My current focus is
on improving understanding of space-time dynamics of disease and using new geovisual
analytics methods to explore these dynamics. One projects utilizes
geographic information, network analysis and visualisation techniques to
acquire an improved understanding of transmission potential routes, in a
data-poor environment, such as Niger. This requires mining of data, from a
variety of sources (i.e. news articles, blogs, RSS feeds and social networks
(i.e. Twitter)) to distinguish modes and methods of disease transmission (see Measles in Niger). In conjunction with
this, I am investigating accessibility of health facilities in Niger and how
health services may be enhanced or restricted throughout the country using
different modes of transportation (foot vs. local vehicles) at different times of the year (wet vs. dry season).
I am also building on the work I did for my PhD looking at host-pathogen interactions. Using a large dataset made available from a state agency I am analysing West Nile Virus incidence across Pennsylvania, comparing similarities and differences of species in different geographic locations and how environment may play a role in diversity. In addition, and in a malaria-vector context I am investigating what temporal scale of climate data is suitable to capture transmission potential of vector-borne diseases.
Modeling and evaluating risk in all the aforementioned cases is important, but so too is understanding
information gathered from a variety of sources can be overwhelming and requires
a number of key elements that may involve collecting and synthesizing
information to gain knowledge needed to help improve understanding about a
phenomenon. Visual analytics satisfies this task by combining “automated analysis techniques with
interactive visualisations for an effective understanding, reasoning and
decision making on the basis of very large and complex datasets" (Keim et al., 2011) and is summarized in Figure 1. Essentially this is an iterative process of
information retrieval and processing that was used to improve our understanding of movement in Niger using "Sense of Place' tools developed at GeoVISTA (see Measles in Niger (3)).
Figure 1 illustrates an amendment to the visual analytics process and automatic data analysis methods used to discover
knowledge (based on Keim et al., 2011). The visual analytics process is an
iterative process that is an interaction between information, visualisations,
and models about the data, to aid users in knowledge discovery.
Thus effective visualization and clear,concise maps are necessary. This requires effective symbolization of maps through standardization of common symbols and sharing of these symbols between agencies (see Symbology). This is critical for crisis management and emergency mapping when communication is paramount.
ANALYZING AND MAPPING TWITTER DATA
I am also involved with SensePlace2 which forages place-time-attribute information from the Twitterverse that can support crisis management.
Investigating diversity and standardization of symbols across multiple agencies through the use of a repeatable process for expanding symbol sets to support new needs, and to develop new technology to support symbol sharing and dissemination. Further details can be found at Map Symbology.
Malaria in Africa
Currently working with the Thomas Lab looking at how malaria will be affected with changes in temperature. In particular, I am modeling transmission potential of malaria using different climate resolutions to understand what temporal scale is necessary to model vector-borne diseases both currently and in the future using downscaled data.
West Nile Virus in Pennsylvania
Tracking Turtles in Colombia
Currently working with CIMAD analyzing spatial movement of turtles and improving conservation in Colombia.