COntent Processing by Integrating Logical and Statistical Knowledge (COPILOSK) — LT World

LT World


provided by

dfki logo

with support by

eu star logofp7 logo


meta logo
clarin logo

as well as by

bmbf logo


take logo


This site uses Google Analytics to record statistics about site visits - seeĀ Legal Information.

You are here: Home kb Players & Teams Projects COntent Processing by Integrating Logical and Statistical Knowledge (COPILOSK)

COntent Processing by Integrating Logical and Statistical Knowledge (COPILOSK)

December 01, 2009

March 01, 2009

Humans solve the problems of interpreting multimedia documentation by both exploiting structural regularities in the data, as well asmaking use of common sense and specialized knowledge that explicitly represents the meaning of data. In spite of this simple observation,current approaches to processing content over the Internet are basedeither on a statistical approach, which exploits the regularities of thecontent encoded in a statistical model, or, alternatively, by logicalapproaches, by exploiting logical knowledge encoded via logical theories (such as ontologies). In both cases this unilateral approach leadsto a limitation in the performance or in the quality of the results. Webelieve that combining statistical knowledge with logical knowledgein a unique system would improve the efficiency and the effectivenessof content management applications. The present proposal aims toinvestigate how statistical and logical knowledge can be combinedand exploited in content and knowledge management, by proposingor extending a theoretical paradigm, test it in three use cases in thearea of machine translation, content extraction from text, and imageobject recognition, and in case of positive results to apply it to a realcase scenario.