Hypothesis:

The uses of machine learning in cybersecurity ultimately are becoming more effective through new innovative tactics that are much more capable of thwarting incoming attacks and threats.

Proposal Outline:

Introduction

Batyha, R. M., Aburashed, T. K., & Alshammari, B. R. (2021)
Kaur, S., & Singh, M. (2020)

Measuring Cybersecurity Effectiveness

Aiyanyo, I., Samuel, H., & Lim, H. (2020)
Olowononi, F., Rawat, D., & Liu, C. (2021).

Machine Learning Effectiveness in Cybersecurity (Literature Review):

Sarker, I., Kayes, A., Badsha, S., Alqahtani, H., Watters, P., & Ng, A. (2020)
Jiang, Y., & Atif, Y. (2021)

Improvements in Machine Learning

Sarker, I., Kayes, A., Badsha, S., Alqahtani, H., Watters, P., & Ng, A. (2020)
Batyha, R. M., Aburashed, T. K., & Alshammari, B. R. (2021)

False Positives in Cybersecurity

Muhammad, Musa Abubakar, and Aladdin Ayesh. (2019)
Lo, O., Buchanan, W. J., Griffiths, P., & Macfarlane, R. (2018)

Malware Detection

Abdulla, S., & Altaher, A. (2015)
Li, S., Zhou, Q., & Wei, W. (2021)

Methodology:

Ten companies of similar size, with similar network infrastructure, and similar reliance upon computers and servers will participate in the study.

Five of the companies will be randomly selected to make use of a new innovative technique of behavior profiling-based detection to detect and mitigate network and intrusions. The innovative technique is one I have reviewed upon my past research:

Muhammad, Musa Abubakar, and Aladdin Ayesh (2019)

The other five companies will use their existing intrusion detection infrastructure.

White hat hackers will test the security quality of all the networks in the form of grey-box testing.

After a 2  month time period, the intrusion detection systems logs will be analyzed and compared.

Discussion:

• Compare company results.
• Compare results with my hypothesis.
• Strengths of the research study.
• Weaknesses of the research study.