Please use this identifier to cite or link to this item: http://10.1.7.192:80/jspui/handle/123456789/9525
Full metadata record
DC FieldValueLanguage
dc.contributor.authorSorthiya, Dipali-
dc.date.accessioned2021-01-04T08:07:04Z-
dc.date.available2021-01-04T08:07:04Z-
dc.date.issued2020-06-01-
dc.identifier.urihttp://10.1.7.192:80/jspui/handle/123456789/9525-
dc.description.abstractWith advancement in technology Machine learning algorithms used in many sectors.In that way Machine learning algorithms used in predicting future stock data. In other way machine learning algorithms used in financial sectors. Predicting of future stock market data is very challenging task due to its complex behaviours, many external factors like social media, real time data that affect the future stock prediction of the data. for predicting future stock data various techniques are used. Technical analysis, Fundamental analysis, Machine learning techniques are used for predicting future stock data. In this report various Stock market Prediction techniques like social sentiment analysis, Linear regression, Support vector regression, LSTM,K-nearest neighbours are performed on stock data. different machine learning techniques are performed on specific data-set and predict the future stock of the company. at the end in this report conclude that which technique predict near accurate to future stock data. Various hybrid method is also defined. using integration of machine learning techniques and time series method able to predict stock market data. Machine learning technique support vector regression and time series method moving average, combination of both techniques achieve the better performance than the existing some machine learning techniques. In experiment analysis all machine learning method is performed on same dataset and calculating the RMSE value of all algorithm. among all this algorithm whichever algorithm has least RMSE value that is to be predicted as best algorithm for predicting stock market data. ARIMA model is best among all the time series analysis method. ARIMA model has capability to convert non-stationary data into stationary data. other Hybrid method defined is combination of ARIMA and K-means clustering. this Hybrid method give also efficient output but the combining two model but complexity increased in Hybrid method. Various Deep Learning methods is sed in time series prediction, like LSTM,CNN,RNN, GANs.GANs used in various image processing technique but GANs is also used in time series prediction. Have explore the different GANs model for Future Stock Market Prediction.en_US
dc.publisherInstitute of Technologyen_US
dc.relation.ispartofseries18MCEC13;-
dc.subjectComputer 2018en_US
dc.subjectProject Report 2018en_US
dc.subjectComputer Project Reporten_US
dc.subjectProject Reporten_US
dc.subject18MCEen_US
dc.subject18MCECen_US
dc.subject18MCEC13en_US
dc.titleStock Market Prediction using Machine Learning and Deep Learning Techniquesen_US
dc.typeDissertationen_US
Appears in Collections:Dissertation, CE

Files in This Item:
File Description SizeFormat 
18MCEC13.pdf18MCEC131.39 MBAdobe PDFThumbnail
View/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.