Browsing by Author "Tseng, WC"

Now showing 1 - 2 of 2

A fast packet classification by using enhanced tuple pruning
(2002)
Wang, PC
;
Chan, CT
;
Tseng, WC
;
Chen, YC
;
資訊工程學系
In the packet classification, the route and resources allocated to a packet are determined by the destination address as well as other header fields of the packet such as source/destination address, TCP and UDP port numbers. It has been demonstrated that performing packet classification on a potentially large number of fields is difficult and has poor worst-case performance. In this work, we proposed an enhanced tuple pruning search algorithm called "Tuple Pruning +" that provides fast two-dimension packet classification. With reasonable extra filters added for Information Marker, only one hash access to the tuples is required. Through experiments, about 8 MB memory is required for 100K-filter database and 20 million packet per second (MPPS) is achievable. The results demonstrate that the proposed algorithm is suitable for high-speed packet classification.
Applying extending classifier system to develop an option-operation suggestion model of intraday trading - An example of Taiwan index option
(2005)
Chen, AP
;
Chen, YC
;
Tseng, WC
;
資訊管理與財務金融系註：原資管所+財金所
;
Department of Information Management and Finance
This novel study developed an option-operation suggestion model by applying integrated artificial intelligence technique, extending learning classifier system (XCS), which incorporates reinforcement machine learning method to the dynamical problems to the behavior finance. Due to the history of Behavior Finance, many researches have found that the shape of stock trend is not following random walk model, but the repeated trading patterns exist which are referred to as investors experiences. Furthermore, some classical researches have been merely adopted traditional artificial intelligence to analyze the result. Those methodologies are not sufficiently to resolve the dynamical problem, such as economical trading behaviors. Therefore, the model has been proposed concerning intraday trading but avoiding the system risk in the short-term position to benefit investors. By dynamic learning ability of XCS and general population features, the output operation suggestions could be obtained as a reference strategy for investors to predict the index option trend. As an example of Taiwan Index option, the results of the accuracy and accumulative profit have been exhibited remarkable outcome, and so as the simulations of short term prediction with 10-minute and 20-minute tick data.