M.A. Thesis:

Choosing optimal subset of classifiers in ensemble classification using cuckoo optimization algorithm

Classification is one of the main subdivisions of data mining and machine learning. Classification involves examining the properties of a new object and assigning it to one of the predefined sets. Recently, ensemble classification methods have drawn more attention to the data mining and machine learning community. In most cases, the performance of a set is better than a single classifier. The results of the studies show that when some basic clusters are created some of them can be pruned without reducing the final result and can even improve the results. An important challenge in this section is how to select the appropriate subset from the basic clauses. In this research, using one of the newest meta-heuristic methods called cuckoo Optimization algorithm is used to find a suitable subset of classifiers in a ensemble classification and build a good high-performance logic model. The cuckoo search metadata algorithm is able to search the answer space in a short time and find the answers in the neighborhood of the exact answer, which is very close to the exact answer.

Keywords: Classification, Ensemble Classifier, Cuckoo Optimization Algorithm, select classifiers