Software defect prediction is important for improving software quality. Defect predictors allow software test engineers to focus on defective modules. Cross-Project Defect Prediction (CPDP) uses data from other companies to build defect predictors. However, outliers may lower prediction accuracy. In this study, we propose a transfer learning based model called VAB-SVM for CPDP robust in handling outliers. Notably, this method deals with the class imbalance problem which may decrease the prediction accuracy. Our proposed method computes similarity weights of the training data based on the test data. Such weights are applied to Boosting algorithm considering the class imbalance. VAB-SVM outperformed the previous research more than 10% and showed a sufficient robustness regardless of the ratio of outliers.
{"title":"Improving Prediction Robustness of VAB-SVM for Cross-Project Defect Prediction","authors":"Duksan Ryu, Okjoo Choi, Jongmoon Baik","doi":"10.1109/CSE.2014.198","DOIUrl":"https://doi.org/10.1109/CSE.2014.198","url":null,"abstract":"Software defect prediction is important for improving software quality. Defect predictors allow software test engineers to focus on defective modules. Cross-Project Defect Prediction (CPDP) uses data from other companies to build defect predictors. However, outliers may lower prediction accuracy. In this study, we propose a transfer learning based model called VAB-SVM for CPDP robust in handling outliers. Notably, this method deals with the class imbalance problem which may decrease the prediction accuracy. Our proposed method computes similarity weights of the training data based on the test data. Such weights are applied to Boosting algorithm considering the class imbalance. VAB-SVM outperformed the previous research more than 10% and showed a sufficient robustness regardless of the ratio of outliers.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125143046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the development of communications and network technologies, video streaming service is becoming increasingly important. However, in the transmission of high compressed video, bit-stream is easily damaged by channel errors which will lead to the decline of the video sequence. Error concealment is an effective approach to reduce the influence in error-prone network and notably improve the video quality. This paper presents a novel error concealment algorithm for temporal video error concealment with H.264/AVC standard. This method exploits both spatial and temporal information in the reference and current frame. To increase the accuracy, the macro block (MB) is divided into 16 4x4 blocks. The reference frame is classified by k-means clustering algorithm. The lost blocks are recovered by the neigh boring blocks which belong to the same class of the lost block. The experimental result shows that our method can achieve a better performance than the existing methods.
{"title":"A New Error Concealment Algorithm for H.264/AVC","authors":"Zhixin Shen, Xingang Liu, Lingyun Lu, Xun Wang","doi":"10.1109/CSE.2014.206","DOIUrl":"https://doi.org/10.1109/CSE.2014.206","url":null,"abstract":"With the development of communications and network technologies, video streaming service is becoming increasingly important. However, in the transmission of high compressed video, bit-stream is easily damaged by channel errors which will lead to the decline of the video sequence. Error concealment is an effective approach to reduce the influence in error-prone network and notably improve the video quality. This paper presents a novel error concealment algorithm for temporal video error concealment with H.264/AVC standard. This method exploits both spatial and temporal information in the reference and current frame. To increase the accuracy, the macro block (MB) is divided into 16 4x4 blocks. The reference frame is classified by k-means clustering algorithm. The lost blocks are recovered by the neigh boring blocks which belong to the same class of the lost block. The experimental result shows that our method can achieve a better performance than the existing methods.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"156 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122954850","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Clarence Goh, J. Devlin, Dennis Deng, A. McDonald, M. Kamarudin
Super DARN operating frequencies from 10MHz-15MHz are susceptible to clutter originating from unwanted altitude backscatter, man-made noise exhibiting a Gaussian distribution and correlated speckle noise from pulses emitted from other similar Super DARN radars operating at the same frequency. The paper focuses on removing man-made clutter and uncorrelated speckle noise noise by prior measurement of received noise signal. Based on noise information, an uncorrelated weighted median filter is proposed and used to remove unwanted clutter.
{"title":"Uncorrelated Weighted Median Filtering for Noise Removal in SuperDARN","authors":"Clarence Goh, J. Devlin, Dennis Deng, A. McDonald, M. Kamarudin","doi":"10.1109/CSE.2014.196","DOIUrl":"https://doi.org/10.1109/CSE.2014.196","url":null,"abstract":"Super DARN operating frequencies from 10MHz-15MHz are susceptible to clutter originating from unwanted altitude backscatter, man-made noise exhibiting a Gaussian distribution and correlated speckle noise from pulses emitted from other similar Super DARN radars operating at the same frequency. The paper focuses on removing man-made clutter and uncorrelated speckle noise noise by prior measurement of received noise signal. Based on noise information, an uncorrelated weighted median filter is proposed and used to remove unwanted clutter.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"82 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131414203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Qi Zhou, Junming Zhang, Jinglin Li, Shangguang Wang
Taxicab demand discovering is one of the most fundamental issues of taxicab services. Most of the regions in one city suffer the demand and supply disequilibrium problem. It causes the difficulty in scheduling taxicabs for taxicab companies. It will be solved by modeling the regional demand of taxicabs by using trajectory data. In this paper, we propose a method to model regional taxicab demand. Firstly, the method uses the KS measures to test the distribution of taxicab service rate. Then, it uses the Parzen window to estimate the probability density function of the rate. We have implemented our method with experiments based on real trajectory data. The results show the effectiveness of our method.
{"title":"Discovering Regional Taxicab Demand Based on Distribution Modeling from Trajectory Data","authors":"Qi Zhou, Junming Zhang, Jinglin Li, Shangguang Wang","doi":"10.1109/CSE.2014.296","DOIUrl":"https://doi.org/10.1109/CSE.2014.296","url":null,"abstract":"Taxicab demand discovering is one of the most fundamental issues of taxicab services. Most of the regions in one city suffer the demand and supply disequilibrium problem. It causes the difficulty in scheduling taxicabs for taxicab companies. It will be solved by modeling the regional demand of taxicabs by using trajectory data. In this paper, we propose a method to model regional taxicab demand. Firstly, the method uses the KS measures to test the distribution of taxicab service rate. Then, it uses the Parzen window to estimate the probability density function of the rate. We have implemented our method with experiments based on real trajectory data. The results show the effectiveness of our method.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121751111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Yanjiang Wei, Rui Wang, Danfeng Zhu, Zhongzhi Luan, D. Qian
Caching is an important technique to improve computer system performance by storing the most recently used data and instructions for main memory. Cache is widely used in modern computer systems and will continue to be an irreplaceable unit to narrow the speed gap between processor and main memory. With the increasing capacity of main memory and the number of processor cores, the cache technology has great development. In this paper, we have some lessons of cache hierarchy changes with the memory technology from experimental methodology. We design a serial of experiments and try to answer some questions about cache designs. Our experiments results indicate that more levels of cache does not necessarily means better performance for all benchmarks, that last level cache miss rate has no direct connection with the system performance, that the average performance of exclusive cache hierarchy is more effective than that of inclusive cache.
{"title":"Lessons from Experimental Methodology of Cache Hierarchy Changes with the Memory Technology","authors":"Yanjiang Wei, Rui Wang, Danfeng Zhu, Zhongzhi Luan, D. Qian","doi":"10.1109/CSE.2014.357","DOIUrl":"https://doi.org/10.1109/CSE.2014.357","url":null,"abstract":"Caching is an important technique to improve computer system performance by storing the most recently used data and instructions for main memory. Cache is widely used in modern computer systems and will continue to be an irreplaceable unit to narrow the speed gap between processor and main memory. With the increasing capacity of main memory and the number of processor cores, the cache technology has great development. In this paper, we have some lessons of cache hierarchy changes with the memory technology from experimental methodology. We design a serial of experiments and try to answer some questions about cache designs. Our experiments results indicate that more levels of cache does not necessarily means better performance for all benchmarks, that last level cache miss rate has no direct connection with the system performance, that the average performance of exclusive cache hierarchy is more effective than that of inclusive cache.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"114 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116498705","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Tao Wu, Leiting Chen, Yayong Guan, Xin Li, Yuxiao Guo
Community structure has many practical applications, and identifying communities could help us to understand and exploit networks more effectively. Generally, real-world networks often have hierarchical structures with communities embedded within other communities. However, there are few effective methods can identify these structures. This paper proposes an algorithm HELPA to detect hierarchical community structures. HELPA is based on coreness centrality to update node's possible community labels, and uses communities as nodes to build super-network. By repeat the procedure, the proposed algorithm can effectively reveal hierarchical communities with different size in various network scales. Moreover, it overcomes the high complexity and poor applicability problem of similar algorithms. To illustrate our methodology, we compare it with many classic methods in real-world networks. Experimental results demonstrate that HELPA achieves excellent performance.
{"title":"LPA Based Hierarchical Community Detection","authors":"Tao Wu, Leiting Chen, Yayong Guan, Xin Li, Yuxiao Guo","doi":"10.1109/CSE.2014.65","DOIUrl":"https://doi.org/10.1109/CSE.2014.65","url":null,"abstract":"Community structure has many practical applications, and identifying communities could help us to understand and exploit networks more effectively. Generally, real-world networks often have hierarchical structures with communities embedded within other communities. However, there are few effective methods can identify these structures. This paper proposes an algorithm HELPA to detect hierarchical community structures. HELPA is based on coreness centrality to update node's possible community labels, and uses communities as nodes to build super-network. By repeat the procedure, the proposed algorithm can effectively reveal hierarchical communities with different size in various network scales. Moreover, it overcomes the high complexity and poor applicability problem of similar algorithms. To illustrate our methodology, we compare it with many classic methods in real-world networks. Experimental results demonstrate that HELPA achieves excellent performance.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132589079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Free and open in nature, open cloud platforms have enjoyed their wide acceptance in academic institutions and enterprises. Fair and insightful comparisons of these available open cloud platforms, however, can still be a challenging task, mainly due to lack of an appropriate evaluation methodology. In this paper, we thus attempt to perform a quantitative evaluation on open IaaS platforms, and such a method of evaluation is based on the standpoint and the perspective of the end software consumers. In essence, based on ISO/IEC 25010:2011 and NIST Cloud Computing Reference Model, we develop a number of rules to score the key features of the open cloud platforms. As an example to showcase this method, the quantitative results and the scores of a few well-known IaaS platforms are reported.
{"title":"Evaluating Open IaaS Cloud Platforms Based upon NIST Cloud Computing Reference Model","authors":"Qing Lei, Yingtao Jiang, Mei Yang","doi":"10.1109/CSE.2014.350","DOIUrl":"https://doi.org/10.1109/CSE.2014.350","url":null,"abstract":"Free and open in nature, open cloud platforms have enjoyed their wide acceptance in academic institutions and enterprises. Fair and insightful comparisons of these available open cloud platforms, however, can still be a challenging task, mainly due to lack of an appropriate evaluation methodology. In this paper, we thus attempt to perform a quantitative evaluation on open IaaS platforms, and such a method of evaluation is based on the standpoint and the perspective of the end software consumers. In essence, based on ISO/IEC 25010:2011 and NIST Cloud Computing Reference Model, we develop a number of rules to score the key features of the open cloud platforms. As an example to showcase this method, the quantitative results and the scores of a few well-known IaaS platforms are reported.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133186794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Research on prediction about analog circuits is rarely conducted, and the only methods are prognosis of few special features extracted from output without guarantee of integrity and rationality of prognostic information, which hence influences prognostic precision. In this paper, a novel prediction method for analog circuits is proposed. In this method, time domain output waveforms in initial state and components degradation state are extracted at first, then white noise estimation is conducted to estimate the change between waveforms according to principles of noise estimation based on Kalman filter so as to obtain more reasonable fault indicators from more complete information, thereafter, a novel degradation tendency model of analog circuits is constructed according to newly obtained fault indicators, model adaption is conducted to the new model through particle filter, and prognostic method is conducted to remaining useful performance of analog circuits. Finally, experimental verification is conducted to the above conclusion.
{"title":"A Novel Prediction Method for Analog Circuits Based on Gaussian White Noise Estimation","authors":"Jingyu Zhou, Shulin Tian, B. Long, Chenglin Yang","doi":"10.1109/CSE.2014.52","DOIUrl":"https://doi.org/10.1109/CSE.2014.52","url":null,"abstract":"Research on prediction about analog circuits is rarely conducted, and the only methods are prognosis of few special features extracted from output without guarantee of integrity and rationality of prognostic information, which hence influences prognostic precision. In this paper, a novel prediction method for analog circuits is proposed. In this method, time domain output waveforms in initial state and components degradation state are extracted at first, then white noise estimation is conducted to estimate the change between waveforms according to principles of noise estimation based on Kalman filter so as to obtain more reasonable fault indicators from more complete information, thereafter, a novel degradation tendency model of analog circuits is constructed according to newly obtained fault indicators, model adaption is conducted to the new model through particle filter, and prognostic method is conducted to remaining useful performance of analog circuits. Finally, experimental verification is conducted to the above conclusion.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115537356","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Multi-antenna blind signal separation (BSS) provides a useful method for co-channel mixed signal processing. But the performance of BSS is limited by the condition number of channel matrix in the presence of noise. To overcome this problem, a new BSS algorithm with the aid of lattice reduction (LR) is proposed for synchronous communication systems. Simulation results show that the proposed algorithm improves the symbol error rate (SER) performance in noise environment while keeping the same complexity level.
{"title":"Lattice Reduction Aided Blind Signal Separation Algorithm","authors":"Kun Zhang, Yourong Lu, Wei Wang","doi":"10.1109/CSE.2014.181","DOIUrl":"https://doi.org/10.1109/CSE.2014.181","url":null,"abstract":"Multi-antenna blind signal separation (BSS) provides a useful method for co-channel mixed signal processing. But the performance of BSS is limited by the condition number of channel matrix in the presence of noise. To overcome this problem, a new BSS algorithm with the aid of lattice reduction (LR) is proposed for synchronous communication systems. Simulation results show that the proposed algorithm improves the symbol error rate (SER) performance in noise environment while keeping the same complexity level.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116874274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The goal of Frequent Item set Mining (FIM) is to find the biggest number of frequently used subsets from a big transaction database. In previous studies, using the advantage of multicore computing, the execution time of an Apriori algorithm was sharply decreased: when the size of a data set was more than TBs and a single host had been unable to afford a large number of operations by using a number of computers connected into a super computer to speed up execution as being the obvious solution. Some parallel Apriori algorithms, based on the MapReduce framework, have been proposed. However, with these algorithms, memory would be quickly exhausted and communication cost would rise sharply. This would greatly reduce execution efficiency. In this paper, we present an improved reformative Apriori algorithm that uses the length of each transaction to determine the size of the maximum merge candidate item sets. By reducing the production of low frequency item sets in Map function, memory exhaustion is ameliorated, greatly improving execution efficiency.
{"title":"IOMRA - A High Efficiency Frequent Itemset Mining Algorithm Based on the MapReduce Computation Model","authors":"Sheng-Hui Liu, Shi-Jia Liu, Shi-Xuan Chen, Kun-Ming Yu","doi":"10.1109/CSE.2014.247","DOIUrl":"https://doi.org/10.1109/CSE.2014.247","url":null,"abstract":"The goal of Frequent Item set Mining (FIM) is to find the biggest number of frequently used subsets from a big transaction database. In previous studies, using the advantage of multicore computing, the execution time of an Apriori algorithm was sharply decreased: when the size of a data set was more than TBs and a single host had been unable to afford a large number of operations by using a number of computers connected into a super computer to speed up execution as being the obvious solution. Some parallel Apriori algorithms, based on the MapReduce framework, have been proposed. However, with these algorithms, memory would be quickly exhausted and communication cost would rise sharply. This would greatly reduce execution efficiency. In this paper, we present an improved reformative Apriori algorithm that uses the length of each transaction to determine the size of the maximum merge candidate item sets. By reducing the production of low frequency item sets in Map function, memory exhaustion is ameliorated, greatly improving execution efficiency.","PeriodicalId":258990,"journal":{"name":"2014 IEEE 17th International Conference on Computational Science and Engineering","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117109840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：