Showing papers by "Shu-Chuan Chu published in 2016"

Privacy preservation through a greedy, distortion-based rule-hiding method

Abstract: Various data mining techniques can be used to discover useful knowledge from large collections of data However, there is a risk of disclosing sensitive information when data is shared between different organizations The balance between legitimate mining needs and the protection of confidential knowledge when data is released or shared must be carefully managed In this paper, we study privacy preservation in association rule mining A new distortion-based method is proposed which hides sensitive rules by removing some items in a database to reduce the support or confidence of sensitive rules below specified thresholds In order to minimize side effects on knowledge, the information on non-sensitive itemsets contained by each transaction is used to sort the supporting transactions The candidates that contain fewer non-sensitive itemsets are selected for modification preferably In order to reduce the distortion degree on data, the minimum number of transactions that need to be modified to conceal a sensitive rule is derived Comparative experiments on real datasets showed that the new method can achieve satisfactory results with fewer side effects and data loss

Optimization Localization in Wireless Sensor Network Based on Multi-Objective Firefly Algorithm

Abstract: Specifying the located nodes of a network plays an important role in the success of many wireless sensor network (WSN) applications including object tracking, detecting, monitoring, etc. In this paper, the multi-objective firefly algorithm for estimating the located nodes in a network is proposed to solve the localization issues in WSNs. Objective functions of estimating the locations of all the nodes of WSN are considered based on two criteria including the distance of nodes, and the geometric topology constraint. The simulation results are compared with other methods in the literature show that the proposed method produces the considerable improvement in terms of localization accuracy and convergence rate.

Dynamic Diversity Population Based Flower Pollination Algorithm for Multimodal Optimization

Abstract: Easy convergence to a local optimum, rather than global optimum could unexpectedly happen in practical multimodal optimization problems due to interference phenomena among physically constrained dimensions. In this paper, an altering strategy for dynamic diversity Flower pollination algorithm (FPA) is proposed for solving the multimodal optimization problems. In this proposed method, the population is divided into several small groups. Agents in these groups are exchanged frequently the evolved fitness information by using their own best historical information and the dynamic switching probability is to provide the diversity of searching process. A set of the benchmark functions is used to test the quality performance of the proposed method. The experimental result of the proposed method shows the better performance in comparison with others methods.

A Compact Flower Pollination Algorithm Optimization

Abstract: A restricted hardware condition is difficult for optimization problems. This paper proposes a novel compact flower pollination algorithm for addressing the class of optimization problems in the restricted hardware condition. In this proposed method, the actual population of tentative solutions is not stored, but a novel probabilistic representation on the population is employed based on the single competition. In the simulation, several problems of numerical optimizations in the benchmark are used to evaluate the accuracy, the computational time and the saving memory of the proposed method. The results compared with the original algorithm and the other algorithms in the literature show that the new proposed method provides the effective way of using a limited memory.

A Joint Attitude Control Method for Small Unmanned Aerial Vehicles Based on Prediction Incremental Backstepping.

Abstract: A new second-order joint attitude control method for Small Unmanned Aerial Vehicles (SUAVs) is presented in this paper. Based on lyapunov theory, a second-order backstepping control law is developed. An incremental control approach called Incremental Backstepping (IB) is used to increase the robust performance of the second-order control system. A new joint incremental backstepping attitude controller is proposed for SUAVs based on the movement equations. A prediction filter is added to enhance the accuracy of the sensors data and eliminate the time delay. The simulation results show the approach presented is valid.

An Alignment-Oriented Segmenting Approach for Optimizing Large Scale Ontology Alignments

Abstract: Addressing ontology heterogeneity problem requires identifying correspondences between the entities across different ontologies, which is commonly known as ontology matching. However, the correct and complete identification of semantic correspondences are difficult to achieve with the larger searching space, thus achieving good efficiency is the major challenge for large scale ontology matching technologies. In this paper, we propose a generic alignment-oriented segmenting approach for optimizing the large scale ontology alignments. In particular, our proposal works in three sequential steps: first, using ontology semantic accuracy measure to determine the source ontology from two ontologies to align, and partitioning the source ontology into a set of disjoint segments through a neighbor based bottom-up partition algorithm to partition; then, utilizing a relevant concept filtering approach to determine the target ontology segments according to each source ontology segments; finally, a Memetic Algorithm (MA) based matching technology is introduced to simultaneously match multiple pairs of ontology segments to obtain final alignments. Four datasets in OAEI 2014, i.e., bibliographic benchmarks, anatomy track, library track and large biomedic track, are used to test our approach. The comparison between our approach and the participants in OAEI 2014 shows that our approach is effective.

Bees and Pollens with Communication Strategy for Optimization

Abstract: Due to interference phenomena among constrained dimensions of the multimodal optimization or complex constrained optimization problems, a local optimum is easily converged, rather than for the expected global optimum. The enhanced diversity agent in optimal algorithms is one of the solutions to this issue. This paper proposes a novel optimization algorithm, namely BPO, based on the communication of the bees in artificial bee colony optimization (ABC), with the pollen in flower pollination algorithm (FPA) to solve the multimodal optimization problems. A new communication strategy for Bees and Pollens is presented to explore and exploit the diversity of the algorithm. Six multimodal benchmark functions are used to verify the convergent behavior, the accuracy, and the speed of the proposed algorithm. Experimental results show that the proposed scheme increases the accuracy more than the original algorithms.

Reversible Watermarking Based on Position Determination and Pixel Pairs

Abstract: Reversible watermarking (RW) based on position determination and three-pixel difference is proposed in this paper. The main idea of this paper is to obtain two difference values depending on a pixel pair. To achieve this purpose, for a pixel pair, its one pixel is predicted by the context of this pair to get its predicted value. By this way, we can obtain a three-pixel set containing one pixel pair and one predicted value, and thus obtain two absolute difference values. For a three-pixel set, no modification is allowed to the predicted value. This predicted value along with all the neighbors surrounding one pair constitute a set used for evaluating the intra-pair correlation. The incorporation of the predicted value in this estimation set helps to largely enhance the estimation accuracy. According to the strength of correlation, we determine if this pair is located into a smooth or complex region. When the desired embedding rate is low, we only modify those pairs located in smooth regions while keeping the others unchanged. Therefore, the PSNR (peak signal to noise ratio) value is largely increased. Experimental results also demonstrate that the proposed method is effective.

Similarity Shape Based on Skeleton Graph Matching.

Abstract: Using graphs to match two feature sets through embedded high-order relations points has many possible applications in criminal justice, security, and high technology. In this paper, we analyze the method of using the random walk framework to establish correspondence between two skeleton graphs and find out matching points between two shapes. The graphs are matched using a skeleton graph with the descriptors of the relationship between the two edges of the end-nodes ranked on an association graph. Through adopting individual jumps with a reweighting scheme, the new proposed approach effectively reflects the one-to-one matching constraints during the random walk process. Experiments on several benchmark data sets show that the proposed approach clearly outperforms existing algorithms, especially in the presence of noise and outliers.

Towards Supporting Feature Location Using Syntactic Analysis

Abstract: Feature location has been recognized as one of the most frequent and important activities undertaken by software developers. Aiming at the issue that most existing feature location approaches based on information retrieval are strongly affected by the quality of the documentation of software artifacts, this paper presents an improved IRbased feature location approach by syntactic analysis. In particular, the proposed approach firstly analyzes how terms have been used in the text through syntactic analysis. Then on this basis, the weight of these terms can be adjusted, and the key terms, which are able to describe the behavioral and semantic characteristics of software repositories, can be further extracted. The results of an experimental study conducted on two open source projects show that characterizing the context of software artifacts considering only key terms can effectively eliminate the text noise in software artifacts, and improve the accuracy of IR-based feature location methods.