An up-to-date comparison of state-of-the-art classification algorithms

Abstract

Current benchmark reports of classification algorithms generally concern common classifiers and their variants but do not include many algorithms that have been introduced in recent years. Moreover, important properties such as the dependency on number of classes and features and CPU running time are typically not examined. In this paper, we carry out a comparative empirical study on both established classifiers and more recently proposed ones on 71 data sets originating from different domains, publicly available at UCI and KEEL repositories. The list of 11 algorithms studied includes Extreme Learning Machine (ELM), Sparse Representation based Classification (SRC), and Deep Learning (DL), which have not been thoroughly investigated in existing comparative studies. It is found that Stochastic Gradient Boosting Trees (GBDT) matches or exceeds the prediction performance of Support Vector Machines (SVM) and Random Forests (RF), while being the fastest algorithm in terms of prediction efficiency. ELM also yields good accuracy results, ranking in the top-5 , alongside GBDT, RF, SVM, and C4.5 but this performance varies widely across all data sets. Unsurprisingly, top accuracy performers have average or slow training time efficiency. DL is the worst performer in terms of accuracy but second fastest in prediction efficiency. SRC shows good accuracy performance but it is the slowest classifier in both training and testing.

Can Genetic Programming Do Manifold Learning Too?

“…We refrain from using the fitness function (or similar optimisation criteria) to measure the manifold "quality" so as not to introduce bias towards any specific manifold learning method. The scikit-learn [14] implementation of the Random Forest (RF) classification algorithm (with 100 trees) is used as it is a widely used algorithm with high classification accuracy, is stable across a range of datasets, and has reasonably low computational cost [21]. While other algorithms could also be compared, we found the results to be generally consistent across algorithms, and so do not include these for brevity.…”

Multi-objective genetic programming for manifold learning: balancing quality and dimensionality

“…kNN is used as an example of a very simple, distance-based classification algorithm. RF, in contrast, is much more sophisticated (using an ensemble decision-based approach) and is widely used for its high classification accuracy and applicability to a wide range of datasets [44]. We use the standard default implementations of this classifiers in the scikit-learn package [34], with k = 3 for kNN, and 100 base estimators for RF.…”

Machine learning classification: Case of Higgs boson CP state in H→ττ decay at the LHC

“…[51] can be used as a guidance. The recent study [52] collected extensive comparison of several Machine Learning algorithms. Despite not being able to contribute much to that topic, we show that for the application discussed through this paper it is indeed the case: the DNN technique by far outperforms the more classical approaches.…”