locally optimal solution in approaches like support
vector machine, and neural network. Besides, most
approaches are limited by the cost of the training
process and require a dataset with high quality and
large scale.
In particular, approaches are faced with especial
challenges. The linear regression cannot perform well
in long-term prediction. (Zhang & Yang, 2023)
Consequently, it takes much time for the model to
train and adapt to the new data. The K-means
Clustering is sensitive to the initial cluster centers.
(Renugadevi et al., 2016) Different initializations
could lead to different results of classification. The
Artificial Neural Network can also be greatly
impacted by the initialization. Besides, it gets harder
to explain how the model works, as the network
structure of the model becomes more and more
complicated.
Socially, machine learning is still not widely put
into use nowadays, despite the boom in machine
learning algorithms, and the effectiveness of these
techniques in stock forecasting. Corporate and
individual investors still rely more on the past
experience and subjective judgment of individual
analysts in their forecasting and analysis of the stock
market. Unlike machine learning, which is an
objective analysis based on the characteristics of the
data. There is still a long way before machine learning
technology is widely accepted and adopted by the
market.
4 CONCLUSIONS
Generally speaking, the Stock market is both an
important indicator and an influential factor. Having
reviewed the machine learning technologies of stock
prediction, the approaches developed and put into use
are exposed systematically, including regression,
support vector machine, clustering, artificial neural
network, and naive Bayes. All these approaches are
proven to be effective for their analysis based on
principles of statistics and exploitation of data
features. That is the main reason these methods are
gradually becoming more widely used.
However, the machine learning approaches have
encountered challenges, despite the great progress
researchers have made. On one hand, challenges such
as locally optimal solutions, running costs, and the
requirement of high-quality datasets prevent
algorithms from high precision and efficiency, which
limits the model performance. On the other hand, the
actual application of stock analysis and prediction is
still mostly based on a subjective judgment of
individual analysts according to traditional methods
like fundamental analysis. That means acceptance of
new technologies is still on the way to be improved.
REFERENCES
Bhosle, S., Galande, S., 2023. Development of Stock
Market Analysis and Prediction Methods Based on
Support Vector Machine Algorithm. Bryan House
Publishing, 5(11).
Elmahgop, F. O., Sayed, O.A., 2020. The Effect of Inflation
Rates on Stock Market Returns in Sudan: The Linear
Autoregressive Distributed Lag Model. Asian
Economic and Financial Review, 10(7), 808–815.
Geyer-Klingeberg, J., Hang, M., Walter, M., Rathgeber, A.,
2018. Do stock markets react to soccer games? A meta-
regression analysis. Economics, 50:19, 2171-2189.
Huang, S., 2024. Big data processing and analysis platform
based on deep neural network model. Systems and Soft
Computing, 6.
Liu, L., Dong, M., 2024. Research on Stock Trend
Prediction Method in Financial Markets Based on
Support Vector Machines, Francis Academic Press,
6(6).
Ma, J., 2024. Regression and Classification in Stock Price
Trend Forecasting. Computer Knowledge and
Technology, 20(12), 12-14+23.
Pesaran, M. H., Shin, Y., Smith, R. J., 2001. Bounds
Testing Approaches to the Analysis of Level
Relationships.Journal of Applied Econometrics, 16(3),
289-326.
Renugadevi, T., Ezhilarasie, R., Sujatha, M.,
Umamakeswari, A., 2016. Stock Market Prediction
using Hierarchical Agglomerative and K-Means
Clustering Algorithm. Indian Journal of Science and
Technology, 9(48).
Rouf, N., Malik, M. B., Arif, T., Sharma, S., Singh, S.,
Aich, S., Kim, H.-C., 2021. Stock Market Prediction
Using Machine Learning Techniques: A Decade
Survey on Methodologies, Recent Developments, and
Future Directions. Electronics, 10, 2717
Sáenz, J. V., Quiroga, F. M., Bariviera, A. F., 2023. Data
vs. information: Using clustering techniques to enhance
stock returns forecasting. International Review of
Financial Analysis, 88.
Wang, C., 2024. Stock return prediction with multiple
measures using neural network models. Financ Innov,
10(72).
Zhou, Y., 2021. A Review of Economic Growth and Stock
Returns. Economy and Trade Updates, (05), 62-65.
Zhang, Y., Yang, Y., 2023.Chinese stock price prediction
under COVID-19 period based on linear Regression
Model. (eds.) Proceedings of the 5th International
Conference on Computing and Data Science(part2)
(pp.285-290). The department of artificial intelligence,
Beijing university of chemical technology; The
department of computer science, Xiamen university of
technology.