Communications of the ACM
The Journal of Machine Learning Research
Mining correlated bursty topic patterns from coordinated text streams
Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining
Interactive, topic-based visual text summarization and analysis
Proceedings of the 18th ACM conference on Information and knowledge management
Hi-index | 0.00 |
Correlating content from multiple data fields is one of the key challenges in text mining. In this paper, we propose a visual analytics approach that leverages both content correlation analysis and interactive visualization technologies in analyzing and understanding content correlations. We have applied our work to analyzing NHAMCS data (National Hospital Ambulatory Medical Care Survey), which helps reveal healthcare-related data patterns through the correlations between unstructured data fields (e.g., cause of injury and diagnosis) and between structured and unstructured fields (e.g., gender and cause of injury).