Maurizio Montagnuolo presented two papers entitled “An Efficient Visual Search Engine for Cultural Broadcast Archives” and “Applying Natural Language Processing to Speech Transcriptions for Automated Analysis of Educational Video Broadcasts”.

Nowadays, Artificial Intelligence (AI) is playing an increasingly important role in many applications areas, among which media and digital entertainment, education and teaching, economics and linguistics. These and other themes were discussed at the 16th International Conference of the Italian Association for Artificial Intelligence, which was hosted by the Department of Computer Science at the University of Bari from 14th November to 17th November 2017. For the CRIT, Maurizio Montagnuolo presented two papers about visual search and natural language processing for automatic annotation and retrieval of broadcast archives. Further in detail:

On November 14th it was presented at the 11th Workshop on Artificial Intelligence for Cultural Heritage (AI*CH 2017) the paper entitled “An Efficient Visual Search Engine for Cultural Broadcast Archives” (E. Caimotti, M. Montagnuolo, A. Messina). The paper presens a work done in collaboration with the Turin Polytechnic for the design and development of a Content Retrieval Architecture (CRA) for visual content analysis and search. The architecture was designed to meet requirements given by handling large volume of video contents. Due to its flexibility, the architecture may be implemented in different demanding application, from cultural applications on smartphones to professional catalogs management on servers.

On November 16th , it was presented at the 1st Workshop on Natural Language for Artificial Intelligence (NL4AI) the result of the work done within the Città Educante project, entitled “Applying Natural Language Processing to Speech Transcriptions for Automated Analysis of Educational Video Broadcasts” (M. Montagnuolo, R. Borgotallo, S. Proscia, L. Boch).

The Città Educante project aims to develop new models of learning and teaching by exploring, developing and evaluating innovative technologies for knowledge extraction, management and sharing.

CRIT contributed to the project:
– Defining an ontology for subject representation based on the Scientific Disciplinary Sector (SSD) taxonomy used in Italy,
– Training a classification model for document categorisation based on the SSD taxonomy,
– Training a set of named entity recognisers (NER) for the identification of persons, locations and organisations from spoken documents,
– Evaluating the generated models on a dataset of more than 2,000 episodes (~340H) from the RAI “TGR Leonardo” programme.

Paper 1:
Emanuele Caimotti, Maurizio Montagnuolo and Alberto Messina, “An Efficient Visual Search Engine for Cultural Broadcast Archives”, Proceedings of the 11th Italian Workshop on Artificial Intelligence for Cultural Heritage co-located with 16th International Conference of the Italian Association for Artificial Intelligence (AI*IA 2017)

Paper 2:
Maurizio Montagnuolo, Roberto Borgotallo, Silvia Proscia, Laurent Boch, “Applying Natural Language Processing to Speech Transcriptions for Automated Analysis of Educational Video Broadcasts”, Proceedings of the 1st Workshop on Natural Language for Artificial Intelligence co-located with 16th International Conference of the Italian Association for Artificial Intelligence (AI*IA 2017), pp. 18-29

Link to ontology

Link to AI*IA 2017 website
Link to AI*CH 2017 website
Link to NL4AI 2017 website

Related Projects:

Experimental system for visual search on broadcast archives
Città Educante