Food shortages are increasing in many areas of the world. In this paper, we consider the problem of understanding the causal relationships between socioeconomic factors in a developing-world household and their risk of experiencing famine. We analyse the extent to which it is possible to predict famine in a household based on these factors, looking at a data collected from 5404 households in Uganda. To do this we use a set of causal structure learning algorithms, employed as a committee that votes on the causal relationships between the variables. We contrast prediction accuracy of famine based on feature sets suggested by our prior knowledge and by the models we learn.

Telephone-based information access has the potential to deliver a significant positive impact in the developing world. We discuss some of the most important issues that must be addressed in order to realize this potential, including matters related to resource development, automatic speech recognition, text-to-speech systems, and user-interface design. Although our main focus has been on the eleven official languages of South Africa, we believe that many of these same issues will be relevant for the application of speech technology throughout the developing world.

This paper examines first ideas on the applicability of Linked Data, in particular a subset of the Linked Open Drug Data (LODD), to connect radiology, human anatomy, and drug information for improved medical image annotation and subsequent search. One outcome of our ontology engineering methodology is the alignment between radiology-related OWL ontologies (FMA and RadLex). These can be used to provide new connections in the medicine-related linked data cloud. A use case scenario is provided that demonstrates the benefits of the approach by enabling the radiologist to query and explore related data, e.g., medical images and drugs. The diagnosis is on a special type of cancer (lymphoma).

Over the last several years, the market for speech technology has seen significant developments (Pieraccini and Huerta We learned some lessons which we use as guidelines 2005) and powerful commercial off-the-shelf solutions for in the development of multimodal dialogue systems where speech recognition (ASR) or speech synthesis (TTS). Further users can combine speech and gestures when using multiple application scenarios, more diverse and dynamic information interaction devices. In earlier projects (Wahlster 2003; Reithinger sources, and more complex prototype systems need et al. 2005) we integrated different sub-components to be addressed in the context of QA. Dialogue-based QA allows to multimodal interaction systems. Other lessons served as a user to pose questions in natural speech, followed by guidelines in the development of semantic dialogue systems answers presented in a concise form (Sonntag et al. 2007).

The friend of a friend (FOAF) vocabulary is widely used on the Web to describe individual people and their properties. Since FOAF does not require a unique ID for a person, it is not clear when two FOAF agents should be linked as co-referent, i.e., denote the same person in the world. One approach is to use the presence of inverse functional properties (e.g., foaf:mbox) as evidence that two individuals are the same. Another applies heuristics based on the string similarity of values of FOAF properties such as name and school as evidence for or against co-reference. Performance is limited, however, by many factors: non-semantic string matching, noise, changes in the world, and the lack of more sophisticated graph analytics. We describe a supervised machine learning approach that uses features defined over pairs of FOAF individuals to produce a classifier for identifying co-referent FOAF instances. We present initial results using data collected from Swoogle and other sources and describe plans for additional analysis.

Amongst the best practices that constitute linked data, one of the foremost is to use only HTTP-URIs as identifiers for RDF resources. This is so that the URI will resolve in a Linked Data browser to give information about the named resource. At the same time, Linked Data takes a resource-centric, as opposed to page-centric, approach to resolution. We argue that this approach can, in certain cases, obviate the need for insisting on HTTP-URIs. As a use of our “expanded” notion of Linked Data, we present as an example Life Science Identifiers.

A frequent topic discussed in the Linked Data community, especially when trying to outreach its values, is "What can we do with all this data ?". In this paper, we demonstrate (1) how to measure semantic distance on Linked Data in order to identify relatedness between resources, and (2) how such measures can be used to provide a new kind of self-explanatory recommendations, bringing together Linked Data and Artificial Intelligence principles, and demonstrating how intelligent agents could emerge in the realm of Linked Data.

While recommender systems can greatly enhance the user experience, the entry barriers in terms of data acquisition are very high, making it hard for new service providers to compete with existing recommendation services. This paper proposes to build open recommender systems which can utilise Linked Data to mitigate the new-user, new-item and sparsity problems of collaborative recommender systems. We describe how to aggregate data about object centred sociality from different sources and how to process it for collaborative recommendation. To demonstrate the validity of our approach, we augment the data from a closed collaborative music recommender system with Linked Data, and significantly improve its precision and recall.

Many of the benefits of structured data come about when users can re-use existing vocabularies rather than create new ones, but it is currently difficult for users to find, create, and host new vocabularies. Moreover, the value of any given vocabulary as a foundation for applications depends on the perceived certainty that the vocabulary — both its machine-readable schemas and human-readable specification documents — will remain reliably accessible over time and that its URIs will not be sold, re-purposed, or simply forgotten. This note proposes two approaches for solving these problems: one for multiple Vocabulary Hosting Services and a Vocabulary Preservation System to keep them linked together.

The Web of Data (WoD) is growing at an amazing rate and it will no longer be feasible to deal with it in a global way, by centralising the data or reasoning processes making use of that data. We believe that Computational Intelligence techniques provides the adaptiveness, robustness and scalability that will be required to exploit the full value of ever growing amounts of dynamic Semantic Web data.