Looking for all addresses of potential customers in your business area? Do you want to have a list of all hotels in your city that is always
up-to-date? Do you want to automatically digitalise your collection of articles
ParaBotS might have the software components you need.
Text needs interpretation in order to become useful. A lot of the information
available on the Internet however is presented in an unstructured way, which
makes it difficult to interpret it automatically.
ParaBotS' information extraction components are designed to make sense from
these unstructured texts.
WebID is an application used in many European countries to extract
addresses from web pages and other documents. With minor alterations
WebId can be used in new countries with other address formats.
We also create software that structures newspaper advertisements and
interprets resumes. Adaptive techniques are used to learn how to extract
new information from the annotation of examples. This technology is
applied in various different domains like ontology filling, and the
extraction of semi-structured information, like the opening times of
restaurants, departure times of trains, hotel information, and resumes.
This information is described in a wide variety of ways, but the
annotation of some examples may suffice for the automated extraction
using adaptive techniques.
The opinion mining component is one of the latest additions to our software base. It searches documents for affective utterances about persons or companies. This component is used in the Vox-pop.nl webmonitor where it deduces the general opinion of bloggers and forum participants about e.g. politicians or companies or even about products or groups of people connected by philosophy of life or ethnicity.