Active Media Technology: 10th International Conference, AMT by Dominik Slezak, Gerald Schaefer, Son T. Vuong, Yoo-Sung Kim

By Dominik Slezak, Gerald Schaefer, Son T. Vuong, Yoo-Sung Kim

This publication constitutes the court cases of the tenth overseas convention on energetic Media know-how, AMT 2014, held in Warsaw, Poland, in August 2014, held as a part of the 2014 net Intelligence Congress, WIC 2014. The forty seven complete papers awarded have been rigorously reviewed and chosen from quite a few submissions. the themes of those papers surround energetic desktops; interactive structures and functions of AMT-based platforms; energetic media laptop studying and information mining thoughts; AMT for the semantic net; social networks and cognitive foundations.

Show description

Read or Download Active Media Technology: 10th International Conference, AMT 2014, Warsaw, Poland, August 11-14, 2014. Proceedings PDF

Similar media books

Take Control of Media on Your iPad (2nd Edition)

Subscribe to media-savvy writer Jeff Carlson as he is helping you get pleasure from media in your iPad like by no means ahead of. Jeff explains easy and complicated methods of including media on your iPad, and provides you an intensive grounding in Apple's middle media apps—iBooks, iPod, movies, and photographs. He additionally is helping you combine your iPad with an Apple television and appears at a number of methods of utilizing an iPad as a television distant.

Social Media Balanced Scorecard: Erfolgreiche Social Media-Strategien in der Praxis

Die Investitionen in Onlinemarketing steigen kontinuierlich, so auch die Werbung in sozialen Medien. Bisher ist es allerdings schwierig, Reichweite und Wirkung sozialer Medien zu messen und die Wertschöpfung zu ermitteln, die das advertising im Bereich sozialer Medien auf die wesentlichen Unternehmensziele wie Umsatzentwicklung, Konversionsereignisse und Kundenzufriedenheit hat.

Extra info for Active Media Technology: 10th International Conference, AMT 2014, Warsaw, Poland, August 11-14, 2014. Proceedings

Sample text

Node influence and node neighbors). But these 7 nodes cannot satisfy with our requirement of choosing 10 nodes. Thus we apply supplementary strategy to choose another three nodes in order to gain 10 important nodes in the Enron email network: {v41 , v51 , v77 }. In many cases, these selected nodes are the most significant nodes in a network, which will be estimated in Sec. 4. By studying this case, we illustrate that the FU strategy can obtain the most important k nodes from a network through getting intersection elements from existing methods.

We next integrate MARS method with k-NN regression transfer techniques. MARS is an adaptive stepwise method for regression [14], and its weighted risk form is as follows: ntr (i) (i) (i) W eig(xtr ) · (ytr − f (xtr ))2 min i=1 (8) m (i) f (xtr ) (i) βj hj (xtr ) = β0 + j=1 where hj (x) is a constant denoted by C, or a hinge function with the form max(0, x − C) or max(0, C − x), or a product of two or more the above hinge (i) functions. m is the total steps to get optimal performance, W eig(xtr ) is defined (i) as the above, f (xtr ) denotes the prediction value of training instances, and (i) f (xte ) can be defined similarly.

In the Table 1, transfer SVM behaves only a little better than non-transfer SVM. The reason may be that SVM only utilizes support vectors which is a small part of the whole data, and thus clustering to the whole data doesn’t produce big effect. Local Regression Transfer Learning 31 Table 1. 2608 Secondly, we compare our local (k-NN) regression transfer method with the non-transfer method, global transfer method, and KMM transfer method, respectively. We choose MARS method and LWR method as a part of local and global transfer methods.

Download PDF sample

Rated 4.22 of 5 – based on 48 votes