Type-1 fuzzy forecasting functions with elastic net regularization

• DOI: 10.1016/j.eswa.2022.116916
• Publication Date: 2022-03-21T18:02:19Z
• Authors: Nihat Tak; Deniz İnan
• Keywords: SELECTION; Social Sciences and Humanities; Social Sciences (SOC); Sosyal Bilimler ve Beşeri Bilimler; Sinyal İşleme; Mühendislik; ENGINEERING; 02 engineering and technology; Information Systems, Communication and Control Engineering; Bilgisayarla Görme ve Örüntü Tanıma; Yapay Zeka; Computer Science (miscellaneous); 0202 electrical engineering, electronic engineering, information engineering; Type-1 fuzzy functions; Bilgisayar Bilimi Uygulamaları; ENGINEERING, ELECTRICAL & ELECTRONIC; Computer Sciences; Elektrik ve Elektronik Mühendisliği; General Engineering; Computer Science Applications; Bilgisayar Bilimi (çeşitli); RIDGE; Physical Sciences; Ekonomi ve İş; ECONOMICS & BUSINESS; Ekonometri; Engineering and Technology; Bilgisayar Bilimi; Sosyal Bilimler (SOC); Computer Vision and Pattern Recognition; Bilgi Sistemleri, Haberleşme ve Kontrol Mühendisliği; COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE; OPERATIONS RESEARCH & MANAGEMENT SCIENCE; OPERASYON ARAŞTIRMA VE YÖNETİM BİLİMİ; Organizational Behavior and Human Resource Management; General Computer Science; Mühendislik (çeşitli); Operational Research; TIME-SERIES; Yöneylem; Management Science and Operations Research; algorithms; BİLGİSAYAR BİLİMİ, YAPAY ZEKA; Genel Mühendislik; Artificial Intelligence; Örgütsel Davranış ve İnsan Kaynakları Yönetimi; REGRESSION; Sosyal ve Beşeri Bilimler; Bilgisayar Bilimleri; Econometrics; Social Sciences & Humanities; Electrical and Electronic Engineering; Engineering, Computing & Technology (ENG); ANFIS; Genel Bilgisayar Bilimi; Engineering (miscellaneous); Non-linear forecasting; Mühendislik, Bilişim ve Teknoloji (ENG); COMPUTER SCIENCE; Elastic-net regularization; Yönetim Bilimi ve Yöneylem Araştırması; Fizik Bilimleri; Signal Processing; MÜHENDİSLİK, ELEKTRİK VE ELEKTRONİK; Mühendislik ve Teknoloji; Algoritmalar; Forecasting

Fuzzy functions have recently been used for forecasting problems. The main concepts behind a fuzzy functions are to cluster the inputs using a fuzzy clustering method and to include the obtained membership grades and their non-linear transformations as new variables in the input matrix. Then, multiple linear regression models are solved for different clusters. However, adding related variables to the input matrix leads to the multicollinearity problem. Thus, the main contribution of the proposed method is to employ an elastic net in fuzzy functions to overcome the aforementioned problem. Two regularization terms occur in an elastic net that come from the ridge and the lasso regression. These regularization terms are optimized using the nested cross-validation approach to overcome the multicollinearity problem in the fuzzy functions method. Twelve practical time-series datasets are analyzed to evaluate the performance of the proposed fuzzy functions. The outstanding performance of the proposed method has been verified in terms of root mean squared errors and mean absolute percentage errors for the selected datasets.