Today I wish to share with you some details about a new project I’ll be involved in throughout the next three years. The project is called CUbRIK and addresses advanced search techniques and tools for multimedia and multimodal search, based on content-intensive, time-intensive and human-intensive scalable processes.
The project aims at experimenting the integration of machine tasks and a variety of human and crowd-sourced tasks, orchestrated by programmable pipelines. The obtained processes will allow content processing, query submission and result computation, relevance feedback processing.
CUbRIK will allow complex search on the web and on enterprise content, by encompassing entity-, time-, and space- aware search functions. This means that people will be able to search for multimedia contents that include specific reference to objects, temporal information or locations. This will be possible because CUbRIK will exploit highly accurate repositories of spatio-temporal entities (locations, events, trends, and so on), correlated through rich semantic associations.
I will participate in the joint research unit of Politecnico di Milano and WebRatio on this, lead by Piero Fraternali.
The project is meant to be the continuation of a successful experience with the Pharos research project. The project will have significant synergies with the Search Computing project and with the topics of crowdsourcing and social web addressed within the BPM4People project.
Here is an introduction video on CUbRIK: