Applying data mining to learn system dynamics in a biological model

Data mining consists of a set of powerful methods that have been successfully applied to many different application domains, including business, engineering, and bioinformatics. In this paper, we propose an innovative approach that uses genetic algorithms to mine a set of temporal behavior data output by a biological system in order to determine the kinetic parameters of the system. Analyzing the behavior of a biological network is a complicated task. In our approach, the machine learning method is integrated with the framework of system dynamics so that its findings are expressed in a form of system dynamics model. An application of the method to the cell division cycle model has shown that the method can discover approximate parametric values of the system and reproduce the input behavior.     

融合机器学习与知识推理的可解释性框架

针对基于规则的可解释性模型可能出现的规则无法反映模型真实决策情况的问题, 提出了一种融合机器学习和知识推理两种途径的可解释性框架. 框架演进目标特征结果和推理结果, 在二者相同且都较为可靠的情况下实现可解释性. 目标特征结果通过机器学习模型直接得到, 推理结果通过子特征分类结果结合规则进行知识推理得到, 两个结果是否可靠通过计算可信度来判断. 使用面向液基细胞学检查图像的融合学习与推理的某类宫颈癌细胞识别案例对框架进行验证, 实验表明, 该框架能够赋予模型的真实决策结果以可解释性, 并在迭代过程中提升了分类精度. 这帮助人们理解系统做出决策的逻辑, 以及更好地了解结果可能失败的原因.     

一种面向形式化表格需求模型的测试用例生成方法

现代安全关键性系统的软件规模和复杂性的快速增长给这类安全关键性软件系统的开发带来了很多挑战。传统文本文档的需求描述方法无法保证此类系统的开发进度和系统可靠性要求。为此文中提出了一种兼具可读性和可自动分析的形式化表格需求建模方法。文中介绍了一种针对这种表格模型测试用例的自动生成方法,工作包括对该形式化需求表格模型展开语义分析,建立需求模型的控制树结构,得到其测试等价类;为了减少不必要的测试,定义了不同安全级别的软件需求模型的测试覆盖标准,并针对不同覆盖率准则分别给出基于控制树结构的测试路径约束选择方法;对于每条路径约束测试等价类,提出了基于域错误的测试用例选择方法,能够自动生成所需的检测域错误的测试用例集。最后,通过一个需求模型实例展示了所提方法的有效性。     

A meta-heuristic approach to solve a JIT scheduling problem in hybrid flow shop

In this paper we address a hybrid flow shop scheduling problem considering the minimization of the sum of the total earliness and tardiness penalties. This problem is proven to be NP-hard, and consequently the development of heuristic and meta-heuristic approaches to solve it is well justified. So, we propose an ant colony optimization method to deal with this problem. Our proposed method has several features, including some heuristics that specifically take into account both earliness and tardiness penalties to compute the heuristic information values. The performance of our algorithm is tested by numerical experiments on a large number of randomly generated problems. A comparison with solutions performance obtained by some constructive heuristics is presented. The results show that the proposed approach performs well for this problem.     

Prioritizing Test Cases for Memory Leaks in Android Applications

Mobile applications usually can only access limited amount of memory. Improper use of the memory can cause memory leaks, which may lead to performance slowdowns or even cause applications to be unexpectedly killed. Although a large body of research has been devoted into the memory leak diagnosing techniques after leaks have been discovered, it is still challenging to find out the memory leak phenomena at first. Testing is the most widely used technique for failure discovery. However, traditional testing techniques are not directed for the discovery of memory leaks. They may spend lots of time on testing unlikely leaking executions and therefore can be inefficient. To address the problem, we propose a novel approach to prioritize test cases according to their likelihood to cause memory leaks in a given test suite. It firstly builds a prediction model to determine whether each test can potentially lead to memory leaks based on machine learning on selected code features. Then, for each input test case, we partly run it to get its code features and predict its likelihood to cause leaks. The most suspicious test cases will be suggested to run at first in order to reveal memory leak faults as soon as possible. Experimental evaluation on several Android applications shows that our approach is effective.     

同步语言的时间可预测多线程代码生成方法

能够提供更强计算能力的多核处理器将在安全关键系统中得到广泛应用.但是,由于现代处理器所使用的流水线、乱序执行、动态分支预测、Cache等性能提高机制以及多核之间的资源共享,使得系统的最坏执行时间分析变得非常困难.为此,国际学术界提出时间可预测系统设计的思想,以降低系统的最坏执行时间分析难度.已有研究主要关注硬件层次及其编译方法的调整和优化,而较少关注软件层次,即时间可预测多线程代码的构造方法以及到多核硬件平台的映射.本文提出一种基于同步语言模型驱动的时间可预测多线程代码生成方法,并对代码生成器的语义保持进行证明;提出一种基于AADL(Architecture Analysis and Design Language)的时间可预测多核体系结构模型,作为本文研究的目标平台;最后,给出多线程代码到多核体系结构模型的映射方法,并给出系统性质的分析框架.     

Multi-objective adaptive large neighborhood search for distributed reentrant permutation flow shop scheduling

Factory management plays an important role in improving the productivity and quality of service in the production process. In particular, the distributed permutation flow shop scheduling problem with multiple factories is considered a priority factor in the factory automation. This study proposes a novel model of the developed distributed scheduling by supplementing the reentrant characteristic into the model of distributed reentrant permutation flow shop (DRPFS) scheduling. This problem is described as a given set of jobs with a number of reentrant layers is processed in the factories, which compromises a set of machines, with the same properties. The aim of the study is to determine the number of factory needs to be used, jobs assignment to certain factory and sequence of job assigned to the factory in order to simultaneously satisfy three objectives of minimizing makespan, total cost and average tardiness. To do this, a novel multi-objective adaptive large neighborhood search (MOALNS) algorithm is developed for finding the near optimal solutions based on the Pareto front. Various destroy and repair operators are presented to balance between intensification and diversification of searching process. The numerical examples of computational experiments are carried out to validate the proposed model. The analytical results on the performance of proposed algorithm are checked and compared with the existing methods to validate the effectiveness and robustness of the proposed potential algorithm in handling the DRPFS problem.     

Applying genetic algorithm to a new bi-objective stochastic model for transportation,location, and allocation of hazardous materials

In 2013, approximately 15,600 HAZMAT accidents with 158 injuries and fatalities have been reported in the USA (“Transportation Statistics Bureau”). Managing hazardous material (HAZMAT) transportation and locating the disposal sites for these materials properly can significantly reduce the risk of accidents and its environmental and social aspects. In this research, a new stochastic model for transportation, location, and allocation of hazardous materials is proposed. The cost of transportation is considered to be of a stochastic nature. The objective function minimizes the total cost and risk of locating facilities and transportation of HAZMATs. The decisions which have to be made are: (1) where to open the facilities and disposal sites; (2) to which facilities every customer should be assigned; (3) to which disposal site each facility should be assigned; and (4) which routes a facility should choose to reach the customers and disposal sites. A novel genetic algorithm (GA) is applied to the model. The results show the efficiency of the proposed GA in terms of finding high quality solutions in a short time.     

An effective hybrid teaching–learning-based optimization algorithm for permutation flow shop scheduling problem

Permutation flow shop scheduling (PFSP) is among the most studied scheduling settings. In this paper, a hybrid Teaching–Learning-Based Optimization algorithm (HTLBO), which combines a novel teaching–learning-based optimization algorithm for solution evolution and a variable neighborhood search (VNS) for fast solution improvement, is proposed for PFSP to determine the job sequence with minimization of makespan criterion and minimization of maximum lateness criterion, respectively. To convert the individual to the job permutation, a largest order value (LOV) rule is utilized. Furthermore, a simulated annealing (SA) is adopted as the local search method of VNS after the shaking procedure. Experimental comparisons over public PFSP test instances with other competitive algorithms show the effectiveness of the proposed algorithm. For the DMU problems, 19 new upper bounds are obtained for the instances with makespan criterion and 88 new upper bounds are obtained for the instances with maximum lateness criterion.