Data streams are everywhere, from F1 racing over electricity networks to news feeds. Data stream mining relies on and develops new incremental algorithms that process streams under strict resource limitations. This paper focuses on, as well as extends the methods implemented in MOA, an open source stream mining software suite currently being developed by the Machine Learning group.
Students will be able to: select and apply appropriate algorithms for data stream mining problems; design and implement new algorithms in a data stream mining framework like MOA or similar; compare and evaluate different algorithms/solutions for a problem and summarize in a report.
Three 300 level Computer Science papers, including
COMP321 Practical Data Mining or
COMP316 Artificial Intelligence Techniques and Applications
COMP523 Data Stream Mining
Official Timetable Information
COMP423 students can skip one of the items of assessment as specified for COMP523. If they attempt all items, their worst one will be removed. The final total will be re-scaled to ensure that the optimal score achievable is 100%.
About 10 hours a week on average.
Knowledge Discovery from Data Streams, by Joao Gama
Internal assessment/final examination ratio 1:0
see Moodle page
see Moodle page
Class attendance is expected. The course notes provided are not comprehensive, additional material will be covered in class. You are responsible for all material covered in class.
Follow this link for Academic Integrity information.
Follow this link for information on Performance Impairment.
Student Concerns and Complaints
Follow this link for Student Concerns and Complaints information.
Application for Extension
Follow this link for information on applying for an Extension.
Review of Grade
Follow this link for information on applying for a Review of Grade.
Your attention is drawn to the following regulations and policies, which are published in the University Calendar: