Abstract Cloud computing is a term referring to the latest new computing paradigm based on the Internet where a program or application runs on a connected server or servers rather than on a local computing device. Software as a Service (SaaS) is one of the most important computing services in cloud computing. The advantages of SaaS products are well recognized including... Cloud computing is a term referring to the latest new computing paradigm based on the Internet where a program or application runs on a connected server or servers rather than on a local computing device. Software as a Service (SaaS) is one of the most important computing services in cloud computing. The advantages of SaaS products are well recognized including lower cost of entry, minimal demands on IT and easy upgrades as well as more rigorous. However, this approach has introduced new challenges in SaaS resource management in data centres. One of the challenges is deploying softwares comprised of service and data components over servers in cloud to provide users with required services easily and quickly in SaaS cloud computing. This paper analyzes the SaaS Placement Problem (SPP) and proposes a mathematical model for Composite SaaS placement in cloud. An Ant Colony Optimization Algorithm (ACOA) with performance matching degree Strategy is applied to the problem with the aim of minimizing the total estimated execution time of all the SaaS components and reducing the algorithm computation time. The ACOA is then evaluated against the Genetic Algorithm in experiments and the results show that the ACOA applied to the SPP not only has quicker computation time but also generates better solutions than the Genetic Algorithm (GA). Read More Read Less

Abstract When disaster strikes, the rapid deployment of Wireless Sensor Networks (WSNs) in the disaster area is helpful for real-time monitoring of environmental indicators and improving the efficiency and quality of emergency rescue. In the meanwhile, seveal factors such as the unpredictability of disaster, harsh environment and the urgency of network system put forward lots of special needs in network architecture,... When disaster strikes, the rapid deployment of Wireless Sensor Networks (WSNs) in the disaster area is helpful for real-time monitoring of environmental indicators and improving the efficiency and quality of emergency rescue. In the meanwhile, seveal factors such as the unpredictability of disaster, harsh environment and the urgency of network system put forward lots of special needs in network architecture, sensor nodes designing, and technical realization. This paper proposes a new routing framework for building WSNs in emergency circumstances. In particular, a mobile access mechanism based on the X-MAC protocol has been designed. An effective path of interest propagation strategy, where the WSN node positions have been used to guide the interest spread, has been proposed. Experimental results demonstrate that our proposed framework can significantly reduce the network traffic and energy consumption in directed diffusion. Read More Read Less

Abstract Text in images is a very important clue for image indexing and retrieving. Unfortunately, it is a challenging work to accurately and robustly extract text from a complex background image. In this paper, a novel region-based text extraction method is proposed. In doing so, the candidate text regions are detected by 8-connected objects detection algorithm based on the edge image.... Text in images is a very important clue for image indexing and retrieving. Unfortunately, it is a challenging work to accurately and robustly extract text from a complex background image. In this paper, a novel region-based text extraction method is proposed. In doing so, the candidate text regions are detected by 8-connected objects detection algorithm based on the edge image. Then the non-text regions are filtered out using shape, texture and stroke width rules. Finally, the remaining regions are grouped into text lines. Since stroke width is the intrinsic and particular characteristics of the text, the accuracy of the non-text filter are notably promoted. The improved Stroke Width Transform in the paper is less computing complexities and more accurate. Experimental results on sample ICDAR competition Dataset and our dataset show that the proposed method has the best performance compared with other five methods. Read More Read Less

Abstract Collaborative business making is emerging as a possible solution for the difficulties that Small and Medium Enterprises (SMEs) are having in the current difficult economic scenarios. Collaboration, as opposed to competition, provides a competitive advantage to companies and organizations that operate in a joint business structure. When dealing with multiple organizations, managers must access unified strategic information obtained from the... Collaborative business making is emerging as a possible solution for the difficulties that Small and Medium Enterprises (SMEs) are having in the current difficult economic scenarios. Collaboration, as opposed to competition, provides a competitive advantage to companies and organizations that operate in a joint business structure. When dealing with multiple organizations, managers must access unified strategic information obtained from the knowledge repositories of each individual organization; unfortunately, traditional Business Intelligence (BI) tools are not designed with the aim of collaboration so the task becomes difficult from a managerial, organizational and technological point of view. To deal with this shortcoming, we provide an integration, mapping-based, methodology for heterogeneous Data Warehouses that aims at facilitating business stakeholders access to unified strategic information obtained from a network of heterogeneous collaborating SMEs. A complete formalization, based on graph theory and the RELEVANT clustering approach is provided together with experimental evaluation of the proposed methodology over real DW instances. Read More Read Less

Abstract With the development of information technology, there has been an increased popularity in the use of electronic texts. Topic detection and tracking can identify hot information from isolated texts. Obtaining hot topics has become an important issue in recent years. The combination of statistics and natural language processing was utilized in the current study to discover hot topics from texts.... With the development of information technology, there has been an increased popularity in the use of electronic texts. Topic detection and tracking can identify hot information from isolated texts. Obtaining hot topics has become an important issue in recent years. The combination of statistics and natural language processing was utilized in the current study to discover hot topics from texts. First, the statistics technique was adopted to obtain frequent and high weighted words. Then, the linguistic grammar rules were used to generate the candidate phrases. Finally, the hot phrase topics were obtained based on the weight computation method of phrases. Experiment results showed that the proposed approach was effective in that the extracted topics can express more comprehensive information. This study's results are meaningful in the areas of text classification, text clustering, information retrieval, and construction of high-quality Tibetan corpus. Read More Read Less

Abstract Although community, one of the general characteristics of complex networks, has obvious hierarchical structure, in-depth research on its application in the current community detection algorithms is limited. In this paper we present a novel hierarchical community detection algorithm which starts from the node similarity calculation based on local adjacency in networks. Then we find the initial communities and joint them... Although community, one of the general characteristics of complex networks, has obvious hierarchical structure, in-depth research on its application in the current community detection algorithms is limited. In this paper we present a novel hierarchical community detection algorithm which starts from the node similarity calculation based on local adjacency in networks. Then we find the initial communities and joint them one by one as starting point from the bottom all the way to the top until all nodes fall within the same community. The concept of community similarity is defined for better community incorporation and each time, initial communities with the most similarity are integrated. Taking the advantage of time efficiency in local community mining, this kind of algorithm recognizes the community structure accurately when revealing the hierarchical structure of community, as the experiment shows. Read More Read Less