Cleaning Missing Data Based on the Bayesian Network

  • Liang Duan
  • Kun Yue
  • Wenhua Qian
  • Weiyi Liu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7901)


To guarantee the data quality, it is necessary to clean the missing data that prevalently exist in real world databases. By incorporating additional information, such as functional dependencies or integrity constraints, the correct value for each missing data item can be derived in many existing data cleaning methods. In this paper, we propose a method for cleaning the missing data item without additional information by adopting Bayesian network (BN) as the framework of the representation and inferences of probability distributions. First, we learn a Bayesian network from the complete part of the given incomplete database, called IBN. Then, we infer the probability distributions of each missing data item based on Gibbs sampling upon the IBN. Consequently, we obtain all possible values with their corresponding probability distributions (i.e., confidence degrees), by which we clean the incomplete databases. Experimental results showed the efficiency, accuracy and precision of our methods.


Missing data cleaning Bayesian network Probabilistic database Gibbs sampling Probabilistic inference 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Muller, H., Freytag, J.C.: Problems, Methods, and Challenges in Comprehensive Data Cleansing. Technical report, Humboldt-Universitat zu Berlin (2003)Google Scholar
  2. 2.
    Arasu, A., Chaudhuri, S., Chen, Z., Ganjam, K., et al.: Experiences with using Data Cleaning Technology for Bing Services. IEEE Data Engineering Bulletin, 14–23 (2012)Google Scholar
  3. 3.
    Beskales, G., Ilyas, I.F., Golab, L.: Sampling the repairs of functional dependency violations under hard constraints. PVLDB 3(1), 197–207 (2010)Google Scholar
  4. 4.
    Bohannon, P., Fan, W., Geerts, F., Jia, X., Kementsietsidis, A.: Conditional Functional Dependencies for Data Cleaning. In: Chirkova, R., Dogac, A., Ozsu, M.T., Sellis, T.K. (eds.) Proc. of ICDE 2007, Istanbul, Turkey, pp. 746–755. IEEE Computer Society (2007)Google Scholar
  5. 5.
    Chen, H., Ku, W.S., Wang, H.: Cleansing Uncertain Databases Leveraging Aggregate Constraints. In: Workshops Proc. of ICDE 2010, California, USA, pp. 128–135. IEEE Computer Society (2010)Google Scholar
  6. 6.
    Srivastava, D.: Analyzing Data Quality Using Data Auditor. In: Chen, L., Tang, C., Yang, J., Gao, Y. (eds.) WAIM 2010. LNCS, vol. 6184, pp. 1–1. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  7. 7.
    Mayfield, C., Neville, J., Prabhakar, S.: ERACER: A Database Approach for Statistical Inference and Data Cleaning. In: Elmagarmid, A.K., Agrawal, D. (eds.) Proc. of SIGMOD 2010, Indiana, USA, pp. 75–86. ACM (2010)Google Scholar
  8. 8.
    Stoyanovich, J., Davidson, S., Milo, T., Tannen, V.: Deriving Probabilistic Databases with Inference Ensembles. In: Abiteboul, S., Bohm, K., Koch, C., Tan, K.L. (eds.) Proc. of ICDE 2011, Hannover, Germany, pp. 303–314. IEEE Computer Society (2011)Google Scholar
  9. 9.
    Darwiche, A.: Modeling and Reasoning with Bayesian Networks. Cambridge University Press (2009)Google Scholar
  10. 10.
    Cheng, J., Greiner, R., Bell, D., Liu, W.: Learning Bayesian Networks from Data: An Efficient Approach Based on Information Theory. Artificial Intelligence 137(1-2), 43–90 (2002)MathSciNetzbMATHCrossRefGoogle Scholar
  11. 11.
    Russell, S., Norvig, P.: Artificial Intelligence: A Modern Approach, 3rd edn. Prentice Hall (2009)Google Scholar
  12. 12.
    Cavallo, R., Pittarelli, M.: The Theory of Probabilistic Databases. In: Stocker, P.M., Kent, W., Hammersley, P. (eds.) Proc. of VLDB 1987, Brighton, England, pp. 71–81. Morgan Kaufmann (1987)Google Scholar
  13. 13.
    Huang, J., Antova, L., Koch, C., Olteanu, D.: MayBMS: A Probabilistic Databases Management System. In: Cetintemel, U., Zdonik, S.B., Kossmann, D., Tatbul, N. (eds.) Proc. of SIGMOD 2009, Rhode Island, USA, pp. 1071–1074. ACM (2009)Google Scholar
  14. 14.
    Benjelloun, O., Sarma, A., Halevy, A., Widom, J.: ULDBs: Databases with Uncertainty and Lineage. In: Dayal, U., Whang, K.Y., Lomet, D.B., Alonso, G.A., Lohman, G.M., Kersten, M.L., Cha, S.K., Kim, Y.K. (eds.) Proc. of VLDB 2006, Seoul, Korea, pp. 953–964. Morgan Kaufmann (2006)Google Scholar
  15. 15.
    Norsys Software Corporation,
  16. 16.
    Cover, T., Thomas, J.: Elements of Information Theory. Wiley and Sons (2006)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Liang Duan
    • 1
  • Kun Yue
    • 1
  • Wenhua Qian
    • 1
  • Weiyi Liu
    • 1
  1. 1.Department of Computer Science and Engineering, School of Information Science and EngineeringYunnan UniversityKunmingChina

Personalised recommendations