Hybrid support vector machine rule extraction method for discovering the preferences of stock market investors: Evidence from Montenegro

Kašćelan, Ljiljana; Kašćelan, Vladimir; Jovanovic, Mladjan

doi:10.1080/10798587.2014.971500

Cited by 8 publications

(7 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The smaller depth, the larger the minimum size for the split, the larger the leaf size and the higher minimum gain, lead to less complex tree, but also a tree with smaller accuracy. SVM rule extraction is not a new method in literature and it was applied in some previous economic studies [36,38,42], but for the topic of direct marketing, i.e., to solve the problem of the minor class of the most valuable customers in customer classification, it is applied for the first time in our research in this way.…”

Section: Svm Rule Extraction Methodsmentioning

confidence: 99%

Class balancing in customer segments classification using support vector machine rule extraction and ensemble learning

Rogić

Kašćelan

2021

ComSIS

View full text Add to dashboard Cite

An objective and data-based market segmentation is a precondition for efficient targeting in direct marketing campaigns. The role of customer segments classification in direct marketing is to predict the segment of most valuable customers who is likely to respond to a campaign based on previous purchasing behavior. A good-performing predictive model can significantly increase revenue, but also, reduce unnecessary marketing campaign costs. As this segment of customers is generally the smallest, most classification methods lead to misclassification of the minor class. To overcome this problem, this paper proposes a class balancing approach based on Support Vector Machine-Rule Extraction (SVM-RE) and ensemble learning. Additionally, this approach allows for rule extraction, which can describe and explain different customer segments. Using a customer base from a company?s direct marketing campaigns, the proposed approach is compared to other data balancing methods in terms of overall prediction accuracy, recall and precision for the minor class, as well as profitability of the campaign. It was found that the method performs better than other compared class balancing methods in terms of all mentioned criteria. Finally, the results confirm the superiority of the ensemble SVM method as a preprocessor, which effectively balances data in the process of customer segments classification

show abstract

Section: Svm Rule Extraction Methodsmentioning

confidence: 99%

Class balancing in customer segments classification using support vector machine rule extraction and ensemble learning

Rogić

Kašćelan

2021

ComSIS

View full text Add to dashboard Cite

show abstract

“…Support vector machines, artificial neural networks, hybrid mechanisms, optimization and ensemble methods are among the methods that are surveyed. Kaelan et al [14] developed a method that uses support vector machine and decision tree models to extract rules for stock market decisionmakers. Chen et al [5] developed a method that uses artificial neural network model to predict the Taiwan Stock Index.…”

Section: Machine Learning and Evolutionary Solutionsmentioning

confidence: 99%

“…Machine learning techniques are used in different areas such as image/video processing, classification and recognition processes, natural language processing, and time series data analysis. ML approaches can also be used to analyze financial time-series data [28,24,14,8]. Cavalcante et al [3] reviewed all possible prediction model techniques that are used in financial time-series data.…”

Section: Machine Learning and Evolutionary Solutionsmentioning

confidence: 99%

Financial Trading Model with Stock Bar Chart Image Time Series with Deep Convolutional Neural Networks

Sezer¹,

Özbayoğlu²

2018

AUTOSOFT

View full text Add to dashboard Cite

Even though computational intelligence techniques have been extensively utilized in financial trading systems, almost all developed models use the time series data for price prediction or identifying buy-sell points. However, in this study we decided to use 2-D stock bar chart images directly without introducing any additional time series associated with the underlying stock. We propose a novel algorithmic trading model CNN-BI (Convolutional Neural Network with Bar Images) using a 2-D Convolutional Neural Network. We generated 2-D images of sliding windows of 30-day bar charts for Dow 30 stocks and trained a deep Convolutional Neural Network (CNN) model for our algorithmic trading model. We tested our model separately between 2007-2012 and 2012-2017 for representing different market conditions. The results indicate that the model was able to outperform Buy and Hold strategy, especially in trendless or bear markets. Since this is a preliminary study and probably one of the first attempts using such an unconventional approach, there is always potential for improvement. Overall, the results are promising and the model might be integrated as part of an ensemble trading model combined with different strategies.

show abstract

“…Comparing the "FRFE + WTT + SVM" in Table 6 with "FRFE + WTT + NBC" in Table 5, another finding is SVM is superior to NBC. The reason is SVM works well for large dimensional problems with relative few instances due to its regularization form [58].…”

Section: Comparison With State-of-the-artmentioning

confidence: 99%

Pathological Brain Detection by a Novel Image Feature—Fractional Fourier Entropy

Wang

Zhang

Yang

et al. 2015

Entropy

View full text Add to dashboard Cite

Aim: To detect pathological brain conditions early is a core procedure for patients so as to have enough time for treatment. Traditional manual detection is either cumbersome, or expensive, or time-consuming. We aim to offer a system that can automatically identify pathological brain images in this paper. Method: We propose a novel image feature, viz., Fractional Fourier Entropy (FRFE), which is based on the combination of Fractional Fourier Transform (FRFT) and Shannon entropy. Afterwards, the Welch's t-test (WTT) and Mahalanobis distance (MD) were harnessed to select distinguishing features. Finally, we introduced an advanced classifier: twin support vector machine (TSVM). Results: A 10ˆK-fold stratified cross validation test showed that this proposed "FRFE + WTT + TSVM" yielded an accuracy of 100.00%, 100.00%, and 99.57% on datasets that contained 66, 160, and 255 brain images, respectively. Conclusions: The proposed "FRFE + WTT + TSVM" method is superior to 20 state-of-the-art methods.

show abstract

Hybrid support vector machine rule extraction method for discovering the preferences of stock market investors: Evidence from Montenegro

Cited by 8 publications

References 35 publications

Class balancing in customer segments classification using support vector machine rule extraction and ensemble learning

Class balancing in customer segments classification using support vector machine rule extraction and ensemble learning

Financial Trading Model with Stock Bar Chart Image Time Series with Deep Convolutional Neural Networks

Pathological Brain Detection by a Novel Image Feature—Fractional Fourier Entropy

Contact Info

Product

Resources

About