Archive for the ‘Concept Maps’ Category

ConceptNet5 [Herein of Hypergraphs]

Friday, May 17th, 2013

ConceptNet5

From the website:

ConceptNet is a semantic network containing lots of things computers should know about the world, especially when understanding text written by people.

It is built from nodes representing concepts, in the form of words or short phrases of natural language, and labeled relationships between them. These are the kinds of things computers need to know to search for information better, answer questions, and understand people’s goals. If you wanted to build your own Watson, this should be a good place to start!

ConceptNet contains everyday basic knowledge:

(…)

ConceptNet 5 is a graph

To be precise, it’s a hypergraph, meaning it has edges about edges. Each statement in ConceptNet has justfications pointing to it, explaining where it comes from and how reliable the information seems to be.

Previous versions of ConceptNet has been distributed as idiosyncratic database structures plus some software to interact with them. ConceptNet 5 is not a piece of software or a database; it is a graph. It’s a set of nodes and edges, which we can represent in multiple formats including JSON. You probably know better than we do what software you want to use to interact with it!

(That said, you can have our idiosyncratic Solr index if you want, but that’s not ConceptNet, it’s just a system for quickly looking things up in ConceptNet.)

Some other interesting properties:

  • The ConceptNet graph is ID-less. Every node and assertion contains all the information necessary to identify it and no more in its URI, and does not rely on arbitrarily-assigned IDs. The advantage of this is that if multiple branches of ConceptNet are developed in multiple places, we can later merge them simply by taking the union of the nodes and edges. (And we hope for this to happen!)
  • ConceptNet supports linked data: you can download a list of links to the greater Semantic Web, via DBPedia and via RDF/OWL WordNet. For example, our concept cat is linked to the DBPedia node at http://dbpedia.org/resource/Cat.

In addition to being a data source, interesting notion of “ID-less” nodes and edges.

Information on the software setup, Solr and Python to deliver ConceptNet5 as a hypergraph is also available.

I first saw this in Max De Marzi’s Knowledge Bases in Neo4j. You will find that Max’s approach involves dumbing down the hypergraph.

Concept Maps – Pharmaceuticals

Monday, January 21st, 2013

Designing concept maps for a precise and objective description of pharmaceutical innovations by Maia Iordatii, Alain Venot and Catherine Duclos. (BMC Medical Informatics and Decision Making 2013, 13:10 doi:10.1186/1472-6947-13-10)

Abstract:

Background

When a new drug is launched onto the market, information about the new manufactured product is contained in its monograph and evaluation report published by national drug agencies. Health professionals need to be able to determine rapidly and easily whether the new manufactured product is potentially useful for their practice. There is therefore a need to identify the best way to group together and visualize the main items of information describing the nature and potential impact of the new drug. The objective of this study was to identify these items of information and to bring them together in a model that could serve as the standard for presenting the main features of new manufactured product.

Methods

We developed a preliminary conceptual model of pharmaceutical innovations, based on the knowledge of the authors. We then refined this model, using a random sample of 40 new manufactured drugs recently approved by the national drug regulatory authorities in France and covering a broad spectrum of innovations and therapeutic areas. Finally, we used another sample of 20 new manufactured drugs to determine whether the model was sufficiently comprehensive.

Results

The results of our modeling led to three sub models described as conceptual maps representing: i) the medical context for use of the new drug (indications, type of effect, therapeutical arsenal for the same indications), ii) the nature of the novelty of the new drug (new molecule, new mechanism of action, new combination, new dosage, etc.), and iii) the impact of the drug in terms of efficacy, safety and ease of use, compared with other drugs with the same indications.

Conclusions

Our model can help to standardize information about new drugs released onto the market. It is potentially useful to the pharmaceutical industry, medical journals, editors of drug databases and medical software, and national or international drug regulation agencies, as a means of describing the main properties of new pharmaceutical products. It could also used as a guide for the writing of comprehensive and objective texts summarizing the nature and interest of new manufactured product. (emphasis added)

We all design categories starting with what we know, as pointed out under methods above.

And any three authors could undertake a such a quest, with equally valid results but different terminology and perhaps even a different arrangement of concepts.

The problem isn’t the undertaking, which is a useful.

The problem is a lack of a binding between such undertakings, which enables users to migrate between such maps, as they develop over time.

A problem that topic maps offer an infrastructure to solve.