Projects

Students will be expected to work in teams of up to 4 members on course projects. Students are welcome to (and usually do) come up with their own projects; alternately they may also choose from a list of projects that are proposed by various collaborators. The list of proposed projects will be presented in the first week of February. We expect at least some of these projects to lead to papers, or to contribute to other applications or projects.

Students are expected to submit a one-page abstract by February 28th 2015. The project abstarcat must specify:

• A list of team members (teams of 2-4 expected)
• Main idea and motivation for the project
• Database to be used
• Outline of experiments expected to run
• Methods and algorithms planned to be explored

The projects are expected to involve some signal analysis and employ and study machine learning algorithms using a large enough, non-trivial dataset.

Some interesting datasets to explore :

• IMDB dataset
• Several challenges and interesting datasets on speech and audio can be found at INTERSPEECH 2015
• Text retrieval datasets TREC datasets
• Wikipedia dump
• Amazon AWS public datasets
• Stanford social networks datasets
• Twitter public streams
• Bioinformatics datsets
• UCI repositories
• Whisper data CHAINS audio databases
• German emotional dataset EMODB
• Neuroscience datasets
• Images and other datasets Deep learning datasets
• Kaggle Datasets
• Other datasets
• Doppler and financial datasets - contact Dr. Bhiksha Raj