDD Cup Competition Dataset: Engaging the machine learning community in predictive learning analytics
This dataset, consisting of over 9GB of student data, is believed to have been the most often cited from an educational technology platform five years later. Expand
Comparison of Collaborative-Filtering Techniques for Small-Scale Student Performance Prediction Task
This work investigates the usability of CF techniques in student performance prediction for small universities or courses with only a few of students and shows that in such cases the predictive accuracy of these models is not so good and they need to utilize more information about students or tasks. Expand
Applying Representation Learning for Educational Data Mining
Educational Data Mining is an area of growing interest, given the increase of available data and generalization of online learning environments. In this paper we present a first approach toExpand
Multi-relational Factorization Models for Student Modeling in Intelligent Tutoring Systems
This work proposes using multi-relational factorization approach, which has been successfully applied in recommender systems area, for student modeling in the Intelligent Tutoring Systems and shows that the proposed approach can improve the prediction results and could be used for student modeled. Expand
Using factorization machines for student modeling
This work proposes using Factorization Machines which combine the advantages of Support Vector Machines with factorization models for the problem of PSP, and shows that this approach can improve the prediction results over the standard matrix factorization. Expand
Performance Prediction for Students: A Multi-Strategy Approach
The experimental results have demonstrated that unlike the PSP in e-Learning systems, the regression- based approach should give better performance than the recommender system-based approach. Expand
Using Knowledge Components for Collaborative Filtering in Adaptive Tutoring Systems
This paper based their similarity measure on the performance on questions that require the same “knowledge components” (or skills) of similar students, which increases the amount of users with whom it is possible to compare performance, and increases the likelihood of finding similar students. Expand
EduRank: A Collaborative Filtering Approach to Personalization in E-learning
A new algorithm for personliazing educational content to students that combines collaborative filtering algorithms with social choice theory is presented that infers a difficulty ranking directly over the questions for a target student, rather than ordering them according to predicted performance, which is prone to error. Expand
Factorization Techniques for Predicting Student Performance
Recommender systems are widely used in many areas, especially in ecommerce. Recently, they are also applied in e-learning for recommending learning objects (e.g. papers) to students. This chapterExpand


Combining predictions for accurate recommender systems
It is shown that linearly combining a set of CF algorithms increases the accuracy and outperforms any single CF algorithm, and how to use ensemble methods for blending predictors in order to outperform a single blending algorithm. Expand
Improving regularized singular value decomposition for collaborative filtering
Different efficient collaborative filtering techniques and a framework for combining them to obtain a good prediction are described, predicting users’ preferences for movies with error rate 7.04% better on the Netflix Prize dataset than the reference algorithm Netflix Cinematch. Expand
Restricted Boltzmann machines for collaborative filtering
This paper shows how a class of two-layer undirected graphical models, called Restricted Boltzmann Machines (RBM's), can be used to model tabular data, such as user's ratings of movies, and demonstrates that RBM's can be successfully applied to the Netflix data set. Expand
GroupLens: an open architecture for collaborative filtering of netnews
GroupLens is a system for collaborative filtering of netnews, to help people find articles they will like in the huge stream of available articles, and protect their privacy by entering ratings under pseudonyms, without reducing the effectiveness of the score prediction. Expand
The BigChaos Solution to the Netflix Prize 2008
The team “BellKor in BigChaos” is a combined team of team BellKor and BigChaos. The solution with a RMSE of 0.8616 is created by a linear blend of the results from both teams. In the following paperExpand
ELF - Ensemble Learning Framework. An open source C++ framework for supervised learning
  • 2010
Collaborative filtering with temporal dynamics
Two leading collaborative filtering recommendation approaches are revamped and a more sensitive approach is required, which can make better distinctions between transient effects and long-term patterns. Expand
ELF -Ensemble Learning Framework. An open source C++ framework for supervised learning.
  • ELF -Ensemble Learning Framework. An open source C++ framework for supervised learning.
  • 2010
Technical report, commendo research & consulting, 2008
  • 11
  • 2008