Semantic Web Jobs: Diffbot

Diffbot, a semantic start-up in Palo Alto, CA, is looking forMachine Learning Interns and Web Development Interns. According to the post, “At Diffbot, we apply computer vision techniques to web documents to extract out semantic metadata. These services are used within hundreds of products at companies such as Cisco, Evernote, StumbleUpon, and AOL. We also offer free access to our technology to developers via an open API. Internally, we are using our technology to develop the next generation semantic results engine for the web. Check out http://diffbot.com for more information about our technology and APIs.”

Machine learning interns should have experience in Java, the implementation of machine learning, and natural language processing algorithms. Web development interns should have experience with Python, Javascript, HTML5, CSS, and interaction design.

Image: Courtesy Diffbot

by http://semanticweb.com

Semantic Reasoners

A semantic reasoner, reasoning engine, rules engine, or simply a reasoner, is a piece of software able to infer logical consequences from a set of asserted facts or axioms. The notion of a semantic reasoner generalizes that of aninference engine, by providing a richer set of mechanisms to work with. The inference rules are commonly specified by means of an ontology language, and often a description language. Many reasoners use first-order predicate logic to perform reasoning; inference commonly proceeds by forward chaining and backward chaining. There are also examples of probabilistic reasoners, including Pei Wang's non-axiomatic reasoning system, Novamente's probabilistic logic network, which tip their hat to reasoning Bayesian networks, and Pronto - probabilistic description logic reasoner.

Some open - source reasoners  than can be used with Protege and OWL API are:

Pellet

Fact ++

and HermiT

Introduction to: Ontologies - semanticweb.com

Introduction to: Ontologies - semanticweb.com

Using (Installing) Jena in Netbeans

When  I first tried to use Jena in Protégé, I was quite confused... I couldn't compile my code and I wasn't sure if my code was wrong or if hadn't installed the right jars in Netbeans.... So, after searching in the web I 've found this great video (I don't know what language is this... but I managed to understand finally) that explains how to use Jena in Netbeans IDE.

GraphViz in Protégé

Hello! Today I will show you an interesting and useful  add - on  for  Protégé.

Protégé is very useful in handling ontologies. With Protégé , you can  define hierarchies between classes and with GraphViz which can be added in Protégé   you can see a graph with these hierarchies. There are two options. In  the first, you can see only the asserted model and in the second you can see the inferred model.

This is a capture  from a graph created with GraphViz in  Protégé.

Instructions for installation:

1. You download  GraphViz from here and install it.

2. In Protege you go in File->Preferences->OWL Viz 

3. You set the Dot Application Path in dot.exe in the folder that Graphviz was installed (example: C:\Program Files\Graphviz2.24\bin\dot.exe ) 

Reasoner - Pellet

Hello!

It's been a long time since my last post (once again.. :( - sorry).

Today I will introduce you Pellet, the reasoner I've been using the last 4-5 months. Pellet is compatible with Protégé (via a plug in ), Jena and OWL API.

As far as I am concerned, I am using OWL and I find Pellet very quick and handy. I've also tried HermiT but I wasn't very satisfied.

From the official site of Pellet I copy - paste this:

For applications that need to represent and reason about information using OWL, Pellet is the leading choice for systems where sound-and-complete OWL DL reasoning is essential. Pellet includes support for OWL 2 profiles including OWL 2 EL. It incorporates optimizations for nominals, conjunctive query answering, and incremental reasoning. There’s more detailed information about the architecture of the system and its features in Pellet Help.

An OWL DL reasoner like Pellet is a core component of ontology-based data management applications; if you need expertise in the use of Pellet for advanced integration or analysis applications, Clark & Parsia LLC can help in a variety of roles: consulting, application development, and OEM licensing.

Cheers,
Angeliki