Luka Jovanovic, Nemanja Milutinovic, Masa Gajevic, Jelena O. Krstovic, Tarik A. Rashid, A. Petrovic
{"title":"简单递归神经网络调整的正弦余弦算法用于股市预测","authors":"Luka Jovanovic, Nemanja Milutinovic, Masa Gajevic, Jelena O. Krstovic, Tarik A. Rashid, A. Petrovic","doi":"10.1109/TELFOR56187.2022.9983694","DOIUrl":null,"url":null,"abstract":"Deep artificial neural networks have recently gained popularity in the time series forecasting literature. Recurrent neural networks’ higher suitability for this type of problem is the reason why this type of network has been chosen over other deep neural network approaches. Due to the number of parameters used the simplicity of these networks is considerable. This characteristic makes deep recurrent neural networks highly suitable for the problems of forecasting. Unfortunately, finding recurrent neural architecture for each specific task is NP-hard, therefore employment of metaheuristics is appropriate. Accordingly, the research proposed in this paper tackles tuning simple recurrent neural networks by sine cosine algorithm for stock market prediction. The proposed method’s performance was compared with other metaheuristics and validated against the Nikkei stock exchange.","PeriodicalId":277553,"journal":{"name":"2022 30th Telecommunications Forum (TELFOR)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":"{\"title\":\"Sine Cosine Algorithm for Simple recurrent neural network Tuning for Stock Market Prediction\",\"authors\":\"Luka Jovanovic, Nemanja Milutinovic, Masa Gajevic, Jelena O. Krstovic, Tarik A. Rashid, A. Petrovic\",\"doi\":\"10.1109/TELFOR56187.2022.9983694\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Deep artificial neural networks have recently gained popularity in the time series forecasting literature. Recurrent neural networks’ higher suitability for this type of problem is the reason why this type of network has been chosen over other deep neural network approaches. Due to the number of parameters used the simplicity of these networks is considerable. This characteristic makes deep recurrent neural networks highly suitable for the problems of forecasting. Unfortunately, finding recurrent neural architecture for each specific task is NP-hard, therefore employment of metaheuristics is appropriate. Accordingly, the research proposed in this paper tackles tuning simple recurrent neural networks by sine cosine algorithm for stock market prediction. The proposed method’s performance was compared with other metaheuristics and validated against the Nikkei stock exchange.\",\"PeriodicalId\":277553,\"journal\":{\"name\":\"2022 30th Telecommunications Forum (TELFOR)\",\"volume\":\"81 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-11-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"4\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 30th Telecommunications Forum (TELFOR)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/TELFOR56187.2022.9983694\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 30th Telecommunications Forum (TELFOR)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TELFOR56187.2022.9983694","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Sine Cosine Algorithm for Simple recurrent neural network Tuning for Stock Market Prediction
Deep artificial neural networks have recently gained popularity in the time series forecasting literature. Recurrent neural networks’ higher suitability for this type of problem is the reason why this type of network has been chosen over other deep neural network approaches. Due to the number of parameters used the simplicity of these networks is considerable. This characteristic makes deep recurrent neural networks highly suitable for the problems of forecasting. Unfortunately, finding recurrent neural architecture for each specific task is NP-hard, therefore employment of metaheuristics is appropriate. Accordingly, the research proposed in this paper tackles tuning simple recurrent neural networks by sine cosine algorithm for stock market prediction. The proposed method’s performance was compared with other metaheuristics and validated against the Nikkei stock exchange.