financial times moving to linked data

“We are working on republishing the architecture on the back end, basically
engineering the whole of the backend stack to use the RDF model” for data
interchange, he says. Its work involves modeling ontologies for companies,
organizations, brands, exchanges, shares, financial instruments and other
key business terms.

It has decided not to use the Financial Industry Business Ontology
<http://www.edmcouncil.org/financialbusiness> (FIBO) model, instead
modeling its own representation of business and finance. The reason,
Rayfield says, is that the FIBO model doesn’t “really fit the news case
particularly well.”

The semantic relationships in the financial domain represented by its
ontologies “are key for us moving forward,” says Rayfield, as are APIs that
will make it possible to get data stored within its triple stores, or from
other sources to which its semantic model maps, out for use in interesting
products. For example, this can power a user’s navigation from a story
about a particular company to information the Financial Times stores about
its directors. Or, “if readers are on a story about [the company] BP, we
know its ticket from our semantic model, we go to the Data API and can get
a graph of a particular instrument from a third-party data source,” he says.

Among other semantically-infused capabilities available now are recommended
reads, based on the concept of semantic fingerprints, he explains. That is,
the Financial Times leverages a concept extraction mechanism to understand
what published stories cover and annotates them with identifiers found
within the content, which is matched against users’ reading habits to
identify what they’d likely want to peruse next. A similar approach
semantically matches readers to ads. “Semantic advertising is good In terms
of getting very targeted ads to the right people and profiles so that we
can charge more for ads,” he says. “And serving the right content to the
right people gets them to click to read more content so people stay on the
site longer.”

In the coming year, he expects the focus to be on user-facing products as a
new version of the site is being built with front-end apps to sit atop its
semantic model and APIs.

The Financial Times is using Ontotext t <http://www.ontotext.com/>echnology
for its effort, and it’s been working with the vendor on features to
improve performance. For instance, The Financial Times works across three
data centers – two in the UK and one in the US – and faced the task of
efficiently operating consistent transactions across geographies with the
ongoing changes to its triple store architecture that are based on requests
affecting its fairly large datasets. Its list of companies alone contains
about 200 million triples statements, and there are frequent changes to
those every day as company hierarchies change or businesses merge or other
transactions take place. “Business and finance changes frequently,” he says.

Semantics, he thinks, is of age in the publishing industry. “I think
editorial and business [sides] have seen the benefits of the [semantic]
approach and they want to follow it,” he says. “A number of other
publishers already are following the same model.”

http://semanticweb.com/financial-times-builds-semantic-publishing-infrastructure_b45386