The Cuban Theater Digital Archive (CTDA) was developed to enable long-term preservation of a digital scholarly record of theater performance. The archive is a resource for research, teaching, and learning, with a special focus on Cuban theater produced by Cuban communities in the United States, but includes materials from Cuba, as well as other areas outside of the island. Read more “Cuban Theater Digital Archive”
The University of Miami Libraries has received a two-year $260,000 grant from the Andrew W. Mellon Foundation to support software development that will expand the functionality of the Cuban Theater Digital Archives. Dr. Mitsunori Oghihara, Associate Dean for Digital Library Innovation and Center for Computational Science Director of Data Mining will oversee the development of various software components that will add metadata extraction from the digital media (videos and graphics), add community tags and a search mechanism, and make a publication tool available that allows scholars to more easily access archived materials for research. Read more “Mellon Foundation Grant to Expand UM Libraries’ Cuban Theater Digital Archive”
This presentation was addressed to a specialized audience of people in Data Mining and Machine Learning. The talk provided a theory that showed how clustering stability can be used to choose the correct number of clusters, as well as demonstrate the importance of cluster stability and discussed the use of stability for clustering evaluation. Read more “Clustering Stability: Impossibility and Possibility”
This presentation by Bertrand Salem Clarke (UM) entitled “Dimenson Reduction (Mostly)” was a review of the basics of machine learning with commentary on the major dimension reduction techniques. Read more “Dimension Reduction (Mostly) Presentation by Bertrand Clarke”
This paper “Nextone Player: A Music Recommendation System Based on User Behavior” modeled music listeners listening patterns using a “forgetting” factor—a slow exponential decrease in the freshness of music that was listened to. The effectiveness of the model was confirmed using a simple music jukebox program that makes a recommendation for the next piece to listen to and allows the listener to “skip” to the next song. Read more “Nextone Player: A Music Recommendation System Based on User Behavior”
Dr. Mitsunori Ogihara joined the University of Miami in 2007 as Professor in the Department of Computer Science and as Program Director of the Big Data Analytics & Data Mining Program within the Center for Computational Science. More recently, he was appointed Associate Dean for Digital Library Innovation in the College of Arts and Sciences.