Nowadays, most users carry high computing power mobile devices where speech recognition is certainly one of the main technologies available in every modern smartphone, although battery draining and application performance (resource shortage) have a big impact on the experienced quality. Shifting applications and services to the cloud may help to improve mobile user satisfaction as demonstrated by several ongoing efforts in the mobile cloud area. However, the quality of speech recognition is still not sufficient in many complex cases to replace the common hand written text, especially when prompt reaction to short-term provisioning requests is required. To address the new scenario, this paper proposes a mobile cloud infrastructure to support the extraction of semantics information from speech recognition in the Social Care domain, where carers have to speak about their patients conditions in order to have reliable notes used afterward to plan the best support. We present not only an architecture proposal, but also a real prototype that we have deployed and thoroughly assessed with different queries, accents, and in presence of load peaks, in our experimental mobile cloud Platform as a Service (PaaS) testbed based on Cloud Foundry.

Outer joins are ubiquitous in many workloads and Big Data systems. The question of how to best execute outer joins in large parallel systems is particularly challenging, as real world datasets are characterized by data skew leading to performance issues. Although skew handling techniques have been extensively studied for inner joins, there is little published work solving the corresponding problem for parallel outer joins, especially in the extremely popular Cloud computing environment. Conventional approaches to the problem such as ones based on hash redistribution often lead to load balancing problems while duplication-based approaches incur significant overhead in terms of network communication. In this paper, we propose a new approach for efficient skew handling in outer joins over a Cloud computing environment. We present an efficient implementation of our approach over the Spark framework. We evaluate the performance of our approach on a 192-core system with large test datasets in excess of 100 GB and with varying skew. Experimental results show that our approach is scalable and, at least in cases of high skew, significantly faster than the state-of-the-art.

Optimal heuristic search has been successful in many domains, including journey planning, route planning and puzzle solving. Existing work typically assumes that the cost of each action can easily be obtained. However, in many problems, the exact edge cost is expensive to compute. Existing search algorithms face a significant performance bottleneck, due to an excessive overhead associated with dynamically calculating exact edge costs. We present DEA*, an algorithm for problems with expensive edge cost computations. DEA* combines heuristic edge cost evaluations with delayed node expansions, reducing the number of exact edge computations. We formally prove that DEA* is optimal and it is efficient with respect to the number of exact edge cost computations. We empirically evaluate DEA* on multiple-worker routing problems where the exact edge cost is calculated by invoking an external multi-modal journey planning engine. The results demonstrate the effectiveness of our ideas in reducing the computational time and improving the solving ability. In addition, we show the advantages of DEA* in domain-independent planning, where we simulate that accurate edge costs are expensive to compute.

Efficiently detecting conversation threads from a pool of messages, such as social network chats, emails, comments to posts, news etc., is relevant for various applications, including Web Marketing, Information Retrieval and Digital Forensics. Existing approaches focus on text similarity using keywords as features that are strongly dependent on the dataset. Therefore, dealing with new corpora requires further costly analyses conducted by experts to find out new relevant features. This paper introduces a novel method to detect threads from any type of conversational texts overcoming the issue of previously determining specific features for each dataset. To automatically determine the relevant features of messages we map each message into a three dimensional representation based on its semantic content, the social interactions in terms of sender/recipients and its timestamp; then clustering is used to detect conversation threads. In addition, we propose a supervised approach to detect conversation threads that builds a classification model which combines the above extracted features for predicting whether a pair of messages belongs to the same thread or not. Our model harnesses the distance measure of a message to a cluster representing a thread to capture the probability that a message is part of that same thread. We present our experimental results on seven datasets, pertaining to different types of messages, and demonstrate the effectiveness of our method in the detection of conversation threads, clearly outperforming the state of the art and yielding an improvement of up to a 19%.

The Semantic Web comprises enormous volumes of semi-structured data elements. For interoperability, these elements are represented by long strings. Such representations are not efficient for the purposes of applications that perform computations over large volumes of such information. A common approach to alleviate this problem is through the use of compression methods that produce more compact representations of the data. The use of dictionary encoding is particularly prevalent in Semantic Web database systems for this purpose. However, centralized implementations present performance bottlenecks, giving rise to the need for scalable, efficient distributed encoding schemes. In this paper, we propose an efficient algorithm for fast encoding large Semantic Web data. Specially, we present the detailed implementation of our approach based on the state-of-art asynchronous partitioned global address space (APGAS) parallel programming model. We evaluate performance on a cluster of up to 384 cores and datasets of up to 11 billion triples (1.9 TB). Compared to the state-of-art approach, we demonstrate a speed-up of$2.6 - 7.4\times$and excellent scalability. In the meantime, these results also illustrate the significant potential of the APGAS model for efficient implementation of dictionary encoding and contributes to the engineering of more efficient, larger scale Semantic Web applications.

Cloud businesses need comprehensive visibility on hardware and software components, their utilization and their configuration. In addition, they need to integrate such information with their asset management systems and publicly available information such as hardware specifications. In this paper, we present an approach for cloud management and monitoring based on a semantic layer that unifies different interfaces and representations, and makes all relevant information accessible from a single point. We show a proof-of-concept based on OpenStack and Linked Data technologies and evaluate it in terms of overhead, query execution times, and effectiveness in the data gathering phase. Our findings indicate that a semantics-based approach is indeed feasible and advantageous for providing uniform access across different cloud environments and levels.

Knowledge on the Web relies heavily on multi-relational representations, such as RDF and Schema.org. Automatically extracting knowledge from documents and linking existing databases are common approaches to construct multi-relational data. Complementary to such approaches, there is still a strong demand for manually encoding human expert knowledge. For example, human annotation is necessary for constructing a common-sense knowledge base, which stores facts implicitly shared in a community, because such knowledge rarely appears in documents. As human annotation is both tedious and costly, an important research challenge is how to best use limited human resources, whiles maximizing the quality of the resulting dataset. In this paper, we formalize the problem of dataset construction as active learning problems and present the Active Multi-relational Data Construction (AMDC) method. AMDC repeatedly interleaves multi-relational learning and expert input acquisition, allowing us to acquire helpful labels for data construction. Experiments on real datasets demonstrate that our solution increases the number of positive triples by a factor of 2.28 to 17.0, and that the predictive performance of the multi-relational model in AMDC achieves the highest or comparable to the best performance throughout the data construction process.

Distributed RDF data management systems become increasingly important with the growth of the Semantic Web. Regardless, current methods meet performance bottlenecks either on data loading or querying when processing large amounts of data. In this work, we propose efficient methods for processing RDF using dynamic data re-partitioning to enable rapid analysis of large datasets. Our approach adopts a two-tier index architecture on each computation node: (1) a lightweight primary index, to keep loading times low, and (2) a series of dynamic, multi-level secondary indexes, calculated as a by-product of query execution, to decrease or remove inter-machine data movement for subsequent queries that contain the same graph patterns. In addition, we propose methods to replace some secondary indexes with distributed filters, so as to decrease memory consumption. Experimental results on a commodity cluster with 16 nodes show that the method presents good scale-out characteristics and can indeed vastly improve loading speeds while remaining competitive in terms of performance. Specifically, our approach can load a dataset of 1.1 billion triples at a rate of 2.48 million triples per second and provide competitive performance to RDF-3X and 4store for expensive queries.

Outer joins are ubiquitous in databases and big data systems. The question of how best to execute outer joins in large parallel systems is particularly challenging as real world datasets are characterized by data skew leading to performance issues. Although skew handling techniques have been extensively studied for inner joins, there is little published work solving the corresponding problem for parallel outer joins. Conventional approaches to this problem such as ones based on hash redistribution often lead to load balancing problems while duplication-based approaches incurs significant overhead in terms of network communication. In this paper, we propose a new algorithm, query with counters (QC), for directly handling skew in outer joins on distributed architectures. We present an efficient implementation of our approach based on the asynchronous partitioned global address space (APGAS) parallel programming model. We evaluate the performance of our approach on a cluster of 192 cores (16 nodes) and datasets of 1 billion tuples with different skew. Experimental results show that our method is scalable and, in cases of high skew, faster than the state-of-the-art.

The performance of joins in parallel database management systems is critical for data intensive operations such as querying. Since data skew is common in many applications, poorly engineered join operations result in load imbalance and performance bottlenecks. State-of-the-art methods designed to handle this problem offer significant improvements over naive implementations. However, performance could be further improved by removing the dependency on global skew knowledge and broadcasting. In this paper, we propose PRPQ (partial redistribution & partial query), an efficient and robust join algorithm for processing large-scale joins over distributed systems. We present the detailed implementation and a quantitative evaluation of our method. The experimental results demonstrate that the proposed PRPQ algorithm is indeed robust and scalable under a wide range of skew conditions. Specifically, compared to the state-of-art PRPD method, we achieve 16% - 167% performance improvement and 24% - 54% less network communication under different join workloads.

We organize and present the 5th version of the International Workshop on Web-scale Knowledge Representation, Retrieval and Reasoning (Web-KR 2014) as a continuous effort to discuss and provide possible theories and techniques to deal with the barriers for knowledge processing at Web scale. This workshop was held in conjunction with the 2014 ACM International Conference on Information and Knowledge Management (CIKM 2014) on November 3rd, 2014 in Shanghai, China. Compared to previous workshops under the same title, accepted papers of this workshop covers even wider topics in the field. The contributions focus on semantic knowledge extraction, representation, knowledge clustering, inconsistency checking, entity relatedness and linking, query suggestions, etc. Many new approaches are proposed to investigate these topics in the context of Web-scale resources. This summary introduces the major contributions of accepted papers in the Web-KR 2014 workshop.

A growing number of applications require continuous pro-cessing of high-throughput data streams, e.g., financial anal-ysis, network traffic monitoring, or Big Data analytics for smart cities. Stream processing applications typically re-quire specific quality-of-service levels to achieve their goals; yet, due to the high time-variability of stream characteris-tics, it is often inefficient to statically allocate the resources needed to guarantee application Service Level Agreements (SLAs). In this paper, we present LAAR, a novel method for adaptive replication that trades fault tolerance for in-creased capacity during load spikes. We have implemented and validated LAAR as a middleware layer on top of IBM In-foSphere Streamsr. We have performed a wide set of exper-iments on an industrial-quality 60-core cluster deployment and we show that, under the assumption of only statistical knowledge of streams load distribution, LAAR can reduce resource consumption while guaranteeing an upper-bound on information loss in case of failures.

High-performance analytical data processing systems often run on servers with large amounts of memory. A common data structure used in such environment is the hash tables. This paper focuses on investigating efficient parallel hash algorithms for processing large-scale data. Currently, hash tables on distributed architectures are accessed one key at a time by local or remote threads while shared-memory approaches focus on accessing a single table with multiple threads. A relatively straightforward “bulk-operation” approach seems to have been neglected by researchers. In this work, using such a method, we propose a high-level parallel hashing framework, Structured Parallel Hashing, targeting efficiently processing massive data on distributed memory. We present a theoretical analysis of the proposed method and describe the design of our hashing implementations. The evaluation reveals a very interesting result - the proposed straightforward method can vastly outperform distributed hashing methods and can even offer performance comparable with approaches based on shared memory supercomputers which use specialized hardware predicates. Moreover, we characterize the performance of our hash implementations through extensive experiments, thereby allowing system developers to make a more informed choice for their high-performance applications.

Academia and industry are investigating novel approaches for processing vast amounts of data coming from enterprises, the Web, social media and sensor readings in an area that has come to be known as Big Data. Logic programming has traditionally focused on complex knowledge structures/programs. The question arises whether and how it can be applied in the context of Big Data. In this paper, we study how the well-founded semantics can be computed over huge amounts of data using mass parallelization. Specifically, we propose and evaluate a parallel approach based on the X10 programming language. Our experiments demonstrate that our approach has the ability to process up to 1 billion facts within minutes.

We present an approach to access and consolidate complex information spanning multiple specialist domains and make it available to non-experts. We are using a combination of business rules and contextual exploration to reduce interface complexity and improve consumability. We present a use case and a prototype on top of a real-world enterprise solution for coordinating Social care and Health care. We evaluate our system through a user study. Our results indicate that our approach reduces the time required to obtain business results compared to a baseline graph exploration approach.

As a continuous effort for organizing discussions and providing possible theories and techniques to deal with the barriers for knowledge processing at Web scale, the 2013 International Workshop on Web-scale Knowledge Representation, Retrieval and Reasoning (Web-KR 2013) was held in conjunction with the 2013 ACM International Conference on Information and Knowledge Management (CIKM 2013) on November 1st, 2013 at Burlingame, CA, United States. This is the 4th version of the Web-KR workshop. As in previous workshops under the same title, accepted papers of this workshop cover many important topics in the field. This year, the contributions focus on multi-faceted understanding of Web knowledge sources, Web entity linking, deep Web knowledge acquisition, and Web-scale stream reasoning. Many new approaches are proposed to deal with these problems in the context of large scale Web resources. This summary introduces the major contributions of accepted papers in the Web-KR 2013 workshop.

The rapid and perpetual growth of knowledge on the Web has given rise to many grand challenges (such as scalability, inconsistency, uncertainty, distribution and dynamics) for traditional knowledge processing methods and systems. Knowledge representation, retrieval and reasoning methods need to evolve and adapt to the Web to face these challenges and make this vast, heterogenous knowledge useful and accessible. In this light, the International Workshop on Web-scale Knowledge Representation, Retrieval, and Reasoning (Web-KR) is initiated. This workshop serves as the third one in this workshop series. This summary discusses the scope of Web-KR and introduces the advances in this field through the accepted papers in the Web-KR 2012 workshop, co-located with CIKM 2012.

Semantic Web data exhibits very skewed frequency distributions among terms. Efficient large-scale distributed reasoning methods should maintain load-balance in the face of such highly skewed distribution of input data. We show that term-based partitioning, used by most distributed reasoning approaches, has limited scalability due to load-balancing problems.We address this problem with a method for data distribution based on clustering in elastic regions. Instead of as- signing data to fixed peers, data flows semi-randomly in the network. Data items speed-date while being temporarily collocated in the same peer. We introduce a bias in the routing to allow semantically clustered neighborhoods to emerge. Our approach is self-organising, efficient and does not require any central coordination.We have implemented this method on the MaRVIN platform and have performed experiments on large real-world datasets, using a cluster of up to 64 nodes. We compute the RDFS closure over different datasets and show that our clustering algorithm drastically reduces computation time, calculating the RDFS closure of 200 million triples in 7.2 minutes.

The increasing popularity of Web services (WS) has exemplified the need for scalable and robust discovery mechanisms. Although decentralized solutions for discovering WS promise to fulfill these needs, most make limiting assumptions concerning the number of nodes and the topology of the network and rely on having information on the data a-priori (e.g. categorizations or popularity distributions). In addition, most systems are tested via simulations using artificial datasets. In this paper we introduce a lightweight, scalable and robust WSDL discovery mechanism based on real-time calculation of term popularity. In order to evaluate this mechanism, we have collected and analyzed real data from deployed WS and performed a large-scale emulation on the DAS-3 distributed supercomputer. Results show that we can achieve Web-scale service discovery based on term search and we also sketch an extension of this mechanism to support a fully-fledged WS query language.