In many machine learning applications,data are not free,and there is a test cost for each data item. For the economical reason,some existing works try to minimize the test cost and at the same time,preserve a particul...In many machine learning applications,data are not free,and there is a test cost for each data item. For the economical reason,some existing works try to minimize the test cost and at the same time,preserve a particular property of a given decision system. In this paper,we point out that the test cost one can afford is limited in some applications. Hence,one has to sacrifice respective properties to keep the test cost under a budget. To formalize this issue,we define the test cost constraint attribute reduction problem,where the optimization objective is to minimize the conditional information entropy. This problem is an essential generalization of both the test-cost-sensitive attribute reduction problem and the 0-1 knapsack problem,therefore it is more challenging. We propose a heuristic algorithm based on the information gain and test costs to deal with the new problem. The algorithm is tested on four UCI（University of California-Irvine） datasets with various test cost settings. Experimental results indicate the appropriate setting of the only user-specified parameter λ.展开更多
A novel test access mechanism （TAM） architecture with multi test-channel （TC） based on IEEE Standard 1500 is proposed instead of the traditional sub-TAM structure. The cost model of an area-time associated test an...A novel test access mechanism （TAM） architecture with multi test-channel （TC） based on IEEE Standard 1500 is proposed instead of the traditional sub-TAM structure. The cost model of an area-time associated test and the corresponding lower bound of system-on-chip （SoC） test time are established based on this TAM architecture. The model provides a more reliable method to control the SoC scheduling and reduces the complexity in related algorithm research. The result based on the area time associated test cost model has been validated using the ITC02 test benchmark.展开更多
In order to deliver a complete reliable software product, testing is performed. As testing phase carries on, cost of testing process increases and it directly affects the overall project cost. Many a times it happens ...In order to deliver a complete reliable software product, testing is performed. As testing phase carries on, cost of testing process increases and it directly affects the overall project cost. Many a times it happens that the actual cost becomes more than the estimated cost. Cost is considered as the most important parameter with respect to software testing, in software industry. In recent year’s researchers have done a variety of work in the area of Cost optimization by using various concepts like Genetic Algorithm, simulated annealing and Automation in generation of test data etc. This paper proposes an efficient cost effective approach for optimizing the cost of testing using Tabu Search (TS), which will provide maximum code coverage along with the concepts of Dijkstra’s Algorithm which will be implemented in Aspiration criteria of Tabu Search in order to optimize the cost and generate a minimum cost path with maximum coverage.展开更多
Background and Objective: HIV, hepatitis B virus (HBV) and hepatitis C virus (HCV) are very widespread in the world, however, less than 20% of the people affected are diagnosed and treated. This study aimed to determi...Background and Objective: HIV, hepatitis B virus (HBV) and hepatitis C virus (HCV) are very widespread in the world, however, less than 20% of the people affected are diagnosed and treated. This study aimed to determine the prevalence of HIV, HCV and HBV co-infections in pregnant women at Bangui Community University Hospital and the cost of screening. Methods: A cross-sectional study involving consenting pregnant women who came for antenatal care was performed. HIV, HCV antibodies and HBV antigens were detected using Exacto Triplex<sup>?</sup> HIV/HCV/HBsAg rapid test, cross-validated by ELISA tests. Sociodemographic and professional data, the modes of transmission and prevention of HIV and both hepatitis viruses were collected in a standard sheet and analyzed using the Epi-Info software version 7. Results: Pregnant women aged 15 to 24 were the most affected (45.3%);high school girls (46.0%), and pregnant women living in cohabitation (65.3%) were the most represented. Twenty-five (16.7%) worked in the formal sector, 12.7% were unemployed housewives and the remainder in the informal sector. The prevalence of HIV, HBV, and HCV viruses was 11.8%, 21.9% and 22.2%, respectively. The prevalence of co-infections was 8.6% for HIV-HBV, 10.2% for HIV-HCV, 14.7% for HBV-HCV and 6.5% for HIV-HBV-HCV. All positive results and 10% of negative results by the rapid test were confirmed by ELISA tests. The serology of the three viruses costs 39,000 FCFA (60 Euros) by ELISA compared to 10,000 FCFA (15.00 Euros) with Exacto Triplex<sup>?</sup> HIV/HCV/AgHBs (BioSynex, Strasbourg, France). Conclusion: The low level of education and awareness of hepatitis are barriers to development and indicate the importance of improving the literacy rate of women in the Central African Republic (CAR). Likewise, the high prevalence of the three viruses shows the need for the urgent establishment of a national program to combat viral hepatitis in the CAR.展开更多
The problem of sequential fault diagnosis is to construct a diagnosis tree that can isolate the failure sources with minimal test cost. Pervious sequential fault diagnosis strategy generating algorithms only consider ...The problem of sequential fault diagnosis is to construct a diagnosis tree that can isolate the failure sources with minimal test cost. Pervious sequential fault diagnosis strategy generating algorithms only consider the execution cost at application stage, which may result in a solution with poor quality from the view of life cycle cost. Furthermore, due to the fact that uncertain information exists extensively in the real-world systems, the tests are always imperfect. In order to reduce the cost of fault diagnosis in the realistic systems, the sequential fault diagnosis problem with imperfect tests considering life cycle cost is presented and formulated in this work, which is an intractable NP-hard AND/OR decision tree construction problem. An algorithm based on AND/OR graph search is proposed to solve this problem. Heuristic search based on information theory is applied to generate the sub-tree in the algorithm. Some practical issues such as the method to improve the computational efficiency and the diagnosis strategy with multi-outcome tests are discussed. The algorithm is tested and compared with previous algorithms on the simulated systems with different scales and uncertainty. Application on a wheel momentum system of a spacecraft is studied in detail. Both the simulation and application results suggest that the cost of the diagnosis strategy can be reduced significantly by using the proposed algorithm, especially when the placement cost of the tests constitutes a large part of the total cost.展开更多
Accelerated destructive degradation tests(ADDTs)are powerful to provide reliability information in the degradation processes with destructive measurements.In order to carry out an ADDT efficiently,both the estimation ...Accelerated destructive degradation tests(ADDTs)are powerful to provide reliability information in the degradation processes with destructive measurements.In order to carry out an ADDT efficiently,both the estimation precision of parameters and the test cost should be considered.On the basis of the given degradation model and failure criterion,a multiple-objective optimization model for the design of ADDTs is proposed.Under constrains of the maximum measurement time,the total sample size and the number of stress levels,a comprehensive target function is suggested to reflect both the precision of lifetime estimation and total cost,and the optimal test plan is obtained,which is composed by optimal choices for samples size,measurement frequency,and the number of measurements at each stress level.A real example is illustrated to demonstrate the implementation of the proposed approach.展开更多
基金supported by the National Natural Science Foundation of China under Grant No. 60873077/F020107
文摘In many machine learning applications,data are not free,and there is a test cost for each data item. For the economical reason,some existing works try to minimize the test cost and at the same time,preserve a particular property of a given decision system. In this paper,we point out that the test cost one can afford is limited in some applications. Hence,one has to sacrifice respective properties to keep the test cost under a budget. To formalize this issue,we define the test cost constraint attribute reduction problem,where the optimization objective is to minimize the conditional information entropy. This problem is an essential generalization of both the test-cost-sensitive attribute reduction problem and the 0-1 knapsack problem,therefore it is more challenging. We propose a heuristic algorithm based on the information gain and test costs to deal with the new problem. The algorithm is tested on four UCI（University of California-Irvine） datasets with various test cost settings. Experimental results indicate the appropriate setting of the only user-specified parameter λ.
基金Project supported by the SDC Project of Science and Technology Commission of Shanghai Municipality (Grant No.08706201000)the AM Foundation Project of Science and Technology Commission of Shanghai Municipality (Grant No.08700741000)+1 种基金the Leading Academic Discipline Project of Shanghai Education Commission (Grant No.J50104)the Innovation Foundation Project of Shanghai University
文摘A novel test access mechanism （TAM） architecture with multi test-channel （TC） based on IEEE Standard 1500 is proposed instead of the traditional sub-TAM structure. The cost model of an area-time associated test and the corresponding lower bound of system-on-chip （SoC） test time are established based on this TAM architecture. The model provides a more reliable method to control the SoC scheduling and reduces the complexity in related algorithm research. The result based on the area time associated test cost model has been validated using the ITC02 test benchmark.
文摘In order to deliver a complete reliable software product, testing is performed. As testing phase carries on, cost of testing process increases and it directly affects the overall project cost. Many a times it happens that the actual cost becomes more than the estimated cost. Cost is considered as the most important parameter with respect to software testing, in software industry. In recent year’s researchers have done a variety of work in the area of Cost optimization by using various concepts like Genetic Algorithm, simulated annealing and Automation in generation of test data etc. This paper proposes an efficient cost effective approach for optimizing the cost of testing using Tabu Search (TS), which will provide maximum code coverage along with the concepts of Dijkstra’s Algorithm which will be implemented in Aspiration criteria of Tabu Search in order to optimize the cost and generate a minimum cost path with maximum coverage.
文摘Background and Objective: HIV, hepatitis B virus (HBV) and hepatitis C virus (HCV) are very widespread in the world, however, less than 20% of the people affected are diagnosed and treated. This study aimed to determine the prevalence of HIV, HCV and HBV co-infections in pregnant women at Bangui Community University Hospital and the cost of screening. Methods: A cross-sectional study involving consenting pregnant women who came for antenatal care was performed. HIV, HCV antibodies and HBV antigens were detected using Exacto Triplex<sup>?</sup> HIV/HCV/HBsAg rapid test, cross-validated by ELISA tests. Sociodemographic and professional data, the modes of transmission and prevention of HIV and both hepatitis viruses were collected in a standard sheet and analyzed using the Epi-Info software version 7. Results: Pregnant women aged 15 to 24 were the most affected (45.3%);high school girls (46.0%), and pregnant women living in cohabitation (65.3%) were the most represented. Twenty-five (16.7%) worked in the formal sector, 12.7% were unemployed housewives and the remainder in the informal sector. The prevalence of HIV, HBV, and HCV viruses was 11.8%, 21.9% and 22.2%, respectively. The prevalence of co-infections was 8.6% for HIV-HBV, 10.2% for HIV-HCV, 14.7% for HBV-HCV and 6.5% for HIV-HBV-HCV. All positive results and 10% of negative results by the rapid test were confirmed by ELISA tests. The serology of the three viruses costs 39,000 FCFA (60 Euros) by ELISA compared to 10,000 FCFA (15.00 Euros) with Exacto Triplex<sup>?</sup> HIV/HCV/AgHBs (BioSynex, Strasbourg, France). Conclusion: The low level of education and awareness of hepatitis are barriers to development and indicate the importance of improving the literacy rate of women in the Central African Republic (CAR). Likewise, the high prevalence of the three viruses shows the need for the urgent establishment of a national program to combat viral hepatitis in the CAR.
基金Project(C1320063131)supported by China Civil Space Foundation
文摘The problem of sequential fault diagnosis is to construct a diagnosis tree that can isolate the failure sources with minimal test cost. Pervious sequential fault diagnosis strategy generating algorithms only consider the execution cost at application stage, which may result in a solution with poor quality from the view of life cycle cost. Furthermore, due to the fact that uncertain information exists extensively in the real-world systems, the tests are always imperfect. In order to reduce the cost of fault diagnosis in the realistic systems, the sequential fault diagnosis problem with imperfect tests considering life cycle cost is presented and formulated in this work, which is an intractable NP-hard AND/OR decision tree construction problem. An algorithm based on AND/OR graph search is proposed to solve this problem. Heuristic search based on information theory is applied to generate the sub-tree in the algorithm. Some practical issues such as the method to improve the computational efficiency and the diagnosis strategy with multi-outcome tests are discussed. The algorithm is tested and compared with previous algorithms on the simulated systems with different scales and uncertainty. Application on a wheel momentum system of a spacecraft is studied in detail. Both the simulation and application results suggest that the cost of the diagnosis strategy can be reduced significantly by using the proposed algorithm, especially when the placement cost of the tests constitutes a large part of the total cost.
文摘Accelerated destructive degradation tests(ADDTs)are powerful to provide reliability information in the degradation processes with destructive measurements.In order to carry out an ADDT efficiently,both the estimation precision of parameters and the test cost should be considered.On the basis of the given degradation model and failure criterion,a multiple-objective optimization model for the design of ADDTs is proposed.Under constrains of the maximum measurement time,the total sample size and the number of stress levels,a comprehensive target function is suggested to reflect both the precision of lifetime estimation and total cost,and the optimal test plan is obtained,which is composed by optimal choices for samples size,measurement frequency,and the number of measurements at each stress level.A real example is illustrated to demonstrate the implementation of the proposed approach.
