Information Sciences Letters
Abstract
The stock market serves as an attractive investment venue that draws interest from a broad cross section of people. At the same time, while it continues to be a substantial source of income, it is frequently seen as one of the riskiest investing options due to the fundamental characteristics of the financial industry and several other elements that frequently escape the notice of inexperienced investors. No one can accurately forecast how well a stock will behave in the times to come, although several factors can aid in stock analysis. To determine the ideal moment to buy stocks and the specific stocks to buy, a decision support system (DSS) that incorporates market patterns, economic analyses, and tactics is thus, urgently needed. This study uses machine learning (ML) approaches to handle various issues presented by the assessment of market data. So, using the hyper-tree kernel-adaptive support vector machine (HKSVM) technique, this study introduces an automatic stock DSS to anticipate the top and bottom stock prices in the forthcoming years. The Z-score normalization method is first used in raw trading statistics to retrieve the data without repeated or redundant information. Then, by using the Latent Dirichlet Allocation (LDA) approach, feature extraction is carried out. By offering a reliable and automatic framework for research on stock trading data, the experimental findings and comparisons proved good interpretability and prediction effectiveness for the suggested HKSVM approach.
Recommended Citation
S. I. Alsalamah, M.
(2023)
"HKSVM-DSS: Novel Machine Learning-Based Approach for Decision Support System in Stock Market,"
Information Sciences Letters: Vol. 12
:
Iss.
5
, PP -.
Available at:
https://digitalcommons.aaru.edu.jo/isl/vol12/iss5/43