In 2003 Kenya abolished user fees in all government primary schools. We find that this Free Primary Education (FPE) policy resulted in a decline in public school quality and increased demand for private schooling. However, the former did not reflect a decline in value added by public schools as anticipated if fees contribute to local accountability but rather the selection of weaker pupils into...

This chapter surveys knowledge extraction approaches from structured sources such as relational databases, XML and CSV. A general definition of knowledge extraction is devised that covers structured as well as unstructured sources. We summarize current progress on conversion of structured data to RDF and OWL. As an example, we provide a formalization and description of SparqlMap, which implemen...

We address the problem of information fusion in uncertain environments. Imagine there are multiple experts building probabilistic models of the same situation and we wish to aggregate the information they provide. There are several problems we may run into by naively merging the information from each. For example, the experts may disagree on the probability of a certain event or they may disagr...

The DBpedia project has extracted a rich knowledge base from Wikipedia and serves this knowledge base as Linked Data on the Web. DBpedia’s knowledge base currently provides 274 million pieces of information about 2.6 million concepts. As DBpedia covers a wide range of domains and has a high degree of conceptual overlap with various openlicense datasets that are already available on the Web, an ...

We present a novel methodology for manipulating sources in a knowledge integration scenario. First we define and exploit an appropriate data model – Knowledge Oriented Specification – for representing and querying the data sources without having to align their background knowledge. Second we propose a structured knowledge representation formalism – Layered Conceptual Graphs – which present the ...

Systems that attempt to integrate and analyze data from multiple data sources are greatly aided by the addition of specific semantic and metadata "context" that explicitly describes what a data value means. In this paper, we describe a systematic approach to constructing models of data and their context. Our approach provides a generic "template" for constructing such models. For each data sour...

We predict discourse segment boundaries from linguistic features of utterances, using a corpus of spoken narratives as data. We present two methods for developing seg-mentation algorithms from training data: hand tuning and machine learning. When multiple types of features are used, results approach human performance on an independent test set (both methods), and using cross-validation (machine...

The objective of the Multi-Knowledge project is the development and validation of a collaborative IT platform for knowledge management, allowing geographically dispersed groups of researchers, dealing with different data sources as well as technological and organisational contexts, to create, exchange and manipulate new knowledge in a seamless fashion. The ambition is also to define a methodolo...

Ontology mediation is enabled through interoperability of semantic data sources. It helps data sharing between heterogeneous knowledgebase and reuse by semantic applications. Ontology mediation includes operations such as, mapping, alignment, matching, merging and integration. After briefly describing these operations, this study selectively discusses set of methods, tools and data integration ...