Feature Selection in Learning Common Sense Associations Using Matrix Factorization

Abstract

We propose a computational model to learn the common sense association between a pair of concept classes based on a bipartite network and matrix factorization methods. We view the concept-pair association as a bipartite network so that the autoassociation mappings can become similarity constraints. We impose the additional similarity and regularity constraints on the optimization objectives so that a mapping matrix can be found in the matrix factorization to best fit the observation data. We extract 139 locations and 436 activities and 667 location–activity pairs from ConceptNet (http://conceptnet5.media.mit.edu/). We evaluate the performance in terms of F-factor, precision, and recall using a common sense association problem between locations and activities against four feature selection strategies in the matrix factorization optimization. The comparison between the performances with and without human judgment reveals that matrix factorization method tends to show good generalization even under very little observation evidence. Among the four feature selection methods, the maximal entropy method performs better in terms of F-score and recall when the feature number is more than 30 % while SVD method performs better in terms of F-score and recall when the feature number is less than 30 %. Random selection can have a higher precision given “enough” features, but it tends to be the worst performer in the recall and F-score.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

References

  1. 1.

    Brewka, G.: Nonmonotonic Reasoning: Logical Foundations of Commonsense. Cambridge University Press, Cambridge (1991)

    Google Scholar 

  2. 2.

    Iwasaki, Y.: Real-world applications of qualitative reasoning. IEEE Expert 12(3), 16–21 (1997)

    Article  Google Scholar 

  3. 3.

    De Kleer, J., Brown, J.S.: A qualitative physics based on confluences. Artif Intell 24(1–3), 7–83 (1984)

    Article  Google Scholar 

  4. 4.

    Sloman, S.A.: The empirical case for two systems of reasoning. Psychol Bull 119(1), 3–22 (1996)

    Article  Google Scholar 

  5. 5.

    Bruck, J.: On the convergence properties of the hopfield model. In: Proceedings of the IEEE, 78(10), October (1990)

  6. 6.

    Kosko, B.: Bidirectional associative memory. IEEE Trans. Syst. Man Cybern. 18(1), 49–60 (1988)

    MathSciNet  Article  Google Scholar 

  7. 7.

    Liou, C.-Y., Yuan, S.-K.: Error tolerant associative memory. Biol. Cybern. 81, 331–342 (1999)

    Article  MATH  Google Scholar 

  8. 8.

    Koren, Y., Bell, R., Volinsky, C.: Matrix factorization techniques for recommender systems. IEEE Comput. 42(8), 30–37 (2009)

    Article  Google Scholar 

  9. 9.

    Lee, D.D., Seung, H.S.: Learning the parts of objects by non-negative matrix factorization. Nature 401, 788–791 (1999)

    Article  Google Scholar 

  10. 10.

    Chen, T.-C., Soo, V.-W.: Learning common sense associations based on a bipartite network using matrix factorization. In: Proceedings of ISIS (2015)

  11. 11.

    https://en.wikipedia.org/wiki/Feature_selection

  12. 12.

    ConceptNet, http://conceptnet5.media.mit.edu/

  13. 13.

    Tremblay, C., et al.: Bidirectional associative memory and learning of nonlinearly separable tasks. In: Proceedings of ICCM, (2013)

  14. 14.

    Acevedo-Mosqueda, M.E., Yanez-Marquez, C., Acevedo-Mosqueda, M.A.: Bidirectional associative memories: different approaches. ACM Comput. Surv. 45(2), 18:1–18:30 (2013)

  15. 15.

    Vedantam, R., et al.: Learning common sense through visual abstraction. In: ICCV (2015)

  16. 16.

    Zang, L.-J., et al.: A survey of commonsense knowledge acquisition. J. Comput. Sci. Technol. 28(4), 689–719 (2013)

    MathSciNet  Article  MATH  Google Scholar 

Download references

Acknowledgments

This research is supported in part by Ministry of Science and Technology of Taiwan, ROC under the Grant Number 104-2221-E007-094-MY2.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Von-Wun Soo.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Chen, TC., Soo, VW. Feature Selection in Learning Common Sense Associations Using Matrix Factorization. Int. J. Fuzzy Syst. 19, 1217–1226 (2017). https://doi.org/10.1007/s40815-016-0235-4

Download citation

Keywords

  • Common sense
  • Association reasoning
  • Matrix factorization
  • Bipartite network
  • Active learning
  • Feature selection
  • Location and actions
  • Entropy