The synergy of our research insights concerning the role of uncertainty and imprecision in deriving information abstractions from the underlying raw data has motivated our research focus on the development of a unifying framework of Granular Computing. Granular Computing arose as a response to the urgent need for intelligent processing of empirical data, that are now commonly available in vast quantities, into a humanly manageable abstract knowledge. In this sense, granular computing offers a landmark change from the machine-centric to human-centric approach to information and knowledge. The theoretical foundations of granular computing are exceptionally sound and involve set theory (interval mathematics), fuzzy sets, rough sets and random sets utilised for the representation of uncertainty and information abstractions.
There has been a tremendous increase in the research interest in Granular Computing since the publication of our first research monograph (Granular Computing: An Introduction, Kluwer, 2002). An important reason behind this has been a realisation that the real world does not lend itself to a manageable description/modelling using numerical data only. The vast quantities of raw data that are now available, need to be processed through intelligent abstraction, into higher-level entities, i.e. information granules. Such an approach mirrors what humans do when faced with the task of processing complex information. That is why we have highlighted in our monograph what we perceived to be the main challenge to granular computing research; namely the development of human-centric computer technologies that would overcome the limitations of the current machine-centric approaches.
The research agenda of human centricity has certainly gained visibility and prominence in the last 10 years. It is quite remarkable that the spectrum of application and research areas that have adopted information granulation as a successful strategy for dealing with information complexity covers such diverse fields as bioinformatics, image understanding, environmental monitoring, urban sustainability, to mention few most visible in the literature. In our recent book (Human-centric information processing through granular modelling, Springer, 2009) we document the milestone contributions to human-centric information processing research and demonstrate the emerging computational methods and the processing environments that arose from these research insights. The chapters, written by experts in the field, cover the fundamental methodologies, the new information processing paradigms, functional architectures of granular information processing and granular modeling applications.
However, there are more challenges that need to be faced by
the granular computing researchers. While the inspiration of human information
processing is valuable, we do not understand fully the mechanisms
for human abstraction of information. In this sense the interaction
between granular computing research and the psychology, biological and medical sciences
research needs to be enhanced to the benefit of all collaborating researchers.
Also the spectrum of applications of granular computing needs to be broadened
to include not only the science domain applications but also applications
from arts and humanities. It seems that the improvement of understanding of
the higher-level information processing that occurs in the process of
human interaction with fine arts could offer a valuable guidance regarding
the computer-based information granulation. Similarly, granular computing
research can both benefit from and contribute to research in social sciences
by formalising abstraction forming processes in social interactions.
In short, granular computing promises to be a fruitful area for
interdisciplinary research and we believe that it will underpin much
of the scientific progress in the 21-st century.