Comparing the Effectiveness of Data Visualization Techniques for Discovering Disease Relationships in a Complex Network Dataset

Abstract

In this study, we compare various data visualization methods for exploring a complicated network dataset containing details on illnesses, symptoms, and safety measures. The dataset was obtained from Kaggle and split into train and test subsets at a 4:1 ratio. It has 269 nodes and 483 edges. To evaluate the network data, we used Neo4j and Gephi, two data visualization tools. The dataset was queried and visually analyzed using Neo4j, and graphical representations of the network were produced using Gephi. We tested the potency of different visualization methods for finding patterns and correlations in the data, including force-directed layouts, node-link diagrams, and matrix views. Moreover, Neo4j's querying capabilities allowed us to analyze sub-networks and their connections in greater detail. Overall, our study shows the value of using a variety of visualization methods to have a deeper understanding of complicated network data. Researchers, medical experts, and public health officials attempting to comprehend and manage illness linkages will find the findings of this study to be quite insightful.