IQ-DAM (Intelligent Qualitative Data Analysis and Mapping) is a novel knowledge-based, ontology-driven Qualitative-GIS system for qualitatively analysing and visualising geographical data. By “geographical data” we do not refer to the strict quantitative definition of the term but, rather, to qualitative data of various types and abstraction levels that, however, possess some geographical property (e.g., an association with a geo-location). The main novelty of IQ-DAM compared to similar Qualitative-GIS approaches is that the former attempts to exploit computational Ontologies and related semantic technologies, coupled with state-of-the-art GIS tools, in order to (a) represent the knowledge surrounding the data in question, (b) semantically analyse and reason on the input data in order to infer novel qualitative information, and (c) present the inferred data to the user in a comprehensible, structured visual form. In this way, IQ-DAM aims to assist the social scientist in formalising, investigating and drawing conclusions upon various research questions on a given research subject and within a given knowledge domain, in a way never attempted before.
To facilitate the research objectives of the Transcapes project, IQ-DAM investigates the concept of the so-called hotspots in the context of the recent migration influx to Europe. In particular, the proof-of-concept version of IQ-DAM exposes the degree of function-outsourcing in the Moria hotspot (situated on Lesbos, Greece), i.e., the extend at which the various hotspot functions have been assigned to external actors. The conceptualisation surrounding the underlying research question encapsulates the actors involved, the functions they perform and the infrastructures in which those actors are based. The automatic reasoning on the input data is conducted based on this very conceptualisation, which has been modeled and formalised into a computational ontology in order to accommodate data analysis. Specifically, IQ-DAM’s reasoning engine determines the extend at which hotspot functions are performed by actors who operate inside or on behalf of the hotspot but are actually based in external locations. Additional concepts, such as the scale at which actors perform by default and the sector they belong to, have also been included in the conceptualisation, setting the ground for more fruitful reasoning that enables the investigation of additional research questions.
The basic functionality of the proof-of-concept version of IQ-DAM can be summed up as follows: the researcher asserts information regarding actors operating in the hotspot, the functions they perform and the infrastructures those actors are based in. As a response, the system semantically analyses and reasons on the input information, and generates a visualisation in the form of a mapped radial graph (i.e., a radial graph geo-placed on a geographical map), which exposes the hotspot’s function-outsourcing extent. The center of the graph represents the hotspot, and each ray visualises an outsourced function. The visual metaphor of the visualisation is quite straightforward: the more and longer the rays, the more and more remote the outsourced functions provided by the hotspot. From there, the researcher can input additional data or modify/delete existing data, and the system responds by updating the visualisation accordingly. By clicking on the various visual elements, the user can obtain information about what exactly each element represents visually. Moreover, the user is enabled to zoom in or out of the visualisation, thereby acquiring detailed or overview information. It should be noted that this particular visualisation design (i.e., the radial graph) allows for the visualisation to also function visually at a more abstract level, that is by discarding the geographical dimension (i.e., the map) and presenting the abstract graph alone. So far, IQ-DAM has been successfully tested in Windows and Linux environment.
A more matured, more generalised version of IQ-DAM would allow for the investigation and visual representation of multiple research questions, possibly in juxtaposition with one another, but also the parallel processing of multiple geographical points of interest (hotspots or other). Such a multilevel, multi-subject functionality would generate multiple visualisations as discrete visual information layers, which could be presented either individually or simultaneously, in the latter case synthesised in a unified multilayered visualisation, thereby conveying combined information and facilitating comparative research.