Livestock disease in Ethiopia (LDE) project is based on a systematic map dataset provided by the SEBI team, which contains 716 articles related to 44 livestock diseases of 4 species in Ethiopia from 2010 to 2019. We have created a 3D interactive website platform for the public to explore livestock diseases in Ethiopia in an engaging way, in which process we apply front-end developments (Three.js, HTML/CSS), data analysis (Python, D3.js)and 3D modelling and animation (Blender, CAD) techniques for final realizations. We would like to introduce three main rationales behind our design, including the choice of audience, the interpretations of data visualization formats and 3D modelling.
The general public is our last audience of choice for three reasons. Firstly, for funders and researchers, the LD4D team has already created a comprehensive interactive Tableau visualization based on the dataset, which can offer a solid evidence base to inform livestock policy or research decisions. Meanwhile , the lack of domain knowledge about livestock obstructed us from in-depth data research, while the shortage of related economic data stopped us from presenting more potential linkages between funding and prevalence. Although assisted by our data provider, innovation attempts for researchers and funders failed due to practical constraints. Most importantly, we found that for the general public, the end consumers of livestock industries, the communication of key findings in livestock disease fields can be insufficient, which turns into the starting point of our project.
We choose a tree-map to display the hierarchical data, comparing the differences in research focus across the eleven regions of Ethiopia and, furthermore , the disease studies in different species and their prevalence . Each diagram represents a tree structure using nested polygons , with each branch representing a species and tiled with sub-branches representing diseases . It is worth noting that the dimension of "percentage " is not representative and therefore not sufficient as evidence , so the area of the leaf node is proportional to the number of studies . Additionally , we use green , yellow , and purple to distinguish the shared relationships of diseases among four species, which are all-shared, partially shared, and unique.
The 3D pages present the first layer of information on the entire website and serve as the introduction interface for the entire storytelling, aiming to provide a visual understanding of the current state of the Ethiopian livestock industry even for non-professional users. At the same time, to better bring users into the story, the entire 3D model is based on mountainous plateaus and deserts, referencing the real Ethiopian environment, visualizing the number of livestock on a 3D Ethiopian ranch . Through interactive operations , users can view the number of livestock in each state and get an overview of the distribution of livestock in Ethiopia , making it more intuitive, clear and immersive.
explore more in https://miyinan.github.io/livestock-disease-ethiopia/