DiMaC: a disguised missing data cleaning tool

ABSTRACT

In some applications such as filling in a customer information form on the web, some missing values may not be explicitly represented as such, but instead appear as potentially valid data values. Such missing values are known as disguised missing data, which may impair the quality of data analysis severely. The very limited previous studies on cleaning disguised missing data highly rely on domain background knowledge in specific applications and may not work well for the cases where the disguise values are inliers.

Recently, we have studied the problem of cleaning disguised missing data systematically, and proposed an effective heuristic approach [2]. In this paper, we present a demonstration of DiMaC, a Disguised Missing Data Cleaning tool which can find the frequently used disguise values in data sets without any domain background knowledge. In this demo, we will show (1) the critical techniques of finding suspicious disguise values; (2) the architecture and user interface of DiMaC system; (3) an empirical case study on both real and synthetic data sets, which verifies the effectiveness and the efficiency of the techniques; and (4) some challenges arising from real applications and several direction for future work.

REFERENCES

• 1: D. DesJardins. Outliers, inliers, and just plain liars - new graphical EDA+ (EDA Plus) techniques for understanding data. In Proc. SAS User's Group International Conference (SUGI26), Long Beach, CA, 2001.
• 2: Ming Hua , Jian Pei, Cleaning disguised missing data: a heuristic approach, Proceedings of the 13th ACM SIGKDD international conference on Knowledge discovery and data mining, August 12-15, 2007, San Jose, California, USA [doi>10.1145/1281192.1281294]
• 3: B. Kégl and L. Wang. Boosting on manifolds: Adaptive regularization of base classifiers. In Lawrence K. Saul, Yair Weiss, and Leon Bottou, editors, Advances in Neural Information Processing Systems 17, pages 665--672, Cambridge, MA, 2005. MIT Press.
• 4: Steffen L. Lauritzen, The EM algorithm for graphical association models with missing data, Computational Statistics & Data Analysis, v.19 n.2, p.191-201, Feb. 1995 [doi>10.1016/0167-9473(93)E0056-A]
• 5: Roderick J A Little , Donald B Rubin, Statistical analysis with missing data, John Wiley & Sons, Inc., New York, NY, 1986
• 6: R. Pearson. Mining imperfect data: Dealing with contamination and incomplete records. In Proc. 2005 SIAM Int. Conf. Data Mining, New Port Beach, CA, April 2005.
• 7: R. K. Pearson. Data mining in the face of contaminated and incomplete records. In Proc. 2002 SIAM Int. Conf. Data Mining, Arlington, VA, April 2002.
• 8: Ronald K. Pearson, The problem of disguised missing data, ACM SIGKDD Explorations Newsletter, v.8 n.1, p.83-92, June 2006 [doi>10.1145/1147234.1147247]
• 9: G. Webb. Further experimental evidence against the utility of occam's razor. The Journal of Artifial Intelligence Research, 4:397--417, 1996.

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.2 DATABASE MANAGEMENT
      H.2.8 Database applications
          Subjects: Data mining

General Terms:
Algorithms, Design, Experimentation

Keywords:
data cleaning, data quality, disguised missing data