Recent Advances in the Evaluation of Ontology Quality

Ontology or domain specific vocabulary is indispensable to a semantic web-based application; therefore, its evaluation assumes critical importance for maintaining the quality. A modular ontology is intuitively preferred to as a monolithic ontology. A good quality modular ontology, in turn, promotes reusability. This chapter is directed at summarizing the efforts towards ontology evaluation, besides defining the process of evaluation, various approaches to evaluation and underlying motivation. In particular, a modular ontology's cohesion and coupling metrics have been discussed in detail as a case study. The authors believe that the body of knowledge in this chapter will serve as a beginning point for ontology quality engineers and at the same time acquaint them with the state-of-art in this field.

Learning to Suggest Hashtags

Twitter is one of the primary online social networks where users share messages and contents of interest to those who follow their activities. To effectively categorize and give audience to their tweets, users try to append appropriate hashtags to their short messages. However, the hashtags usage is very small and very heterogeneous and users may spend a lot of time searching the appropriate hashtags. Thus, the need for a system to assist users in this task is very important to increase and homogenize the hashtagging usage. In this chapter, the authors present a hashtag recommendation system on microblogging platforms by leveraging semantic features. Furthermore, they conduct a detailed study on how the semantic-based model influences the final recommended hashtags using different ranking strategies. Moreover, they propose a linear and a machine learning based combination of these ranking strategies. The experiment results show that their approach improves content-based recommendations, achieving a recall of more than 47% on recommending 5 hashtags.

Data Replication Impact on DDBS System Performance

In distributed database systems (DDBS), the utmost purpose of data distribution and replication aims at shrinking transmission costs (TC), including communication costs, and response time. In this chapter, therefore, an enhanced heuristic clustering-based technique for data fragmentation and replicated based allocation is efficaciously presented. This work is mainly sought to further enhance an existing technique so TC is to be significantly minimized. In fact, the approached enhancement is applied by suggesting different replication scenarios. Off these scenarios, one scenario is to be selected based on competitive performance evaluation process. DDBS performance is measured via its being exposed on objective function (TC). Despite the fact that this work is mildly improved, yet evaluation results show that it has been promising, particularly as TC being the foremost design objective of DDBS System. Experimental results have been analyzed under all presented scenarios as an internal evaluation and are vividly provided to demonstrate the undeniable impact of data replication on DDBS performance.

Company Acquisition Relations Extraction From Web Pages

In this chapter, we study the problem of extracting company acquisition relation from huge amounts of webpages, and propose a novel algorithm for a company acquisition relation extraction. Our algorithm considers the tense feature of Web content and classification technology of semantic strength when extracting company acquisition relation from webpages. It first determines the tense of each sentence in a webpage, where a CRF model is employed. Then, the tense of sentences is applied to sentences classification so as to evaluate the semantic strength of the candidate sentences in describing company acquisition relation. After that, we rank the candidate acquisition relations and return the top-k company acquisition relation. We run experiments on 6144 pages crawled through Google, and measure the performance of our algorithm under different metrics. The experimental results show that our algorithm is effective in determining the tense of sentences as well as the company acquisition relation.

A Semantic Sensor Web Architecture in the Internet of Things

The Internet of Things (IoT) vision is connecting uniquely identifiable devices to the internet, best described through ontologies. Furthermore, new emerging technologies such as wireless sensor networks (WSN) are recognized as essential enabling component of the IoT today. Hence, the interest is to provide linked sensor data through the web either following the semantic web enablement (SWE) standard or the linked data approach. Likewise, a need exists to explore those data for potential hidden knowledge through data mining techniques utilized by a domain ontology. Following that rationale, a new lightweight IoT architecture has been developed. It supports linking sensors, other devices and people via a single web by mean of a device-person-activity (DPA) ontology. The architecture is validated by mean of three rich-in-semantic services: contextual data mining over WSN, semantic WSN web enablement, and linked WSN data. The architecture could be easily extensible to capture semantics of input sensor data from other domains as well.

The OSGI SIB

The research activity in the IoT field caused a proliferation of information brokers with different features and targeted at different information abstraction levels. The OSGI Semantic Information Broker (SIB) is a portable and extendable solution for providing an IoT system with semantic support, a publish subscribe paradigm, and expressive primitives for information modeling. In this chapter the authors explain the main reasons for defining a new SIB version, substituting the previously used RedSIB, its main features and comparative evaluation against both ad hoc and standard benchmarks. Furthermore, recently defined primitives and experimental work on the portability to mobile devices and resiliency are proposed and discussed.

Overview of MERA

The proliferation of large databases with potentially repeated entities across the World Wide Web drives into a generalized interest to find methods to detect duplicated entries. The heterogeneity of the data cause that generalist approaches may produce a poor performance in scenarios with distinguishing features. In this paper, we analyze the particularities of music related-databases and we describe Musical Entities Reconciliation Architecture (MERA). MERA consists of an architecture to match entries of two sources, allowing the use of extra support sources to improve the results. It makes use of semantic web technologies and it is able to adapt the matching process to the nature of each field in each database. We have implemented a prototype of MERA and compared it with a well-known music-specialized search engine. Our prototype outperforms the selected baseline in terms of accuracy.

Leveraging Context-Awareness in Duty-Cycled Broadcast Wireless Sensor Networks

Operating under duty-cycle mode allows wireless sensor networks to prolong their lifetime. However, this working pattern, with the temporary unavailability of nodes, brings challenges to the network design, mainly for a fundamental service like flooding. The challenging task is to authorize sensors to adopt a duty-cycle mode without inflicting any negative impact on the network performances. Context-awareness offers to sensors the ability to adapt their functional behavior according to many contexts in order to cope with network dynamics. In this context, the authors propose an Enhanced-Efficient Context-Aware Multi-hop Broadcast (E-ECAB) protocol, which relies on multi contextual information to optimize resources usage and satisfy the application requirements in a duty-cycled environment. The authors proved that only one transmission is required to achieve the broadcast operation in almost all situations. Simulation results show that E-ECAB achieves a significant improvement compared to previous work in terms of throughput and end-to-end delay without sacrificing energy efficiency.

Semantic Extension for the Linked Data Based on Semantically Enhanced Annotation and Reasoning

Linked Data, a new form of knowledge representation and publishing described by RDF, can provide more precise and comprehensible semantic structures. However, the current RDF Schema (RDFS) and SPARQL-based query strategy cannot fully express the semantics of RDF since they cannot unleash the implicit semantics between linked entities, so they cannot unleash the potential of Linked Data. To fill this gap, this chapter first defines a new semantic annotating and reasoning method which can extend more implicit semantics from different properties and proposes a novel general Semantically-Extended Scheme for Linked Data Sources to realize the semantic extension over the target Linked Data source. Moreover, in order to effectively return more information in the process of semantic data retrieval, we then design a new querying model which extends the SPARQL pattern. Lastly, experimental results show that our proposal has advantages over the initial Linked Data source and can return more valid results than some of the most representative similarity search methods.

Executing, Comparing, and Reusing Linked-Data-Based Recommendation Algorithms With the Allied Framework

Data published on the web following the principles of linked data has resulted in a global data space called the Web of Data. These principles led to semantically interlink and connect different resources at data level regardless their structure, authoring, location, etc. The tremendous and continuous growth of the Web of Data also implies that now it is more likely to find resources that describe real-life concepts. However, discovering and recommending relevant related resources is still an open research area. This chapter studies recommender systems that use linked data as a source containing a significant amount of available resources and their relationships useful to produce recommendations. Furthermore, it also presents a framework to deploy and execute state-of-the-art algorithms for linked data that have been re-implemented to measure and benchmark them in different application domains and without being bound to a unique dataset.