Abstract:
Social media and bibliographic datasets can be graphed. Clustering graphs reveals data structure. Node attributes in attributed graphs can improve clustering. Attributed graph clustering methods typically separate structural and attribute similarity.
This paper models attributed graphs as star-schema heterogeneous graphs with different types of graph nodes. This makes personalized pagerank (PPR) a unified distance measure that captures structural and attribute similarities. DBSCAN clustering and iterative edge weight updates balance attribute importance.
Distributed clustering algorithms are needed due to data volume growth. Thus, we use Blogel, a popular distributed graph computing system, to develop four exact and approximate methods for efficient PPR score computation when edge weights change.
A simple entropy-based edge weight update strategy improves clustering. We also present a game theory-based method for trading efficiency for quality. Our proposals are tested on real-world datasets.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
