Learning from Patterns: Information Retrieval and Visualisation Issues Between Bioimage Informatics and Digital Humanities

The large amount of data generated in different fields, among which bioimage informatics and digital humanities, is increasingly requiring appropriate automatic processing techniques, such as computer vision, data mining and particular visualisation tools, to extract information out of complexity and to clearly display it. This has led, in digital humanities, to the use of pattern recognition techniques similar to those applied in biology, chemistry and medical studies, but where patterns to be analysed and segmented are extracted from texts, images, audiovisual and online media rather than from cells and tissues. Regularities can be recognised through machine learning, based on artificial neural networks that are modelled, to some extent, after the brain's structure, showing a variety of analogies between natural and artificial world. These processes can also add information to 3D models for cultural heritage: data mining technologies allow information retrieval from archives and repositories, as well as the comparison of data in order to better understand the context of-and relationships between-works of art, thus producing knowledge enhancement. Various tools to describe complexity are here analysed not only for their educational aim, but also for their heuristic value, allowing new discoveries and connections between different disciplines.