|
|
|
|
|
|
Help | Seminars List | Add Seminar | Edit Seminars | Tips for organisers | RSS | ics Calendar | Search | Send comments about this website to seminar-master@cecs.anu.edu.au
Contact: Michelle.Moravec@anu.edu.au CS PHD MONITORING
Literature review on Multi-labeling and Topic Model with some ideasMr Lan Du (School of Computer Science, CECS)DATE: 2009-09-24 TIME: 09:30:00 - 10:00:00 LOCATION: RSISE Seminar Room, ground floor, building 115, cnr. North and Daley Roads, ANU ABSTRACT: Multi-labeling problems are not uncommon in real-world situation, such as patent classification, and web-page categorization. However, most of existing multi-label classifiers simply assume the predefined classes are independently distributed. The intrinsic correlations and the underlying semantic structure are ignored, which will definitely hinder the classification performance. In our research, we will solve the multi-labeling problems by carefully considering the relationship among classes. Besides, the generative topic model has been receiving more and more attentions in machine learning and data mining. However, fitting unsupervised topic model may not be a good solution because of discarding the label information. Thus, another intention of our research is to extend the current topic model to handle multi-labeling problem, along with exploiting the meta data, such as author.
|