Bayesian networks for incomplete data analysis in form processing

  • Emilie PhilippotEmail author
  • K. C. SantoshEmail author
  • Abdel Belaïd
  • Yolande Belaïd
Original Article


In this paper, we study Bayesian network (BN) for form identification based on partially filled fields. It uses electronic ink-tracing files without having any information about form structure. Given a form format, the ink-tracing files are used to build the BN by providing the possible relationships between corresponding fields using conditional probabilities, that goes from individual fields up to the complete model construction. To simplify the BN, we sub-divide a single form into three different areas: header, body and footer, and integrate them together, where we study three fundamental BN learning algorithms: Naive, Peter & Clark and maximum weighted spanning tree. Under this framework, we validate it with a real-world industrial problem i.e., electronic note-taking in form processing. The approach provides satisfactory results, attesting the interest of BN for exploiting the incomplete form analysis problems, in particular.


Bayesian networks Electronic note-taking Form processing 


Conflict of interest

None declared.

Author Contributions

Ms. Emilie Philoppot performed experiments, under the supervision of Dr. Abdel Belaïd and Dr. Yolande Belaïd. Dr. K.C. Santosh analysed data and results, and wrote a complete paper and responses to the anonymous reviewers in addition to the supplementary results.


  1. 1.
    Belaïd A (2001) Recognition of table of contents for electronic library consulting. Int J Document Analysis Recogn 4(1):35–45CrossRefGoogle Scholar
  2. 2.
    Cho S-J, Kim JH (2003) Bayesian network modeling of hangul characters for on-line handwriting recognition. In: Proceedings of the IAPR International Conference on Document Analysis and Recognition, pp 207–2011Google Scholar
  3. 3.
    Chow C, Liu, C (1968) Approximating discrete probability distributions with dependence trees. IEEE Trans Inf Theory 14(3):462–467CrossRefzbMATHMathSciNetGoogle Scholar
  4. 4.
    Davis J, Goadrich M (2006) The relationship between precision-recall and roc curves. In: Proceedings of the International Conference on Machine Learning, pp 233–240. ACMGoogle Scholar
  5. 5.
    Denoyer L, Gallinari P (2004) Bayesian network model for semi-structured document classification. Inf Process Manage 40(5):807–827CrossRefGoogle Scholar
  6. 6.
    François O, Leray P (2006) Learning the tree augmented naive bayes classifier from incomplete datasets. In: Proceedings of European Workshop on Probabilistic Graphical Models, pp 91–98Google Scholar
  7. 7.
    Friedman N, Goldszmidt M (1996) Building classifiers using bayesian networks. In Proceedings of the national conference on Artificial intelligence, vol 2 AAAI’96, pp 1277–1284Google Scholar
  8. 8.
    Friedman N, Geiger D, Goldszmidt M, Provan G, Langley P, Smyth P (1997) Bayesian network classifiers. 29:131–163Google Scholar
  9. 9.
    Hallouli K, Likforman-Sulem L, Sigelle M, Sigelle M (2002) A comparative study between decision fusion and data fusion in markovian printed character recognition. In: Proceedings of the IAPR International Conference on Pattern Recognition, pp 147–150Google Scholar
  10. 10.
    He Y-L, Wang R, Kwong S, Wang X-Z (2014) Bayesian classifiers based on probability density estimation and their applications to simultaneous fault diagnosis. Inf Sci 259:252–268CrossRefMathSciNetGoogle Scholar
  11. 11.
    Hirayama J, Shinjo H, Takahashi T, Nagasaki T (2011a) Development of template-free form recognition system. In: Proceedings of the IAPR International Conference on Document Analysis and Recognition, pp 237–241Google Scholar
  12. 12.
    Hirayama J, Shinjo H, Takahashi T, Nagasaki T (2011b) Development of template-free form recognition system. In: Proceedings of the IAPR International Conference on Document Analysis and Recognition, pp 237–241Google Scholar
  13. 13.
    Jensen FV (1996) Introduction to Bayesian Networks. Springer, New York, Inc., Secaucus, NJ, 1st editionGoogle Scholar
  14. 14.
    Jensen FV, Lauritzen SL, Olesen KG (1990) Bayesian updating in causal probabilistic networks by local computations. Comput Stat Q 4:269–282MathSciNetGoogle Scholar
  15. 15.
    Jiang L, Zhang H, Cai Z, Su J (2005) Learning tree augmented naive bayes for ranking. In: Zhou L, Ooi B, Meng X (eds.) Database Systems for Advanced Applications volume 3453 of Lecture Notes in Computer Science. Springer Berlin Heidelberg, 688–698. Google Scholar
  16. 16.
    Jiang L, Wang D, Cai Z (2007) Scaling up the accuracy of bayesian network classifiers by m-estimate. In: Huang D-S, Heutte L, Loog M (eds.) Advanced Intelligent Computing Theories and Applications. With Aspects of Artificial Intelligence volume 4682 of Lecture Notes in Computer Science. Springer Berlin, Heidelberg, 475–484Google Scholar
  17. 17.
    Jiang L, Zhang H, Cai Z (2009) A novel bayes model: Hidden naive bayes. IEEE Trans Knowledge Data Eng 21(10):1361–1371CrossRefGoogle Scholar
  18. 18.
    Jiang L, Wang D, Cai Z (2012) Discriminatively weighted naive bayes and its application in text classification. Int J Artif Intell Tools 21(1):1250007Google Scholar
  19. 19.
    Jiang L, Cai Z, Wang D, Zhang H (2013) Bayesian citation-knn with distance weighting. Int J Mach Learn Cybernetics. doi: 10.1007/s13042-013-0152-x
  20. 20.
    Kebairi S, Taconet B, Zahour A, Ramdane S (1998) A statistical method for an automatic detection of form types. In Proceedings of International Workshop on Document Analysis Systems, pp 84–98Google Scholar
  21. 21.
    Keogh E, Pazzani M (1999) Learning augmented bayesian classifiers: A comparison of distribution-based and classification-based approaches. In: Proceedings of the seventh international workshop on artificial intelligence and statistics, pp 225–230Google Scholar
  22. 22.
    Langley P, Iba W, Thompson K (1992) An analysis of bayesian classifiers. In: AAAI, pp 223–228Google Scholar
  23. 23.
    Likforman-Sulem L, Sigelle M (2008) Recognition of degraded characters using dynamic bayesian networks. Pattern Recogn 41(10):3092–3103CrossRefzbMATHGoogle Scholar
  24. 24.
    Likforman-Sulem L, Sigelle M (2009) Combination of dynamic bayesian network classifiers for the recognition of degraded characters. In: Proceedings of the SPIE International Symposium on Document Recognition and Retrieval, pp 1–10Google Scholar
  25. 25.
    Mahjoub MA, Jayech K (2010) ndexation de structures de documents par rseaux baysiens, pp 163–178Google Scholar
  26. 26.
    Naïm PW, Leray POP, Becker A (2007) Réseaux bayésiens. EyrollesGoogle Scholar
  27. 27.
    Neapolitan R (2004) Learning Bayesian Networks. Prentice Hall, Upper Saddle RiverGoogle Scholar
  28. 28.
    Pearl J (1988) Probabilistic reasoning in intelligent systems: networks of plausible inference. Morgan Kaufmann Publishers Inc., San FranciscoGoogle Scholar
  29. 29.
    Philippot E, Belaïd Y, Belaïd A (2010) Bayesian networks learning algorithms for online form classification. In: Proceedings of the IAPR International Conference on Pattern Recognition, pp 1981–1984Google Scholar
  30. 30.
    Piwowarski B, Denoyer L, Gallinari P (20002) Un modle pour la recherche d’information sur des documents structurs. In: Journes internationales d’Analyse statistique des Donnes Textuelles (JADT)Google Scholar
  31. 31.
    Russell SJ, Norvig P (2003) Artificial Intelligence: a modern approach. Pearson EducationGoogle Scholar
  32. 32.
    Santosh KC, Nattee C, Lamiroy B (2012) Relative positioning of stroke-based clustering: a new approach to online handwritten devanagari character recognition. Int J Image Graphics 12(2):1250016CrossRefMathSciNetGoogle Scholar
  33. 33.
    Sebastiani F (2002) Machine learning in automated text categorization. ACM Computing Surveys, pp 1–47Google Scholar
  34. 34.
    Souafi-Bensafi S, Parizeau M, Lebourgeois F, Emptoz H (2002) Bayesian networks classifiers applied to documents. In: Proceedings of the IAPR International Conference on Pattern Recognition, pp 483Google Scholar
  35. 35.
    Spirtes P, Glymour C, Scheines R (2001) Causation, prediction, and search. The MIT Press, Cambridge, second editionGoogle Scholar
  36. 36.
    Subrahmanya N, Shin Y (2013) A variational bayesian framework for group feature selection. Int J Mach Learn Cybern 4(6):609–619CrossRefGoogle Scholar
  37. 37.
    Tran DC, Franco P, Ogier J-M (2010) Form recognition from ink strokes on tablet. In: Proceedings of International Workshop on Document Analysis Systems, pp 293–300Google Scholar
  38. 38.
    Verron S, Tiplica T, Kobi A (2007) Multivariate control charts with a bayesian network. In: ICINCO-ICSO, pp 228–233Google Scholar
  39. 39.
    Wang X.-Z., He Y.-L., Wang D. (2014) Non-naive bayesian classifiers for classification problems with continuous attributes. IEEE Trans Cybern 44(1):21–39CrossRefGoogle Scholar
  40. 40.
    Webb GI, Boughton JR, Wang Z (2005) Not so naive bayes: aggregating one-dependence estimators. Mach Learn 58(1):5–24CrossRefzbMATHGoogle Scholar
  41. 41.
    Weissenbacher D (2006) Bayesian network, a model for nlp? In: Eleventh Conference of the European Chapter of the Association for Computational Linguistics EACL, pp 195–198Google Scholar
  42. 42.
    Weissenbacher D, Nazarenko A (2011) Understand the effects of erroneous annotations produced by nlp pipelines, a case study on the pronominal anaphora resolution. Traitement Automatique des Langues 52(1):161–185Google Scholar
  43. 43.
    Wong ML, Leung KS (2004) An efficient data mining method for learning bayesian networks using an evolutionary algorithm-based hybrid approach. IEEE Trans Evol Comput 8(4):378–404CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  1. 1.Université de Lorraine - LORIA (UMR 7503)Vandoeuvre-lés-Nancy CedexFrance
  2. 2.Communications Engineering Branch, US National library of Medicine (NLM)National Institutes of Health (NIH)BethesdaUSA

Personalised recommendations