Concept lattice based composite classifiers for high predictability. (English) Zbl 1024.68028
Summary: Concept lattice model, the core structure in formal concept analysis, has been successfully applied in software engineering and knowledge discovery. This paper integrates the simple base classifier (Naïve Bayes or Nearest Neighbour) into each node of the concept lattice to form a new composite classifier. Two new classification systems are developed, CLNB and CLNN, which employ efficient constraints to search for interesting patterns and voting strategy to classify a new object. CLNB integrates the Naïve Bayes base classifier into concept nodes while CLNN incorporates the Nearest Neighbour base classifier into concept nodes. Experimental results indicate that these two composite classifiers greatly improve the accuracy of their corresponding base classifier. In addition, CLNB even outperforms three other state-of-the-art classification methods, NBTree, CBA and C4.5 Rules.
References:
| [1] | Aha D. W., Lazy Learning (1997) |
| [2] | Blanzieri, E. and Ricci, F. A minimum risk metric for nearest neighbor classification. Proceedings of the Sixteenth International Conference on Machine Learning. Bled, Slovenia. pp.22–31. |
| [3] | Breiman L., Machine Learning 24 pp 123– (1996) |
| [4] | Dasarathy B., Nearest Neighbor (NN) Norms: NN Pattern Classification Techniques (1991) |
| [5] | DOI: 10.1023/A:1007413511361 · Zbl 0892.68076 · doi:10.1023/A:1007413511361 |
| [6] | Duda R., Pattern Classification and Scene Analysis (1973) · Zbl 0277.68056 |
| [7] | Fayyad, U. M. and Irani, K. B. Multi-interval discretization of continuous-valued attributes for classification learning. Proceedings of IJCAI-93, pp. pp.1022–1027. |
| [8] | DOI: 10.1006/inco.1995.1136 · Zbl 0833.68109 · doi:10.1006/inco.1995.1136 |
| [9] | DOI: 10.1023/A:1007465528199 · Zbl 0892.68077 · doi:10.1023/A:1007465528199 |
| [10] | Ganter B., Formal Concept Analysis: Mathematical Foundations (1999) · Zbl 0909.06001 |
| [11] | Godin R., Theoretical Computer Science, Special Issue on Formal Methods in Databases and Software Engineering 133 pp 387– (1994) |
| [12] | Kohavi, R. Scaling up the accuracy of Naïve-Bayes classifiers: A decision-tree hybrid. Proceedings of the Second International Conference on Knowledge Discovery and Data Mining. Menlo Park, CA. pp.202–207. The AAAI Press. |
| [13] | Liu, B., Hsu, W. and Ma, Y. Integrating classification and association rule mining. Proceedings of the Fourth International Conference on Knowledge Discovery and Data Mining. New York, USA. pp.80–86. |
| [14] | Mephu-Nguifo, E. Galois lattice: a framework for concept learning, design, evaluation and refinement. Proceedings of 6th International Conference on Tools with Artificial Intelligence. New Orleans, USA. pp.461–467. IEEE Press. |
| [15] | Merz C. J., UCI repository of machine learning database (1996) |
| [16] | DOI: 10.1016/0031-3203(90)90123-3 · doi:10.1016/0031-3203(90)90123-3 |
| [17] | Njiwoua, P. and Mephu-Nguifo, E. Back from experimentation: a study of learning bias in LEGAL-E. Proceedings of BENELEARN-96. pp.57–68. Maastricht: University of Limburg. |
| [18] | Pasquier N., Information Systems 19 pp 33– (1999) |
| [19] | Quinlan J. R., C4.5: Programs for Machine Learning (1993) |
| [20] | Sahami, M. Learning classification rules using lattices (Extended Abstract). Proceedings of the Eighth European Conference on Machine Learning. Berlin, Germany. pp.343–346. Springe-Verlag. |
| [21] | DOI: 10.1109/TIT.1981.1056403 · Zbl 0464.62047 · doi:10.1109/TIT.1981.1056403 |
| [22] | Wille R., Ordered Sets pp 445– (1982) |
| [23] | DOI: 10.1023/A:1007613203719 · doi:10.1023/A:1007613203719 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
