Yungho Leu

A novel hybrid feature selection method for microarray data analysis

Recently, many methods have been proposed for microarray data analysis. One of the challenges for microarray applications is to select a proper number of the most relevant genes for data analysis. In this paper, we propose a novel hybrid method for feature selection in microarray data analysis. This method first uses a genetic algorithm with dynamic parameter setting (GADP) to generate a number of subsets of genes and to rank the genes according to their occurrence frequencies in the gene subsets. Then, this method uses the @g^2-test for homogeneity to select a proper number of the top-ranked genes for data analysis. We use the support vector machine (SVM) to verify the efficiency of the selected genes. Six different microarray datasets are used to compare the performance of the GADP method with the existing methods. The experimental results show that the GADP method is better than the existing methods in terms of the number of selected genes and the prediction accuracy.

A distance-based fuzzy time series model for exchange rates forecasting

Fuzzy time series model has been successfully employed in predicting stock prices and foreign exchange rates. In this paper, we propose a new fuzzy time series model termed as distance-based fuzzy time series (DBFTS) to predict the exchange rate. Unlike the existing fuzzy time series models which require exact match of the fuzzy logic relationships (FLRs), the distance-based fuzzy time series model uses the distance between two FLRs in selecting prediction rules. To predict the exchange rate, a two factors distance-based fuzzy time series model is constructed. The first factor of the model is the exchange rate itself and the second factor comprises many candidate variables affecting the fluctuation of exchange rates. Using the exchange rate data released by the Central Bank of Taiwan, we conducted several experiments on exchange rate forecasting. The experiment results showed that the distance-based fuzzy time series outperformed the random walk model and the artificial neural network model in terms of mean square error.

Effects of local autonomy of global concurrency control in heterogeneous distributed database systems

A heterogeneous distributed database system (HDDBS) is a system which integrates preexisting database systems to support global applications accessing more than one database. The difficulties and general approaches of global concurrency control in HDDBSs are studied. In particular, the author discusses the difficulties of maintaining global serializability in HDDBSs. The need for new strategies is established, and counterexamples to existing algorithms are given. An assumption usually made in addressing multidatabase systems is that the element databases are autonomous. The meaning of autonomy and its effects on designing global concurrency control algorithms are addressed. It is the goal of this study to motivate other researchers to realize the need for new correctness criteria by which concurrency control algorithms can be validated.<<ETX>>

Extending the transaction model to capture more meaning

Since it has been first introduced, the concept of transaction has become one of the fundamental abstractions used in the design and analysis of information systems that provide concurrent access to share information. The basic idea of transactions is to divide application programs into well defined units that provide semantically correct transitions between consistent states of the information system. The fundamental properties of transactions, as defined by Gray [GraB1], namely atornicity, isolation and durability turned out to be very useful in the modeling of real world applications. In particular, most work on concurrency control, commitment and recovery has been performed using transactions as a basic unit of work.

An effective Boolean algorithm for mining association rules in large databases

In this paper, we present in effective Boolean algorithm for mining association rules in large databases of sales transactions. Like the a priori algorithm, the proposed Boolean algorithm mines association rules in two steps. In the first step, logic OR and AND operations are used to compute frequent itemsets. In the second step, logic AND and XOR operations are applied to derive all interesting association rules based on the computed frequent itemsets. By only scanning the database once and avoiding generating candidate itemsets in computing frequent itemsets, the Boolean algorithm gains a significant performance improvement over the a priori algorithm. We propose two efficient implementations of the Boolean algorithm, the bitstream approach and the sparse-matrix approach. Through comprehensive experiments, we show that both the bitstream approach and the sparse-matrix approach outperform the a priori algorithm in all database settings. The sparse-matrix approach in particular shows a very significant performance improvement over the a priori algorithm.

Specification and execution of transactions for advanced database applications

Abstract Autonomous multidatabases, Computer Aided Design (CAD) databases, and Object Oriented Databases have requirements which traditional transactions cannot meet. Users of these advanced applications are more sophisticated than users envisioned for On Line Transaction Processing (OLTP) applications a few decades ago. The need to relax the properties of complex database transactions is urgent. Frameworks for flexible transaction systems are needed. This paper is a step towards this objective. This paper addresses the formalization of a new transaction model called Flex transactions, along with execution control and analysis protocols. The algorithm is formalized through the use of Predicate Transition Nets (PTN) and reachability trees.

Constructing gene regulatory networks from microarray data using GA/PSO with DTW

Recently, many methods have been proposed for constructing gene regulatory networks (GRNs). However, most of the existing methods ignored the time delay regulatory relation in the GRN predictions. In this paper, we propose a hybrid method, termed GA/PSO with DTW, to construct GRNs from microarray datasets. The proposed method uses test of correlation coefficient and the dynamic time warping (DTW) algorithm to determine the existence of a time delay relation between two genes. In addition, it uses the particle swarm optimization (PSO) to find thresholds for discretizing the microarray dataset. Based on the discretized microarray dataset and the predicted types of regulatory relations among genes, the proposed method uses a genetic algorithm to generate a set of candidate GRNs from which the predicted GRN is constructed. Three real-life sub-networks of yeast are used to verify the performance of the proposed method. The experimental results show that the GA/PSO with DTW is better than the other existing methods in terms of predicting sensitivity and specificity.

Efficient data broadcast schemes for mobile computing environments with data missing

Data broadcast is an efficient method for disseminating data items in a mobile computing environment. With the data broadcast method, data items are broadcast periodically according to a predetermined schedule. If a data item is retrieved from a storage device with a nondeterministic access time, the content of the data item may not be ready when it is required in a broadcast cycle. We call this problem the data-missing problem. To handle this problem, we revise the existing data broadcast schemes using two approaches. The reaccess approach requires a mobile client to access a missing data item in the next broadcast cycle, while the add-missing approach allows a mobile client to access a missing data item in an attached missing data segment. We compare the performance of the revised schemes in terms of access time and tuning time. The comparison shows that, except latency_opt, the access time of an add-missing-based revised scheme is shorter than that of its reaccess-based counterpart. On the contrary, the tuning time of the former is longer than that of the latter. For latency_opt, both the access time and the tuning time of its add-missing-based revised scheme are shorter than those of its reaccess-based revised scheme. The comparison also shows that the access times of the revised schemes increases dramatically as the data missing probability increases, while the tuning times of the revised schemes are not so sensitive to the data missing probability.

A hierarchical approach to concurrency control for multidatabases

A multidatabase system is a facility that allows access to data stored in multiple autonomous and possibly heterogeneous database systems. In order to support atomic updates across multiple database systems, a global concurrency control algorithm is required. Hierarchical concurrency control has been proposed as one possible approach for multidatabase systems. However, to apply this approach, some restrictions have to be imposed on the local concurrency control algorithms. In this paper, we identify the restriction. Based on this restriction, we formalize the hierarchical concurrency control approach and prove its correctness. A new global concurrency control algorithm based on this hierarchical approach is also presented.

An effective stock portfolio trading strategy using genetic algorithms and weighted fuzzy time series

Investments in a stock market may incur risk. To reduce the risk of an investment, many portfolio selection methods have been proposed. By buying several stocks together, a portfolio selection method aims at maximizing the return of an investment given a predefined risk level. To build an optimal stock portfolio, one needs to select stocks and decide the proportion of the capital on each stock. Also, it is very important to decide when to buy or sell a stock portfolio. In this paper, we present a genetic algorithm to build stock portfolios. The proposed method comprises a genetic algorithm and a weighted fuzzy time series. The genetic algorithm is used to construct an optimal portfolio while the weighted fuzzy time series is used to predict the return of the portfolio which in turn is used to formulate the fitness function of the genetic algorithm. Furthermore, we propose the periodically checking and stop-loss point policies to decide selling and buying time points of the stock portfolio. The experiments on the stocks of Taiwan 50 show that the proposed method outperforms the Taiwan 50 index and the TAIEX index in terms of the 7-year average return rate.