Pei-Chann Chang

A TSK type fuzzy rule based system for stock price prediction

In this paper, a Takagi-Sugeno-Kang (TSK) type Fuzzy Rule Based System is developed for stock price prediction. The TSK fuzzy model applies the technical index as the input variables and the consequent part is a linear combination of the input variables. The fuzzy rule based model is tested on the Taiwan Electronic Shares from the Taiwan Stock Exchange (TSE). Through the intensive experimental tests, the model has successfully forecasted the price variation for stocks from different sectors with accuracy close to 97.6% in TSE index and 98.08% in MediaTek. The results are very encouraging and can be implemented in a real-time trading system for stock price prediction during the trading period.

Fuzzy Delphi and back-propagation model for sales forecasting in PCB industry

Reliable prediction of sales can improve the quality of business strategy. In this research, fuzzy logic and artificial neural network are integrated into the fuzzy back-propagation network (FBPN) for sales forecasting in Printed Circuit Board (PCB) industry. The fuzzy back propagation network is constructed to incorporate production-control expert judgments in enhancing the models performance. Parameters chosen as inputs to the FBPN are no longer considered as of equal importance, but some sales managers and production control experts are requested to express their opinions about the importance of each input parameter in predicting the sales with linguistic terms, which can be converted into pre-specified fuzzy numbers. The proposed system is evaluated through the real world data provided by a printed circuit board company and experimental results indicate that the Fuzzy back-propagation approach outperforms other three different forecasting models in MAPE measures.

A neural network with a case based dynamic window for stock trading prediction

Stock forecasting involves complex interactions between market-influencing factors and unknown random processes. In this study, an integrated system, CBDWNN by combining dynamic time windows, case based reasoning (CBR), and neural network for stock trading prediction is developed and it includes three different stages: (1) screening out potential stocks and the important influential factors; (2) using back propagation network (BPN) to predict the buy/sell points (wave peak and wave trough) of stock price and (3) adopting case based dynamic window (CBDW) to further improve the forecasting results from BPN. The system developed in this research is a first attempt in the literature to predict the sell/buy decision points instead of stock price itself. The empirical results show that the CBDW can assist the BPN to reduce the false alarm of buying or selling decisions. Nine different stocks with different trends, i.e., upward, downward and steady, are studied and one individual stock (AUO) will be studied as case example. The rates of return for upward, steady, and downward trend stocks are higher than 93.57%, 37.75%, and 46.62%, respectively. These results are all very promising and better than using CBR or BPN alone.

A hybrid model combining case-based reasoning and fuzzy decision tree for medical data classification

In this research, a hybrid model is developed by integrating a case-based data clustering method and a fuzzy decision tree for medical data classification. Two datasets from UCI Machine Learning Repository, i.e., liver disorders dataset and Breast Cancer Wisconsin (Diagnosis), are employed for benchmark test. Initially a case-based clustering method is applied to preprocess the dataset thus a more homogeneous data within each cluster will be attainted. A fuzzy decision tree is then applied to the data in each cluster and genetic algorithms (GAs) are further applied to construct a decision-making system based on the selected features and diseases identified. Finally, a set of fuzzy decision rules is generated for each cluster. As a result, the FDT model can accurately react to the test data by the inductions derived from the case-based fuzzy decision tree. The average forecasting accuracy for breast cancer of CBFDT model is 98.4% and for liver disorders is 81.6%. The accuracy of the hybrid model is the highest among those models compared. The hybrid model can produce accurate but also comprehensible decision rules that could potentially help medical doctors to extract effective conclusions in medical diagnosis.

Combining SOM and fuzzy rule base for flow time prediction in semiconductor manufacturing factory

This paper presents a novel approach by combining SOM and fuzzy rule base for flow time prediction in semiconductor manufacturing factory. Flow time of a new order is highly related to the shop floor status; however, the semiconductor manufacturing processes are highly complicated and involve more than hundred of production steps. There is no governing function identified so far among the flow time of a new order and these shop flow status. Therefore, a simulation model which mimics the production process of a real wafer fab located in Hsin-Chu Science-based Park of Taiwan is built and flow time and related shop floor status are collected and fed into the SOM for classification. Then, corresponding fuzzy rule base is selected and applied for flow time prediction. Genetic process is further applied to fine-tune the composition of the rule base. Finally, using the simulated data, the effectiveness of the proposed method is shown by comparing with other approaches.

Evolving fuzzy rules for due-date assignment problem in semiconductor manufacturing factory

This paper presents a fuzzy modeling method proposed by Wang and Mendel for generation of fuzzy rules using data generated from a simulated model that is built from a real factory located in Hsin-Chu science-based park of Taiwan, R.O.C. The fuzzy modeling method is further evolved by a genetic algorithm for due-date assignment problem in manufacturing. By using simulated data, the effectiveness of the proposed method is shown and compared with two other soft computing techniques: multi-layer perceptron neural networks and case-based reasoning. The comparative results indicate that the proposed method is consistently superior to the other two methods.

Sub-population genetic algorithm with mining gene structures for multiobjective flowshop scheduling problems

According to previous research of Chang et al. [Chang, P. C., Chen, S. H., & Lin, K. L. (2005b). Two phase sub-population genetic algorithm for parallel machine scheduling problem. Expert Systems with Applications, 29(3), 705-712], the sub-population genetic algorithm (SPGA) is effective in solving multiobjective scheduling problems. Based on the pioneer efforts, this research proposes a mining gene structure technique integrated with the SPGA. The mining problem of elite chromosomes is formulated as a linear assignment problem and a greedy heuristic using threshold to eliminate redundant information. As a result, artificial chromosomes are created according to this gene mining procedure and these artificial chromosomes will be reintroduced into the evolution process to improve the efficiency and solution quality of the procedure. In addition, to further increase the quality of the artificial chromosome, a dynamic threshold procedure is developed and the flowshop scheduling problems are applied as a benchmark problem for testing the developed algorithm. Extensive tests in the flow-shop scheduling problem show that the proposed approach can improve the performance of SPGA significantly.

Evolving and clustering fuzzy decision tree for financial time series data forecasting

Stock price predictions have always been a subject of interest for investors and professional analysts. Nevertheless, determining the best time to buy or sell a stock remains very difficult because there are many factors that may influence the stock prices. This paper establishes a novel financial time series-forecasting model by evolving and clustering fuzzy decision tree for stocks in Taiwan Stock Exchange Corporation (TSEC). This forecasting model integrates a data clustering technique, a fuzzy decision tree (FDT), and genetic algorithms (GA) to construct a decision-making system based on historical data and technical indexes. The set of historical data is divided into k sub-clusters by adopting K-means algorithm. GA is then applied to evolve the number of fuzzy terms for each input index in FDT so the forecasting accuracy of the model can be further improved. A different forecasting model will be generated for each sub-cluster. In other words, the number of fuzzy terms in each sub-cluster will be different. Hit rate is applied as a performance measure and the proposed GAFDT model has the best performance of 82% average hit rate when compared with other approaches on various stocks in TSEC.

Two-phase sub population genetic algorithm for parallel machine-scheduling problem

This paper introduces a two@?phase sub population genetic algorithm to solve the parallel machine-scheduling problem. In the first phase, the population will be decomposed into many sub-populations and each sub-population is designed for a scalar multi-objective. Sub-population is a new approach for solving multi-objective problems by fixing each sub-population for a pre-determined criterion. In the second phase, non-dominant solutions will be combined after the first phase and all sub-population will be unified as one big population. Not only the algorithm merges sub-populations but the external memory of Pareto solution is also merged and updated. Then, one unified population with each chromosome search for a specific weighted objective during the next evolution process. The two phase sub-population genetic algorithm is applied to solve the parallel machine-scheduling problems in testing of the efficiency and efficacy. Experimental results are reported and the superiority of this approach is discussed.

A hybrid electromagnetism-like algorithm for single machine scheduling problem

Electromagnetism-like algorithm (EM) is a population-based meta-heuristic which has been proposed to solve continuous problems effectively. In this paper, we present a new meta-heuristic that applies the EM methodology to the single machine scheduling problem. To the best of our knowledge, there are only few researches in solving the combinatorial optimization problem (COP) by EM. This research attempts to employ the random-key concept combining with genetic operators in the hybrid algorithm to obtain the best/optimal schedule for the single machine problems. This new approach attempts to achieve the convergence and diversity effects when it is iteratively applied to solve the problem. This hybrid algorithm is tested on a set of standard test problems available in the literature. The computational results show that this hybrid algorithm performs better than the standard genetic algorithm.