Research On AdaBoost Regression Tree-based Multi-target Prediction Algorithm

With the era of big data,Accurate prediction of the needs of the target has practical significance for the development of the national economy.In recent years,Multi-target regression has become a new young field of data mining,which is closely related to Multi-label classification.Recently,The latest research in Multi-label classification field have stimulated us to apply existing algorithms to Multi-target regression which has appeared in many interesting areas,such as wind noise of vehicle components forecasting,stock price forecasting and ecological model and so on.Real word prediction problems typically involve the simultaneous prediction of multiple target variables using the same set of predictive variables.When the target variables are binary,the prediction task is called Multi-label classification,and when the target variable is real-valued,it is called Multi-target regression.In this paper,we put forward two new Multi-target regression algorithms:Multi-Target Stacking and Ensemble of Regressor Chains which are inspired by two popular Multi-label classification methods.Both MTS and ERC have the same baseline method,which are based on the single-target(ST)prediction model that is established using 100 regression tree AdaBoost iterative algorithms.However,MTS and ERC extend the input space of the second stage by adding the target prediction value of the first stage.Both methods take into account the dependencies between the target variables,Besides,ERC takes into account the order selection between targets.In addition,we also summarize the shortcomings of MTS and ERC methods,and propose the corrected versions denoted as MTS Corrected(MTSC)and ERC Corrected(ERCC).Another important contribution of this paper is the collection of data sets from 12 different domains in the real world.The six prediction methods involved in the experiment are evaluated based on 12 these data sets.Experimental results show that the ERCC algorithm performs best,and its performance is superior to the baseline ST and the most advanced multi-objective random forest(MORF)method.In addition,the modified versions of MTSC and ERCC presented significant improvements compared to MTS and ERC.