查看更多>>摘要：The objective of this paper is to empirically evaluate SKLSEForum, a lightweight forum-based tool, for predicting the likelihood of acceptance of the tool in requirements elicitation practice and finding directions for improvement. We analyzed three data sets from a family of experiments, deriving from two replicated controlled experiments and a survey of expert panel. Results showed that SKLSEForum can improve the quantity of posts and judge satisfaction of requirements. In addition, the tool was perceived as useful and easy to use by the participants, who also expressed their intention to use SKLSEForum in the future. Furthermore, some insights about the improvement of SKLSEForum also have been found.
查看更多>>摘要：Data analysis is a quite important process and has been extensively employed in many areas. Especially in statistics, the distribution type test of data often needs to be handled. This paper presents that linear regression as a type of universal method can be applied to the distribution type test. In reliability engineering due to the failure data is commonly of non-linear relationship so that the linear regression method can not be directly employed. This difficulty can be resolved through linear transformation. Aimed for the linearization procedure of four kind of typical distributions, e.g., exponential distribution, normal distribution, logarithmic normal distribution and two-parameter Weibull distribution, their transformations are respectively different. After linearization transformations, correlation coefficients had been used as a criterion to choose the most matched distribution type. This method can be conveniently operated in MS EXCEL. Several samples and experiments illustrated the detailed transformation and the efficiency than other methods.
查看更多>>摘要：The work presented in this paper measures the contribution of developers towards evolved structural design of software systems. Measuring the contribution of developers is useful for project managers who manage the development process. Project managers can empirically identify developers who made changes to the structural design and compare among them based on their contributions. The proposed measures help to understand the nature of developers' code changes activities. The process of calculating the measures is based on the historical code changes committed by developers. Specifically, code changes that affect the corresponding UML class diagram representation of the source code. Both type and amount of previous changes to class elements are used to measure the design contributions. The proposed measures are helpful for open source projects where no detailed information is available about various developers involved in the development process. A tool has been developed to automatically measure the contributions based on archived historical code changes. The measures have been applied on two open source projects. Results showed that only small number of developers have the major design contributions.
查看更多>>摘要：In this paper, a new method named modular multi-directional two-dimensional principle component analysis (M~2D2DPCA) is proposed for face recognition. First, the original images are rotated at some predetermined angles so that we may extract features from the images in any direction. Then we divide the rotated images into smaller sub-images and apply 2DPCA approach to each of these sub-images. Finally we propose a fusion method named modular multi-directional 2DPCA (M~2D2DPCA) to combine a bank of preliminary results in different directions. Compared with conventional 2DPCA based algorithms, the advantage of the proposed method is that it can extract significant features from the images in any direction and avoid the effects of varying illumination and facial expression. The results of the experiments on ORL and Yale datasets show that the proposed M~2D2DPCA method can obtain a higher recognition rate than the conventional 2DPCA based methods.
查看更多>>摘要：The contract is the bridge to transaction, as well as the core of business practice. Accompanying with the development of electronic commerce, contract negotiation under more complicated environment in advantage of internet is becoming increasingly urgent. This paper briefly introduces the status of electronic signature and electronic contract in China, then presents a reference architecture for an online system for electronic contract negotiation and signature in detail. This architecture is systematically designed on the basis of electronic signature technology to realize online contract signing, electronic contract storage on the trusted third party and online or offline verifications. All above ensures electronic contract's confidentiality, integrity and non-repudiation, makes a breakthrough for electronic signature in business field, and realizes the electronization in the whole process of business negotiation.
查看更多>>摘要：Structured link vector model (SLVM) and its improved version depend on statistical term measures to implement XML document representation. As a result, they ignore the lexical semantics of terms and its mutual information, leading to text classification errors. This paper proposed a XML document representation method, WordNet-based lexical-semantic SLVM, to solve the problem. Using WordNet, this method constructed a data structure for characterizing lexical semantic contents of XML document, and adjusted EM modeling to disambiguate word stems. Then, synset matrix of lexical semantic contents was built in the lexical-semantic feature space for XML document representation, and lexical semantic relations were marked on it to construct the feature matrix in lexical-semantic SLVM. On categorized dataset of Wikipedia XML, using NWKNN classification algorithm, the experimental results show that the feature matrix of our method performs F1 measure better than original SLVM and frequent sub-tree SLVM based on TF-IDF.
查看更多>>摘要：Bug Triaging is a vital part of issue management systems. Bug triaging deals with assigning a developer the task of an incoming bug. This activity is error prone and time consuming if done manually. There is a need for automated support to accelerate this process. The current automated bug triaging systems exploits the text contents of the bug and the tossing relations among the developers. The automated bug triaging systems estimate the optimal bath between the first assignee of the bug and the bug resolver using the tossing relations. The metrics used for assessing the efficiency of bug triaging systems that are based on tossing relations is Mean number of Steps To Resolve (MSTR). This metric quantifies the number of steps reduced by the predicted path compared to the original path. It does not capture how far the retrieved path is in alignment with the actual path. MSTR does reveal the information regarding the extent to which the order of the developers in the retrieved path is in line with that of the original path. In addition, there are no indicators for measuring the strength of the retrieved path. In this paper, we propose two metrics (ⅰ) Path Similarity Metric which quantifies path alignment based on pair wise path alignment and (ⅱ) Path Alignment Indicator that measures the effectiveness of the retrieved path based on degree centrality. The effectiveness of the two proposed metrics is validated using bug reports extracted from the Eclipse project.
查看更多>>摘要：This study presents a case study demonstrating the application of Six Sigma (DMAIC) to successfully reduce the number of online petitions on the e-petition system in Taiwan Power Company (TPC). First, in the define (D) stage, a total of 1,292 petitions for the year 2010 in TPC are received and divided into 9 categories. In the measure (M) stage, the category that received the most petition letters is "line removal", which accounts for 29.3% of the total petitions. The analyze (A) stage indicates that TPC did not provide relevant application procedures for "line removal" service. In the improve (I) stage, this study presents three approaches to speed up the "line removal" application process. Finally, a well-organized standard of procedure and workflow control charts has been developed in order to monitor the overall application and construction processes in the control (C) stage. After implementing Six Sigma (DMAIC), the number of petitions for "line removal" decreased by 66% from 379 in 2010 to 129 in 2011.
查看更多>>摘要：This paper presents a novel energy constrained target K-coverage algorithm, and the proposed algorithm is suitable to be exploited in heterogeneous wireless sensor networks. Particularly, the network activity in heterogeneous wireless sensor networks in this paper is organized in rounds, and each round is constructed by initial step and information sensing step. Furthermore, to prolong the network lifetime in target K-coverage process, the remaining energy and the sensing ability of each sensor are calculated in advance. Afterwards, the sensing unit set is defined to record sensing attributes that can be only covered by the target sensors. Furthermore, the attribute set contains several sensing attributes, and these attributes can cover the targets utilizing the specific sensing unit. In order to guarantee the K-coverage constraints, for each sensing attribute, we set a function to test whether the sensing attribute is covered. If the sensing attribute is covered by at least K sensors, the function return true, and then the decisions which represent if a sensor should be turned on or not are broadcast to each sensor's one-hop neighbors. Finally, a series of experiments are conducted to make performance evaluation. In these experiments, all targets and wireless sensors are randomly allocated in the sensing field, and six types of experimental settings with different number wireless sensor are utilized. Experimental results show that, our proposed algorithm performs better than EF method and KTC_MNL method, and the performance of our algorithm is close to the optimal method.
查看更多>>摘要：How to cluster different query interfaces effectively is one of the most core issues when generating integrated query interface on Deep Web integration domain. However, with the rapid development of Internet technology, the number of Deep Web query interface shows an explosive growth trend. For this reason, the traditional stand-alone Deep Web query interface clustering approaches encounter bottlenecks in terms of time complexity and space complexity. After further study of the Hadoop distributed platforms and Map Reduce programming model, a Deep Web query interface clustering algorithm based on Hadoop platform is designed and implemented, in which the Vector Space Model (VSM) and Latent Semantic Analysis (LSA) are employed to represent "Query Interfaces-Attributes" relationships. The experimental results show that the proposed algorithm has better scalability and speedup ratio by using Hadoop architecture.
NETL
NSTL