10

A semantic network is a graphic notation for representing knowledge in patterns of interconnected nodes and arcs. It is one way for knowledge visualization and presentation. It was used firstly for for artificial intelligence and machine translation in computer science. What is common to all semantic networks is a declarative graphic representation that can ...


6

There are a lot of different approaches that you could take. While the commenters are right, coming up with a distance metric is important, based on my own experience, finding good representations of your words/phrases is going to be significantly more important. Most of the "semantic" clustering algorithms that immediately come to mind are document level, ...


5

I don't know about robotics, but ontologies are part of the standard toolkit for modern expert systems, especially those with a natural language processing component. For example, consider the process of performing literature searches for systematic reviews in medicine. Of the millions of medical studies out there, reviewers need to find the 20 or so high-...


4

Here's a simple technique. Train an LDA using something like MALLET over your collection of texts. For each pair of documents you want to compare, obtain the topic distributions and compute the Hellinger distance between them. Things you can tweak include term weighting, the LDA hyperparameters, and the metric for comparing distributions. Term weighting ...


4

IBM's question-answering system, Watson, used the ontology DBPedia in their evidence scoring algorithms to compete (and win) the Jeopardy Challenge.


3

Ontologies abstractly defined are "representation of knowledge": the object and their relations. Basically the ontology represent the knowledge exploited by the "intelligent" agent. Now in practice, ontologies can be used either as: a structured knowledge representation for domain-specific databases a structured data format for interoperability of ...


3

Ontology gives you a power to model, reason and manage complex data systems from different domains. For example Natural Language Processing (NLP), Molecular Biology and Bioinformatic, Wireless Sensor Networks, to name a few. Considering semantic web (RDF-based) as a special use case of ontology-based systems, there are many companies and commercials on that ...


3

My understanding is that an ontology, at the end of the day, really boils down to some database storing objects/entries, where each entry has a set of known attributes as well as relationships to other entries. The more common understanding is that one can emulate1 any ontology as a theory in single sorted first order predicate logic. Because such a theory ...


3

DBpedia provides structured information from Wikipedia. It uses RDF, and is one of the most known Linked Data datasets. Also of interest (but not using information from Wikipedia) may be Freebase and OpenCyc.


3

Semantic Networks are in no way the same as ontologies. The accepted answer is therefore very misleading. Semantic networks (Quillian 1968) and semantic frames (Minsky 1975) emerged earlier than ontologies, and they were lacking a formal semantics despite the fact that they had semantic in the name. This is some kind of a fun fact among logicians: a ...


2

There is a fairly strong connection between ontologies and APIs. An ontology can be regarded roughly as a word-based API. Many are associated with web-based languages in XML part of the so-called semantic web also defined here. Tim Berners Lee, the inventor of the web, has strongly advocated the semantic web & has written on it. Here are two simple ...


1

If you want to say something about an RDF triple (i.e., an rdf:Statement), you can use reification: @prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> . @prefix voc: <https://example.com/vocabulary#> . @prefix : <https://example.com/instances#> . :Triple42 rdf:type rdf:Statement . :Triple42 rdf:subject :...


1

An entry in an ontology is typically called a 'concept'. It is a common understanding of what something is. Generally concepts are about kinds, although individual things may also be referred to as concepts. A concept is independent of various collections of related concepts (e.g. 'attributes'). Different models may allocate different collections of ...


1

Some more detail can be found p97. Here "ontologies" are basically encodings of database contents (which represent cyber attacks, incidents, and malicious/ impacted software programs) parseable by external sources, e.g. ACT-R. The event data management system keeps track of (real) attacks/ incidents, or possibly simulated ones. TENA is a Testing and Training ...


1

One good way is to use Flickr. For your project you could use available API for downloading photos with related information. If you are using Java you could use the classes on Google Codes.


Only top voted, non community-wiki answers of a minimum length are eligible