Hsinchu, Taiwan

Minghsin University of Science and Technology is a private university in Xinfeng Township, Hsinchu County, Taiwan. Wikipedia.


Time filter

Source Type

Hsu C.-M.,Minghsin University of Science and Technology
Expert Systems with Applications | Year: 2011

Stock price prediction is a very important financial topic, and is considered a challenging task and worthy of the considerable attention received from both researchers and practitioners. Stock price series have properties of high volatility, complexity, dynamics and turbulence, thus the implicit relationship between the stock price and predictors is quite dynamic. Hence, it is difficult to tackle the stock price prediction problems effectively by using only single soft computing technique. This study hybridizes a self-organizing map (SOM) neural network and genetic programming (GP) to develop an integrated procedure, namely, the SOM-GP procedure, in order to resolve problems inherent in stock price predictions. The SOM neural network is utilized to divide the sample data into several clusters, in such a manner that the objects within each cluster possess similar properties to each other, but differ from the objects in other clusters. The GP technique is applied to construct a mathematical prediction model that describes the functional relationship between technical indicators and the closing price of each cluster formed in the SOM neural network. The feasibility and effectiveness of the proposed hybrid SOM-GP prediction procedure are demonstrated through experiments aimed at predicting the finance and insurance sub-index of TAIEX (Taiwan stock exchange capitalization weighted stock index). Experimental results show that the proposed SOM-GP prediction procedure can be considered a feasible and effective tool for stock price predictions, as based on the overall prediction performance indices. Furthermore, it is found that the frequent and alternating rise and fall, as well as the range of daily closing prices during the period, significantly increase the difficulties of predicting. © 2011 Elsevier Ltd. All rights reserved.


Hsu C.-M.,Minghsin University of Science and Technology
International Journal of Systems Science | Year: 2012

The lighting performance of an LED (light-emitting diode) flash is significantly influenced by the geometric form of a reflector. Previously, design engineers have usually determined the geometric design of a reflector according to the principles of optics and their own experience. Some real reflectors have then been created to verify the feasibility and performance of a certain geometric design. This, however, is a costly and time-consuming procedure. Furthermore, the geometric design of a reflector cannot be proven to be actually optimal. This study proposes a systematic approach based on genetic programming (GP) and ant colony optimisation (ACO), called the GP-ACO procedure, to improve the geometric design of a reflector. A case study is used to demonstrate the feasibility and effectiveness of the proposed optimisation procedure. The results show that all the crucial quality characteristics of an LED flash fulfil the required specifications; thus, the optimal geometric parameter settings of the reflector obtained can be directly applied to mass production. Consequently, the proposed GP-ACO procedure can be considered an effective method for resolving general multi-response parameter design problems. © 2012 Taylor & Francis.


Jaing C.-C.,Minghsin University of Science and Technology
Applied Optics | Year: 2011

This study elucidates the effects of columnar angles and deposition angles on the thermal expansion coefficients and intrinsic stress behaviors of MgF 2 films with columnar microstructures. The behaviors associated with temperature-dependent stresses in the MgF2 films are measured using a phase-shifting Twyman-Green interferometer with a heating stage and the application of a phase reduction algorithm. The thermal expansion coefficients of MgF2 films at various columnar angles were larger than those of glass substrates. The intrinsic stress in the MgF2 films with columnar microstructures was compressive, while the thermal stress was tensile. The thermal expansion coefficients of MgF2 films with columnar microstructures and their intrinsic stress evidently depended on the deposition angle and the columnar angle. © 2010 Optical Society of America.


Hsu C.-M.,Minghsin University of Science and Technology
International Journal of Systems Science | Year: 2014

Portfolio optimisation is an important issue in the field of investment/financial decision-making and has received considerable attention from both researchers and practitioners. However, besides portfolio optimisation, a complete investment procedure should also include the selection of profitable investment targets and determine the optimal timing for buying/selling the investment targets. In this study, an integrated procedure using data envelopment analysis (DEA), artificial bee colony (ABC) and genetic programming (GP) is proposed to resolve a portfolio optimisation problem. The proposed procedure is evaluated through a case study on investing in stocks in the semiconductor sub-section of the Taiwan stock market for 4 years. The potential average 6-month return on investment of 9.31% from 1 November 2007 to 31 October 2011 indicates that the proposed procedure can be considered a feasible and effective tool for making outstanding investment plans, and thus making profits in the Taiwan stock market. Moreover, it is a strategy that can help investors to make profits even when the overall stock market suffers a loss. © 2013 Taylor & Francis.


Tsai P.S.M.,Minghsin University of Science and Technology
Expert Systems with Applications | Year: 2010

Association rule mining is an important research topic in the data mining community. There are two difficulties occurring in mining association rules. First, the user must specify a minimum support for mining. Typically it may require tuning the value of the minimum support many times before a set of useful association rules could be obtained. However, it is not easy for the user to find an appropriate minimum support. Secondly, there are usually a lot of frequent itemsets generated in the mining result. It will result in the generation of a large number of association rules, giving rise to difficulties of applications. In this paper, we consider mining top-k frequent closed itemsets from data streams using a sliding window technique. A single pass algorithm, called FCI-max, is developed for the generation of top-k frequent closed itemsets of length no more than max-l. Our method can efficiently resolve the mentioned two difficulties in association rule mining, which promotes the usability of the mining result in practice. © 2010 Elsevier Ltd. All rights reserved.


Li T.,Minghsin University of Science and Technology
International Journal of Advanced Manufacturing Technology | Year: 2010

Innovative design in the development of new product and process has become the core value in most business establishments. These innovative designs are often associated with the long-established trade-off compromises or conflicting performance parameters where speed and reliability or quality and cost are readily acknowledged. The rate of change in technology and the commercial environment suggests that the opportunity for innovative design is accelerating and systematic support for innovation process is needed. This study combines the Russian theory of inventive problem solving (TRIZ) and the analytical hierarchy process (AHP) for designing the automated manufacturing systems. This study applied the contradiction matrix table, 40 innovative principles, and 39 engineering parameters to compromise the trade-off between design contradictions and engineering parameters. The design engineers can acquire more feasible solutions and inspiration through the proposed approach. However, due to vagueness and uncertainty in the decision maker's judgment, an AHP is employed as a decision support tool that can adequately represent qualitative and subjective assessments under the multiple criteria decision making environment. Moreover, the proposed approach can help decision makers facilitate the selection and evaluation of innovative designs in the presence of intangible attributes and uncertainty. In short, the objectives of this research are to use TRIZ to propose the automated design alternatives under the innovative design consideration and to use an AHP to evaluate and select the best feasible alternative under multiple criteria. A case study of designing automated connector assembly line has been used to demonstrate the applicability of the proposed approach. © 2009 Springer-Verlag London Limited.


Hsu C.-M.,Minghsin University of Science and Technology
Proceedings - 2011 6th IEEE Joint International Information Technology and Artificial Intelligence Conference, ITAIC 2011 | Year: 2011

Stock/futures price forecasting is an important financial topic for individual investors, stock fund managers and financial analysts, and is currently receiving considerable attention from both researchers and practitioners. However, the inherent characteristics of stock/futures prices, namely, high volatility, complexity, and turbulence, make forecasting a challenging endeavor. In the past, various approaches have been proposed to deal with the problems of stock/futures price forecasting, that are difficult to resolve by using only a single soft computing technique. In this study, a systematic procedure based on a backpropagation (BP) neural network and a feature selection technique is proposed to tackle stock/futures price forecasting problems with the use of technical indicators. The feasibility and effectiveness of this procedure are evaluated through a case study on forecasting the closing prices of Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) futures of the spot month. Experimental results show that the proposed forecasting procedure is a feasible and effective tool for forecasting stock/futures prices. Furthermore, the statistical hypothesis testing indicates that the forecasting performance of a BP model with feature selection is better than that obtained through a simple BP model. © 2011 IEEE.


Hsu C.-M.,Minghsin University of Science and Technology
Neural Computing and Applications | Year: 2013

Stock/futures price forecasting is an important financial topic for individual investors, stock fund managers, and financial analysts and is currently receiving considerable attention from both researchers and practitioners. However, the inherent characteristics of stock/futures prices, namely, high volatility, complexity, and turbulence, make forecasting a challenging endeavor. In the past, various approaches have been proposed to deal with the problems of stock/futures price forecasting that are difficult to resolve by using only a single soft computing technique. In this study, a hybrid procedure based on a backpropagation (BP) neural network, a feature selection technique, and genetic programming (GP) is proposed to tackle stock/futures price forecasting problems with the use of technical indicators. The feasibility and effectiveness of this procedure are evaluated through a case study on forecasting the closing prices of Taiwan Stock Exchange Capitalization Weighted Stock Index (TAIEX) futures of the spot month. Experimental results show that the proposed forecasting procedure is a feasible and effective tool for improving the performance of stock/futures price forecasting. Furthermore, the most important technical indicators can be determined by applying a feature selection method based on the proposed simulation technique, or solely on the preliminary GP forecast model. © 2011 Springer-Verlag London Limited.


Wang S.-S.,Minghsin University of Science and Technology | Lin Y.-S.,Minghsin University of Science and Technology
Computer Communications | Year: 2013

Vehicular ad hoc networks (VANETs) are a promising architecture for vehicle-to-vehicle communications in the transportation field. However, the frequent topology changes in VANETs create many challenges to data delivery because the vehicle velocity varies with time. Thus, designing an efficient routing protocol for stable and reliable communication is essential. Existing studies show that clustering is an elegant approach to efficient routing in a mobile environment. In particular, the passive clustering (PC) mechanism has been validated as a more efficient approach compared to traditional clustering mechanisms. However, the PC mechanism was primarily designed for mobile ad hoc networks (MANETs), and may be unsuitable for constructing a cluster structure in VANETs because it does not account for vehicle behavior and link quality. In this paper, we propose a passive clustering aided routing protocol, named PassCAR, to enhance routing performance in the one-way multi-lane highway scenario. The main goal of PassCAR is to determine suitable participants for constructing a stable and reliable cluster structure during the route discovery phase. Each candidate node self-determines its own priority to compete for a participant using the proposed multi-metric election strategy based on metrics such as node degree, expected transmission count, and link lifetime. Simulation results show that, compared with the original PC mechanism, PassCAR not only increases the successful probability of route discovery, but also selects more suitable nodes to participate in the created cluster structure. This well-constructed cluster structure significantly improves the packet delivery ratio and achieves a higher network throughput due to its preference for reliable, stable, and durable routing paths. © 2012 Elsevier B.V. All rights reserved.


Wang S.-S.,Minghsin University of Science and Technology | Chen Z.-P.,Minghsin University of Science and Technology
IEEE Sensors Journal | Year: 2013

In wireless sensor networks, nodes in the area of interest must report sensing readings to the sink, and this report always satisfies the report frequency required by the sink. This paper proposes a link-aware clustering mechanism, called LCM, to determine an energy-efficient and reliable routing path. The LCM primarily considers node status and link condition, and uses a novel clustering metric called the predicted transmission count (PTX), to evaluate the qualification of nodes for clusterheads and gateways to construct clusters. Each clusterhead or gateway candidate depends on the PTX to derive its priority, and the candidate with the highest priority becomes the clusterhead or gateway. Simulation results validate that the proposed LCM significantly outperforms the clustering mechanisms using random selection and by considering only link quality and residual energy in the packet delivery ratio, energy consumption, and delivery latency. © 2001-2012 IEEE.

Loading Minghsin University of Science and Technology collaborators
Loading Minghsin University of Science and Technology collaborators