Tommaso Di Noia

Sisinf Lab logo

 
The advent of the Linked Open Data (LOD) initiative gave birth to a variety of open knowledge bases freely accessible on the Web. They provide a valuable source of information that can improve conventional recommender systems, if properly exploited. The evaluation of recommender systems has been, and still is, a fundamental topic in the field. While, several datasets are available for evaluating the performance of collaborative filtering and content-based techniques, for semantic-aware and LOD based recommender systems those datasets are missing. Specifically, these semantic-aware techniques rely on the availability of semantic descriptions about the items and/or users in the system. We provide mappings between the items in well known datasets to DBpedia resources. This can allows practitioners in the field to evaluate and compare their algorithms with existing approaches.

Take a look @ http://sisinflab.poliba.it/semanticweb/lod/recsys/datasets/

 

 
CinemAppy programs your evenings at the cinema for you! CinemAppy will help you to choose not only the best film to watch for you and for your friends but it will also suggest the best cinema to see it. The recommendation comes also with a list of points of interest such as pubs, restaurants, pizzerias so that you can organize your perfect evening!

WATCH A DEMO VIDEO

Try it @ http://sisinflab.poliba.it/cinemappy/

 

 
NOT is a proof-of-concept application for the creation of semantic-based annotation tools. By exploiting the information encoded in DBpedia, NOT is able to suggest related tags to the user and, once selected, to semantically expand them. NOT can be embedded in third-parties applications. To see NOT in action, take a look at http://sisinflab.poliba.it/impakt-revolution/

WATCH A DEMO VIDEO

Try it @ http://sisinflab.poliba.it/not-only-tags/

 

 
LED leverages information available in DBpedia to suggest new terms related to a user query for query refinement and exploratory browsing tasks. Via an intuitive interface, the user is allowed to explore the knowledge domain related to a given query and to refine this latter by using the terms suggested by the system.

WATCH A DEMO VIDEO

Try it @ http://sisinflab.poliba.it/led/

 

 
This is a Facebook application for movie recommendation. The content-based recommender engine behind MORE is fed exclusively with Linked Open Data. All the information related to the domain of movie is retrieved from RDF datasets and used to recommend movies to the end user based on the items she previously selected. The knowledge available in the datasets is also exploited to provide explanations on the reason why a movie has been suggested.

WATCH A DEMO VIDEO

Try it @ http://apps.facebook.com/movie-recommendation/

 

 
IMPAKT-revolution is Web tool to created RDFa-enabled versions of a Curriculum Vitae. It exploits the data coming from DBpedia to allow users to semantically annotate sections of their CV via RDFa.

WATCH A DEMO VIDEO

Try it @ http://sisinflab.poliba.it/impakt-revolution/

 

 
The Semantic Marketplace is a Conversational Recommender System for the Semantic Web. It uses non-standard reasoning tasks over OWL knowledge bases to suggest relevant items to the users and guide them through the query refinement process.

WATCH A DEMO VIDEO

Try it @ http://sisinflab.poliba.it/marketplace/

 

 
OWLed2 is a Knowledge Base Management System for OWL DL developed by Data Over Ontological Models s.r.l. in cooperation with Information Systems Laboratory at Politecnico di Bari (Technical University of Bari).

OWLed2 is not just another Ontology Editor; it embeds and uses the reasoner Pellet in order to manage a knowledge base written in OWL DL. With OWLed2 it is possible both to edit an OWL DL knowledge base and to query it in order to extract new information from explicitly stated one. Click here to see OWLed2 in action on the ontology of Pizzas.

OWLed2 is very easy to use thank to its intensive use of Drag and Drop as well as of an internal clipboard to Cut/Copy/Paste OWL DL expressions in all editable areas. With OWLed2 you can write OWL DL concept expressions with no restriction. For instance you can:

  • write a concept expression as Domain or Range of a property
  • use external XML schema to define new range of a DatatypePropery
  • easily edit axioms
  • create individuals as instances of an anonymous class (class expression)
  • write complex formulas without using your keyboard.
  • perform many many other tasks in a very intuitive way.
Click here to see an overview of OWLed2 editing functionalities.

Try it @ OWLed2

 

 

MAMAS-tng (MAtch MAking Service - The Next Generation) is a Description Logics reasoning engine, in a subset of OWL-DL (ALN), endowed both of standard reasoning services and some unique non-standard inferences.

MAMAS-tng is a multi-user, multi-ontology Java servlet based system and exposes an exended DIG 1.1 interface. It is available as an HTTP service at: http://dee227.poliba.it:8080/MAMAS-tng/DIG. MAMAS-tng has been applied in several fields where a semantic-based Resource Retrieval is needed such as Conversational Recommender Systems.

Non-standard services exposed by MaMaS-tng include: Concept Abduction, Concept Contraction, Rank Potential, Rank Partial, Match Type detection, Concept Covering

Available @ MAMAS-tng
 

 

OwlEd is a Multi Platform OWL manager able to exploit innovative features exposed by MaMaS-tng

OwlEd supports standard inference services of DL reasoners, namely subsumption, satisfiability, equivalence and disjointness for concept expressions, instance checking, retrieval and realization. Furthermore, as its design has been specifically oriented towards semantic matchmaking services it also features non-standard services supported by MaMaS-tng.

Available @ OWLed
 

Valid XHTML 1.1   Valid CSS

:: hosted by SisInf Lab http://sisinflab.poliba.it ::