Acquiring Graphic Design Knowledge with
Nonmonotonic Inductive Learning

Kazuya CHIBA1, Hayato OHWADA2 and Fumio MIZOGUCHI2

1Corporate Research Laboratories, Fuji Xerox Co., Ltd.
Email: chiba.kazuya@fujixerox.co.jp
2Faculty of Science and Engineering, Science University of Tokyo
Email: {ohwada, mizo)@ia.noda.sut.ac.jp



Abstract. In this paper, we present a new method based on nonmonotonic
learning where the Inductive Logic Programming (ILP) algorithm is used twice
and apply our method to acquire graphic design knowledge. Acquiring
design knowledge is a challenging task because such knowledge is complex and
vast. We thus focus on principles of layout and constraints that layouts must
satisfy to realize automatic layout generation. Although we do not have
negative examples in this case, we can generate them randomly by considering
that a page with just one element moved is always wrong. Our nonmonotonic
learning method introduces a new predicate for exceptions. In our method, the
ILP algorithm is executed twice, exchanging positive and negative examples.
From our experiments using magazine advertisements, we obtained rules
characterizing good layouts and containing relationships between elements.
Moreover, the experiments show that our method can learn more accurate rules
than normal ILP can.
References

[Bain 91] M. Bain and S. Muggleton. Non-monotonic learning. In D. Michie, editor, Machine
Intelligence 12:105-120. Oxford University Press, 1991.
[Boming 87] A. Borning, R. Duisberg, B. Freeman-Benson, A. Kramer, and M. Woolf.
Constraint Hierarchies. ACM OOPSLA, Oct. 1987, pp. 48-60.
[BYTE 97] BYTE, 22(12), McGraw-Hill, 1997.
[De Raedt 93] L. De Raedt and M. Bruynooghe, A theory of clausal discovery, in Proceedings
of the 13th International Joint Conference on Artificial Intelligence, pp. 1058-1063, Morgan
Kaufmann, 1993.
[Esposito 94] F. Esposito, D. Malerba and G. Semeraro. Multistrategy Learning for Document
recognition. Applied Artificial Intelligence: An International Journal, 8(1):33-84, 1994.
[Honda 94] K. Honda, C. Kato, H. Ohwada, N. Ichihara and F. Mizoguchi. A Floor Planning
System Using Constraint Logic Programming. In Proc. of The 2nd International Conference
on the Practical Applications of Prolog,1994.
[Ishiba 97] M. Ishiba et al. A Document Generation System using the Kansei Model. IPSJ
SIG Report, 97-HI-70, pp. 7 1-78, 1997. (In Japanese)
[Lieberman 93] H. Lieberman. Mondrian: A Teachable Graphical Editor. in Watch What I Do:
Programming by Demonstration, Allen Cypher, ed., MIT Press, 1993.
[Lieberman 95] H. Lieberman. The Visual Language of Experts in Graphic Design. IEEE
Symposium on Visual Languages, Darmstadt, Germany, September 1995.
[Lieberman 96] H. Lieberman. Intelligent Graphics. Communications of the ACM, 39(8):38-48,
1996.
[Muggleton 91] S. Muggleton. Inductive logic programming. New Generation Computing,
8(4):295-318, 1991.
[Muggleton 95] 5. Muggleton. Inverse entailment and Progol. New Generation Computing,
13:245-286, 1995.
[Rijsbergen 79] C. J. van Rijsbergen. Information Retrieval. chapter 7. Butterworths, 1979.
[Slattery 98] S. Slattery and M. Craven. Combining Statistical and Relational Methods for
Learning in Hypertext Domains. Proceedings of the 8th International Conference on
Inductive Logic Programming, Springer-Verlag, 1998.
[Srinivasan 92] A. Srinivasan, S. Muggleton, and M. Bain. Distinguishing exceptions from
noise in non-monotonic learning. In Proceedings of the Second Inductive Logic
Programming Workshop, pp. 97-107, Tokyo, 1992.
[Wrobel 94] S. Wrobel. Concept Formation During Interactive Theory Revision. Machine
Learning, 14:169-191, 1994.
