D-HOTM: distributed higher order text mining

  • Authors:
  • William M. Pottenger

  • Affiliations:
  • Rutgers University, Piscataway, NJ

  • Venue:
  • dg.o '07 Proceedings of the 8th annual international conference on Digital government research: bridging disciplines & domains
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

Modern data mining algorithms primarily leverage first-order relations, aiming to discover patterns such as association or classification rules only within records. This paper leverages higher order associations - associations between records as well as entities - in both classification and association rule mining algorithms to generate more accurate models. Our preliminary results suggest that by employing higher order paths, we can exploit valuable latent information for both association rule mining and classification.