|
Machine reading for the Semantic Web| old_uid | 13321 |
|---|
| title | Machine reading for the Semantic Web |
|---|
| start_date | 2014/01/24 |
|---|
| schedule | 11h-12h30 |
|---|
| online | no |
|---|
| location_info | salle 267 |
|---|
| summary | Machine reading is a mild version of natural language understanding,
aiming at extracting as many elements as possible from a text,
possibly on a large scale, and with performance appropriate to
industrial applications.
In recent years, machine reading in a pure NLP tradition has been
hybridised with semantic web techniques and standards, contributing
substantially to the productive area of “semantic technologies”.
In this seminar I make an overview of machine reading that applies, or
is used by, semantic web (in that area machine reading is better known
as “knowledge extraction from text”), and I will defend the position
that a closer collaboration and sharing of assumptions and objectives
can be mutually beneficial.
In order to show a concrete case study, I will detail one tool
designed by my group, FRED, which applies deep parsing and
computational semantics to generate well connected and linked RDF/OWL
graphs. |
|---|
| responsibles | Candito |
|---|
| |
|