Machine Learning Techniques for Stock Prediction System: a Comprehensive Review

EasyChair Preprint 10302

Abstract

This paper presents a comprehensive analysis of stock forecast frameworks utilizing machine learning procedures. The objective of this think about is to assess the execution of different machine learning calculations in foreseeing stock costs and to distinguish the challenges and openings related with utilizing machine learning [10] in stock expectation. The paper talks about the distinctive sorts of information utilized in stock forecast frameworks, such as chronicled stock costs, budgetary news, and financial markets. It moreover portrays the different machine learning calculations that have been connected in stock expectation frameworks, counting manufactured neural systems, bolster vector machines, choice trees, and arbitrary woodlands. The paper presents a basic investigation of the precision and adequacy of these calculations in anticipating stock costs. [2] In addition, it examines the significance of highlight determination and accurate [1] and pre-processing methods for progressing the execution of machine learning models. The paper concludes with a talk on long run investigate headings in this region and the potential suggestions of utilizing machine learning in stock prediction for financial specialists, dealers, and money related examiners. [5] In general, this paper gives an important asset for analysts and professionals curious about creating and conveying stock expectation frameworks utilizing machine learning procedure such as underneath examined.

Keyphrases: Forecasting, linear regression, machine learning, stock market

@booklet{EasyChair:10302,
  author    = {Ankit Kumar Mishra and Neha Sansoya and Sahil Shrama and Maulik Jain and Pradeep Chauhan and Nitin Agrahari},
  title     = {Machine Learning Techniques for Stock Prediction System: a Comprehensive Review},
  howpublished = {EasyChair Preprint 10302},
  year      = {EasyChair, 2023}}