Titlebook: Intelligent Data Engineering and Automated Learning -- IDEAL 2013; 14th International C Hujun Yin,Ke Tang,Xin Yao Conference proceedings 20

FECT

Continuous Motion Recognition Using Multiple Time Constant Recurrent Neural Network with a Deep Net the MTRNN according to automatically detected situation change. We apply our approach to 20 motion skeleton units, which is obtained by KINECT, to construct three kinds of human motion sequences. The results show that the proposed method is able to recognize various motions using proper initial state information in a real-time procedure.

仔細(xì)閱讀

CHOIR

Web Service Evaluation Method Based on Time-aware Collaborative Filtering,which computes the time-aware effectiveness of the past data and improves the process of the neighbor users finding. The real-world web service Qos dataset is conducted and the experimental results show that our approach achieves the lower prediction errors than other approaches.

半球

激勵

An Initialized ACO for the VRPTW,e . ., . can reach or reduce . . in 8 out of 18 instances of the Solomon benchmark set, at the cost of increasing travel distance slightly. . can effectively decrease the number of vehicles, travel distance and runtime compared with an ACO without initialization.