Abstract This paper presents a computer-aided diagnosis for pneumoconiosis radiographs obtained with a common CCD scanner. Since the existing diagnosis systems for pneumo coniosis extract abnormalities of pneumoconiosis from im ages obtained with a special scanner which can appropriatelyapplyforchestradiographs, itisdifficulttoapply the systems to images obtained with a CCD scanner due to un clear shadow and the systems are not practical for... This paper presents a computer-aided diagnosis for pneumoconiosis radiographs obtained with a common CCD scanner. Since the existing diagnosis systems for pneumo coniosis extract abnormalities of pneumoconiosis from im ages obtained with a special scanner which can appropriatelyapplyforchestradiographs, itisdifficulttoapply the systems to images obtained with a CCD scanner due to un clear shadow and the systems are not practical for medical doctors due to high costs. In the unclear images, the ab normal levels of pneumoconiosis could depend on density distribution in each of intercostal and rib areas. Therefore, the proposed method measures the abnormalities by extract ing characteristics of the distribution in the areas. Besides, using theabnormalities, theproposedmethodclassifiestheimagesinto the three categories of pneumoconiosis. Exper imental resultsoftheclassificationsfor51right- lungimagesincluding 6 pneumoconiosis images have shown that the proposed abnormalities are well extracted according to the standards of pneumoconiosis categories. Read More Read Less

Abstract The Ensemble classifier has been an active research topic in the area of machine learning. In a classification task, the ensemble scheme determines a final class label from several individual results which are usually generated by several individual classifiers according to two major principles. The first is to use multiple learning algorithms to form the set of individual outcomes, and... The Ensemble classifier has been an active research topic in the area of machine learning. In a classification task, the ensemble scheme determines a final class label from several individual results which are usually generated by several individual classifiers according to two major principles. The first is to use multiple learning algorithms to form the set of individual outcomes, and the second is to train a set of data fractions, which are generated from a given training dataset, with a weak learning algorithm to generate the set of results. The advantage of the first principle is on classification stability, whereas that of the second principle is on the accuracy improvement in terms of classification accuracy. Thus far, most studies in the literature have been based on either a study of the two principles of ensemble. In this paper we propose a new ensemble scheme to combine the two principles simultaneously. We evaluate the performance on a classification task. Experimental results on several UCI benchmark datasets demonstrate that the proposed framework achieves improved performance in terms of classification accuracy compared to the conventional approaches. Read More Read Less

Abstract Extensive use of web services has lead to introduce web service compositions to execute business workflows. Since several function equivalent services, provided by different service providers, may be available in the Web, the problem of selection of web service compositions arises. QoS aware selection means considering the quality of the service composition in choosing constituent services. So a method is... Extensive use of web services has lead to introduce web service compositions to execute business workflows. Since several function equivalent services, provided by different service providers, may be available in the Web, the problem of selection of web service compositions arises. QoS aware selection means considering the quality of the service composition in choosing constituent services. So a method is needed for computing the quality of a composition and determining the composition with the (almost) best quality. The resulting composite has optimal quality and satisfies user’s quality constraints. In this paper, the problem of QoS aware selection of web service composition is modeled as an optimization problem. Then a variation of Harmony Search algorithm is used to find a near optimal composition.Harmony Search is a recently developed optimization algorithm that mimics the improvisation process. The proposed method based on Harmony Search has been applied to a large variety of composition schemas, generated by a simulation software, and the results show that the proposed method works faster, compared with other approaches. Read More Read Less

Abstract A data element specifies one of the characteristics of its parent element. Therefore, the context of a data element is determined by its parent. Non context driven search engines build relationships between data nodes based solely on their labels and proximity to one another while overlooking their contexts. Therefore, they may return faulty answers. This paper investigates the pitfalls... A data element specifies one of the characteristics of its parent element. Therefore, the context of a data element is determined by its parent. Non context driven search engines build relationships between data nodes based solely on their labels and proximity to one another while overlooking their contexts. Therefore, they may return faulty answers. This paper investigates the pitfalls and limitations of non context-driven XML search engines, which caused by overlooking elements’ contexts. We propose a generic context-driven search frame - work, which could be used as a layer on top of the frameworks of non context-driven search systems. Read More Read Less

Abstract In this paper, we propose a query-oriented fragment search approach for efficient and effective XML search engines on relational database systems. Conventional approaches for XML fragment search have only considered statistics of XML documents stored in a database system to rank a search result. As a result, almost all XML fragment search engines have attained negative results in the research... In this paper, we propose a query-oriented fragment search approach for efficient and effective XML search engines on relational database systems. Conventional approaches for XML fragment search have only considered statistics of XML documents stored in a database system to rank a search result. As a result, almost all XML fragment search engines have attained negative results in the research field of information retrieval (IR). To cope with this problem, considering not only statistics in the database system but also ones extracted from a query helps to improve retrieval accuracy of XML fragment search engines; however, it is also faced with a problem related with their efficiencies. Therefore, we have to devise ways of achieving both highly-relevant search results and efficient query processing. We implemented our XML search engine adopting the query-oriented fragment search approach, and found that it allows users to search XML fragments more accurately and effectively than conventional ones on our XML search engine. Read More Read Less

Abstract The introduction of modern communication and computation technologies has greatly changed the traditional approaches processing computationally intensive applications such as Virtual Environments (VEs). Distributed and multi-server architectures are scalable and exploit the concepts of spatial and temporal locality for handling massive Distributed VEs (DVEs). However, using these environments require special attention for scalability, consistency, load balancing, and latency. The existing... The introduction of modern communication and computation technologies has greatly changed the traditional approaches processing computationally intensive applications such as Virtual Environments (VEs). Distributed and multi-server architectures are scalable and exploit the concepts of spatial and temporal locality for handling massive Distributed VEs (DVEs). However, using these environments require special attention for scalability, consistency, load balancing, and latency. The existing mechanisms primarily developed for games have key performance issues in managing large scale virtual worlds. In our previous work (Farooq and Glauert, 2009a), we have proposed a hierarchical infrastructure named JoHNUM for the development of scalable and consistent virtual worlds. Further, we have introduced the concept of aggregate region assignment to minimise resource utilisation and distribute the load as balanced as possible among servers (Farooq and Glauert, 2009a, 2009b). This paper presents our proposed load distribution algorithm that yields regular and contiguous areas for assignment. It examines possible combinations of regions based on proposed strategies but greatly reduces the aggregation process by using different filters. It also presents a simple communication model justifying the proposed strategies and load distribution algorithm. Simulation results with an extended set of experiments show that the proposed algorithm obtains fair distribution of load and has the capability to cope with small, medium and large scale environments while minimising communication and implementation cost. Read More Read Less

Abstract Future wireless communication systems are expected to provide more stable and higher data rate transmissions in the whole network. WiMAX is one of the technologies to cover diverse geographic regions. WiMAX covers the high dense urban areas and provide access over large geographic regions in remote rural area. However, in both scenarios it is challenging to cover the entire service... Future wireless communication systems are expected to provide more stable and higher data rate transmissions in the whole network. WiMAX is one of the technologies to cover diverse geographic regions. WiMAX covers the high dense urban areas and provide access over large geographic regions in remote rural area. However, in both scenarios it is challenging to cover the entire service area. In both cases relays are foreseen to extend the range of the BS and to increase the system throughput allowing for a cost-efficient deployment and service. As multi-hop/relay has complex interactions between the routing and medium access control decisions, the extent to which analytical expressions can be used to explore its benefits are limited. Consequently simulations tend to be the preferred way of assessing the performance of relays. In this paper, we explain, how to deploy fixed-relay and mobile relay by using computer simulation in WiMAX system and we consistently observed that the throughput is increased and the outage is decreased in the relay-augmented WiMAX network, which is converted to range extension without any capacity penalty, for the realistic range of values of the propagation and other system parameters investigated. Read More Read Less

Abstract Citations management is an important task in managing digital libraries. Citations provide valuable information e.g., used in evaluating an author’s influences or scholarly quality (the impact factor of research journals). But although a reliable and effective autonomous citation management is essential, manual citation management can be extremely costly. Automatic citation mining on the other hand is a non-trivial task mainly... Citations management is an important task in managing digital libraries. Citations provide valuable information e.g., used in evaluating an author’s influences or scholarly quality (the impact factor of research journals). But although a reliable and effective autonomous citation management is essential, manual citation management can be extremely costly. Automatic citation mining on the other hand is a non-trivial task mainly due to non-conforming citation styles, spelling errors and the difficulty of reliably extracting text from PDF documents. In this paper we propose a novel rule-based autonomous citation mining technique, to address this important task. We define a set of common heuristics that together allow to improve the state of the art in automatic citation mining. Moreover, by first disambiguating citations based on venues, our technique significantly enhances the correct discovery of citations. Our experiments show that the proposed approach is indeed able to overcome limitations of current leading citation indexes such as ISI Web of Knowledge, Citeseer and Google Scholar. Read More Read Less

Abstract Linked Open Data (LOD) is becoming an essential part of the Semantic Web. Although LOD has amassed large quantities of structured data from diverse, openly available data sources, there is still a lack of user-friendly interfaces and mechanisms for exploring this huge resource. In this paper, we describe a methodology for harvesting relevant information from the gigantic LOD cloud.... Linked Open Data (LOD) is becoming an essential part of the Semantic Web. Although LOD has amassed large quantities of structured data from diverse, openly available data sources, there is still a lack of user-friendly interfaces and mechanisms for exploring this huge resource. In this paper, we describe a methodology for harvesting relevant information from the gigantic LOD cloud. The methodology is based on combination of information: identification, extraction, integration and presentation. Relevant information is identified by using a set of heuristics. The identified information resource is extracted by employing an intelligent URI discovery technique. The extracted information is further integrated with the help of a Concept Aggregation Framework. Then the information is presented to end users in logical informational aspects. Thereby, the proposed system is capable of hiding complex underlying semantic mechanics from end users and reducing the users’ cognitive load in locating relevant information. In this paper, we describe the methodology and its implementation in the CAF-SIAL system, and compare it with the state of the art. Read More Read Less

Abstract In market-like grids, applying an economical method for providing fault-tolerant job scheduling is necessary. Since we have to pay cost for job execution in economic grids, consumers tend to execute their jobs on a grid with minimum cost and acceptable completion time. To do this, many approaches have been proposed, but they unilaterally consider either time/cost or consumer/provider profit. Based... In market-like grids, applying an economical method for providing fault-tolerant job scheduling is necessary. Since we have to pay cost for job execution in economic grids, consumers tend to execute their jobs on a grid with minimum cost and acceptable completion time. To do this, many approaches have been proposed, but they unilaterally consider either time/cost or consumer/provider profit. Based on current research, a new approach has been proposed; the approach is an optimized, fault-tolerant method for meta-scheduler that considers bilateral satisfaction for consumer/provider and a tolerable completion time. We suppose that, if one of the provider nodes fails, there would not be enough time to start a task on a new node from the beginning. The experimental results show a promising improvement in job scheduling with less computation cost, better fault-tolerance, and an acceptable completion time. Read More Read Less