In this paper, we present a prototype application — the Technology Trend Tracker — to measure the reality of technology trends of interest using information on the Web to inform decisions such as when to develop training, when to invest in expertise, and more. This prototype performs this task by integrating several artificial intelligence technologies in an innovative way. These technologies include rich semantic representations, a natural language understanding module, and a flexible semantic matcher. We use our system to augment Accenture's annual technology vision survey and show how our system performs well on measuring the reality of technology trends from this survey. We also show why our system performs well through an ablation study.

This paper introduces the challenge of digital preservation in the   area of engineering design and manufacturing and presents a   methodology to apply knowledge representation and semantic   techniques to develop Digital Engineering Archives.  This work   is part of an ongoing, multi-university, effort to create   Cyber-Infrastructure-Based Engineering Repositories for   Undergraduates (CIBER-U) to support engineering design education.   The technical approach is to use knowledge representation techniques   to create formal models of engineering data elements, workflows and   processes.  With these formal engineering knowledge and processes   can be captured and preserved with some guarantee of long-term   interpretability.  The paper presents examples of how the techniques   can be used to encode specific engineering information     packages and workflows.  These techniques are being integrated   into a semantic Wiki that supports the CIBER-U engineering education   activities across nine universities and involving over 3,500   students since 2006.

As the volume and interconnectedness of corporate data grows, data quality is becoming a business competency essential to success. Existing methods for managing data quality do not scale up to large volumes of data in a way that is directly manageable by the owner of the data. For the past two years a new breed of data quality products, built on applied AI techniques, are empowering non-technical users. Over 150 businesses are benefiting from these products including NASDAQ, Visa, Experian, Oracle, Fidelity, Bank of America, Volvo, Dell, Sabic, and Dassault Systems. The applied AI techniques described include lightweight ontologies to efficiently find inexact textual matches in large data sets.

The motivation of semantic wikis is to make acquisition, maintenance, and mining of formal knowledge simpler, faster, and more flexible. However, most existing semantic wikis have a very technical interface and are restricted to a relatively low level of expressivity. In this paper, we explain how AceWiki uses controlled English -- concretely Attempto Controlled English (ACE) -- to provide a natural and intuitive interface while supporting a high degree of expressivity. We introduce recent improvements of the AceWiki system and user studies that indicate that AceWiki is usable and useful.

A folksonomy refers to a collection of user-defined tags with which users describe contents published  on the Web. With the flourish of Web 2.0, folksonomies have become an important mean to develop the Semantic Web. Because tags in folksonomies are authored freely, there is a need to understand the structure and semantics of these tags in various applications. In this paper, we propose a learning approach to create an ontology that captures the hierarchical semantic structure of folksonomies. Our experimental results on two different genres of real world data sets show that our method can effectively learn the ontology structure from the folksonomies.

Conjunctive queries (CQ) are fundamental for accessing description logic (DL) knowledge bases. We study CQ answering in (extensions of) the DL EL, which is popular for large-scale ontologies and underlies the designated OWL2-EL profile of OWL2. Our main contribution is a novel approach to CQ answering that enables the use of standard relational database systems as the basis for query execution. We evaluate our approach using the IBM DB2 system, with encouraging results.

We present a novel reasoning procedure for Horn SHIQ ontologies—SHIQ ontologies that can be translated to the Horn fragment of first-order logic. In contrast to traditional reasoning procedures for ontologies, our procedure does not build models or model representations, but works by deriving new consequent axioms. The procedure is closely related to the so-called completion-based procedure for EL++ ontologies, and can be regarded as an extension thereof. In fact, our procedure is theoretically optimal for Horn SHIQ ontologies as well as for the common fragment of EL++ and SHIQ. A preliminary empirical evaluation of our procedure on large medical ontologies demonstrates a dramatic improvement over existing ontology reasoners. Specifically, our implementation allows the classification of the largest available OWL version of Galen. To the best of our knowledge no other reasoner is able to classify this ontology.

Effective communication in open environments relies on the ability of agents to reach a mutual understanding of the exchanged message by reconciling the vocabulary (ontology) used. Various approaches have considered how mutually acceptable mappings between corresponding concepts in the agents' own ontologies may be determined dynamically through argumentation-based negotiation (such as Meaning-based Argumentation). However, the complexity of this process is high, approaching π 2 (p) -complete in some cases. As reducing this complexity is non-trivial, we propose the use of ontology modularization as a means of reducing the space over which possible concepts are negotiated. The suitability of different modularization approaches as filtering mechanisms for reducing the negotiation search space is investigated, and a framework that integrates modularization with Meaning-based Argumentation is proposed. We empirically demonstrate that some modularization approaches not only reduce the number of alignments required to reach consensus, but also predict those cases where a service provider is unable to satisfy a request, without the need for negotiation.

This paper provides a decentralized data model and associated algorithms for peer data management systems (PDMS) based on the DL-liteR description logic. Our approach relies on reducing query reformulation and consistency checking for DL-liteR into reasoning in propositional logic. This enables a straightforward deployment of DL-liteR PDMSs on top of SomeWhere, a scalable propositional peer-to-peer inference system. We also show how to use the state-of-the-art Minicon algorithm for rewriting queries using views in DL-liteR in the centralized and decentralized cases.

iMatch is a probabilistic scheme for ontology matching based on Markov networks, which has several advantages over other probabilistic schemes. First, it uses undirected networks, which better supports the non-causal nature of the dependencies. Second, it handles the high computational complexity involved by approximate reasoning, rather then by ad-hoc pruning. Third, the probabilities that it uses are learned from matched data. Finally, iMatch naturally supports interactive semi-automatic matches. Experiments using the standard benchmark tests that compare our approach with the most promising existing systems show that iMatch is one of the top performers.