A Markov chain collaborative filtering model for course enrollment recommendations
Autor: | Zhao Zhenge, Elham S. Khorasani, John Champaign |
---|---|
Rok vydání: | 2016 |
Předmět: |
Computer science
Stochastic modelling Maximum likelihood Markov process Context (language use) 02 engineering and technology Recommender system Markov model Machine learning computer.software_genre Data modeling symbols.namesake ComputingMilieux_COMPUTERSANDEDUCATION 0202 electrical engineering electronic engineering information engineering Collaborative filtering Curriculum Markov chain business.industry 05 social sciences symbols 020201 artificial intelligence & image processing Artificial intelligence 0509 other social sciences 050904 information & library sciences business computer |
Zdroj: | IEEE BigData |
DOI: | 10.1109/bigdata.2016.7841011 |
Popis: | In this paper we detail our initial approach and early results in examining the efficacy of a Markovbased stochastic model to course enrollment recommendations. We outline a Markov-based collaborative filtering model to recommend courses to students at each semester based on the sequence of courses they have taken in the previous semesters. The proposed model is based on the enrollment data and no prior knowledge of the institution, course prerequisites, curriculum or degree requirement is assumed. Using enrollment data from a research university in Canada, we evaluate and compare the Markov model with traditional collaborative filtering approaches for course recommendation. Our initial results show that the Markov-based model significantly outperforms traditional collaborative filtering models when applied to course enrollment recommendation. |
Databáze: | OpenAIRE |
Externí odkaz: |