Scholarlink

<p>The study highlights the significance of cybersecurity in defending significant organizations' sensitive data while confronting the real issue of increasing cyber dangers in the banking sector. Therefore, the study hopes to improve the capability of identifying and mitigating the increasing cases of cyber threats, including DDoS and Malware attacks, utilizing a mix of machine learning algorithms in the context of R Studio. Therefore, the study demonstrates how enhanced big data and proactive analytical models support risk evaluation and offer trustworthy defenses against cyber threats that raise customer confidence in financial services. With the use of R Studio, the study evaluates cybersecurity risks related to the banking sector leveraging secondary data collection approaches. It includes techniques for processing the pre-processing, cleaning, and conversion of data to make room for the 40,000 entries with 25 columns. Multiple methods, including K-Nearest Neighbors, Decision Trees, and Gaussian Naive Bayes, are used to categorize and forecast the risks. R libraries are used for comparisons and searches to improve the precision of the model while searching for any trends. The present study confirms that the banking industry's capability to detect and identify cybersecurity issues is enhanced by the integration of machine learning into R Studio. To model cybersecurity threats, R Studio was used to implement the Gaussian Naïve Bayes (GNB), Decision Tree (DT), and K-Nearest Neighbour (KNN) classifiers. The models' respective accuracies of 34.08%, 34.04%, and 34.05% were extremely near to the random baseline that would be anticipated for a three-class problem. These findings suggest that the current dataset has limited predictive power and offer a repeatable standard for subsequent research using enhanced feature engineering and sophisticated ensemble or deep learning techniques.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cybersecurity threat detection for financial institutions: developing advanced systems using R to safeguard against cyber attacks

  • 期刊论文

  • 发表日期：

    2025年12月16日

查看全文
  • Sagar Ramani,
  • Minal Shukla,
  • Amit Sata,
  • Divya Mobarsa,
  • Slaheddine Jarboui

摘要

The study highlights the significance of cybersecurity in defending significant organizations' sensitive data while confronting the real issue of increasing cyber dangers in the banking sector. Therefore, the study hopes to improve the capability of identifying and mitigating the increasing cases of cyber threats, including DDoS and Malware attacks, utilizing a mix of machine learning algorithms in the context of R Studio. Therefore, the study demonstrates how enhanced big data and proactive analytical models support risk evaluation and offer trustworthy defenses against cyber threats that raise customer confidence in financial services. With the use of R Studio, the study evaluates cybersecurity risks related to the banking sector leveraging secondary data collection approaches. It includes techniques for processing the pre-processing, cleaning, and conversion of data to make room for the 40,000 entries with 25 columns. Multiple methods, including K-Nearest Neighbors, Decision Trees, and Gaussian Naive Bayes, are used to categorize and forecast the risks. R libraries are used for comparisons and searches to improve the precision of the model while searching for any trends. The present study confirms that the banking industry's capability to detect and identify cybersecurity issues is enhanced by the integration of machine learning into R Studio. To model cybersecurity threats, R Studio was used to implement the Gaussian Naïve Bayes (GNB), Decision Tree (DT), and K-Nearest Neighbour (KNN) classifiers. The models' respective accuracies of 34.08%, 34.04%, and 34.05% were extremely near to the random baseline that would be anticipated for a three-class problem. These findings suggest that the current dataset has limited predictive power and offer a repeatable standard for subsequent research using enhanced feature engineering and sophisticated ensemble or deep learning techniques.