Embedded Files
Barnyard | Levels | Workflow | Formats
Our concrete to-do list. Join us and dig in!
Our concrete to-do list. Join us and dig in!
Overview of Red Hen Project Areas
Overview of Red Hen Project Areas
- Audio processing pipeline
- Current state of text tagging
- Machine learning
- Overview of research
- Red Hen corpus data format
- Red Hen data format
- Tweet pipeline
- Video processing pipelines
Red Hen levels of functionality
Red Hen levels of functionality
We invite research proposals and projects at any level in the broader Red Hen infrastructure, including the following levels and goals.
- Data acquisition -- global capture
- US media markets: Los Angeles CA, Cleveland OH
- Europe: Denmark (Denmark, Norway, Sweden), Spain
- Data storage -- distributed
- UCLA
- Case Western Reserve University
- University of Navarra
- Data enhancement -- align captions to speech, transcripts, speech-to-text ...
- Transcripts for CNN, FOX News, and MSNBC
- On-screen text from all shows
- Close analysis and manual coding
- Skilled taggers annotate a training sample
- Analytical methods from the humanities and the social sciences
- Machine learning
- Use the training sample for pattern recognition
- Extend the pattern to a larger dataset
- Data mining research -- joint image / audio / text parsing
- Multiple projects (cf. Audio processing pipeline) (See also http://guidetodatamining.com.)
- Communications research -- cognitive, social
- Multiple projects
- User interface development, publication platforms
- Multiple projects
- Visualization interfaces
- Multiple projects
For some suggestions for possible contributions and collaborations, see What Kind of Red Hen Are You?
Page updated
Google Sites
Report abuse