EVALA: Links Analysis and Impact Evaluation Platform

  • Completed
  • Programme: Other
  • Start date: 16.12.2016
  • End date: 15.12.2019

EVALA (Links Analysis and Impact Evaluation Platform) is a Eurostars-2 project that aims to create a platform for media content analysis and impact evaluation. It visually represents automatically and dynamically analyzed entity correlations, drawing real-time associative semantic graphs on a cognitive map of entities – much like the human brain. Based on frequency, proximity and volume of mentions rather than just prior ontological knowledge, its focus is on evaluating causal relations between entities.

Identifying and presenting these relationships in a graph and additionally enriching them with linked data leads to e-discovery of hidden patterns, factors and models, influencer identification and stakeholder analysis.

Contact: Zlatina Marinova, Nicola Rusinov

Project Overview

The EVALA project’s goal is to demonstrate an easier way of transforming massive amounts of media content into knowledge providing animated dashboards. These represent the variety and intensity of interconnectedness between entities and how they change over time. The idea of the project is also to show that such solutions can be used as a practical and effective instrument for information extraction and analysis in Knowledge Management. The new product targets a wide range of decision-makers from all over the world in both private and public sectors.

Ontotext’s partners in this project are Videntifier and Semantic Interactive.

Ontotext’s Role

The Links Analysis and Impact Evaluation Platform combines and use a set of existing state-of-the-art technologies in an innovative way for automated gathering and heterogeneous source processing, ontology modeling, entity recognition in text, object recognition in video (and images), and interactive visual representation.

As an established еxpert in semantic technologies, Ontotext is responsible for:

  • Structuring, clustering and classifying gathered content with the means of task-specific natural language processing techniques and semantic annotation;
  • Building the semantic map (“blots” of data and metadata);
  • Providing semantic search on the gathered content based on the semantic annotation and enhanced with the capabilities of the Ontotext’s own RDF graph database that infers facts and meaning.

Ontotext Newsletter