- 2012~2013 PivotLab at Purdue University
Big ratings data is hard to understand
A novel method to visualize bipartite datasets, such as user ratings of movies, so that the characteristics of a particular cluster is represented by the density of the corresponding nodes. This dyadic data is displayed as weighted bipartite graphs using scatterplots in two separated visual spaces, where each entity is positioned according to similarity in preferences. Selecting or navigating in one space is reflected in the other space, so that organic visual patterns can be formed to facilitate the understanding of underlying social groupings. We also overlay a contour plot based on kernel density estimation to uncover such groupings. We validated ParallelSpaces by implementing a movie preference visualization application. By displaying the relationship between two domains interactively, this visualization allows an exploration of preference patterns.
probabilistic Latent Semantic Analysis (pLSA)
- Movie Preference Visualization for Recommender System
- Yelp Data Visualization showing the words used to describe the businesses
- Authors, Papers and Conference Visualization for Social Network Analysis