EWMA control chart and machine learning for anomaly detection
Not available
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Bachelor thesis |
| Language: | English |
| Published: |
Universiti Teknologi Malaysia
2025
|
| Subjects: | |
| Online Access: | https://utmik.utm.my/handle/123456789/43021 |
| Abstract | Abstract here |
| _version_ | 1854975069694984192 |
|---|---|
| author | Tan, Sek Fook |
| author2 | Adnan Hassan, supervisor |
| author_facet | Adnan Hassan, supervisor Tan, Sek Fook |
| author_sort | Tan, Sek Fook |
| description | Not available |
| format | Bachelor thesis |
| id | utm-123456789-43021 |
| institution | Universiti Teknologi Malaysia |
| language | English |
| publishDate | 2025 |
| publisher | Universiti Teknologi Malaysia |
| record_format | dspace |
| record_pdf | Abstract |
| spelling | utm-123456789-430212025-08-21T00:34:02Z EWMA control chart and machine learning for anomaly detection Tan, Sek Fook Adnan Hassan, supervisor Mechanical Engineering Not available The amount of data moving across network is increasing exponentially year by year as result of increment of human activities on Internet. However, the network safety remains a concern of many domain experts and all the users. In this present paper, we propose a conventional statistical process control method, exponential weighted moving average (EWMA) control chart and a modern deep learning method, artificial neural network (ANN) for intrusion detection. Since, the network traffic data is often very high dimensional. It is important to conduct feature selection to remove redundancy and capture valuable information underlying the dataset itself. Thus, we employed 4 natural inspired algorithms to select the optimized feature subset. Our paper intended to provide more recent study of EWMA control chart-based network intrusion system and comparison of EWMA control chart with machine learning. The key finding of this research is natural inspired algorithms are great feature selection tools for network traffic dataset. Furthermore, ANN is able to outperform univariate EWMA control chart. zulraizam UTM 97 p. Project Paper (Sarjana Muda Kejuruteraan (Mekanikal)) - Universiti Teknologi Malaysia, 2020 2025-03-11T05:24:30Z 2025-03-11T05:24:30Z 2020 Bachelor thesis https://utmik.utm.my/handle/123456789/43021 vital:148529 valet-20220613-134621 ENG Closed Access UTM Complete Unpublished Completion application/pdf Universiti Teknologi Malaysia |
| spellingShingle | Mechanical Engineering Tan, Sek Fook EWMA control chart and machine learning for anomaly detection |
| thesis_level | Other |
| title | EWMA control chart and machine learning for anomaly detection |
| title_full | EWMA control chart and machine learning for anomaly detection |
| title_fullStr | EWMA control chart and machine learning for anomaly detection |
| title_full_unstemmed | EWMA control chart and machine learning for anomaly detection |
| title_short | EWMA control chart and machine learning for anomaly detection |
| title_sort | ewma control chart and machine learning for anomaly detection |
| topic | Mechanical Engineering |
| url | https://utmik.utm.my/handle/123456789/43021 |
| work_keys_str_mv | AT tansekfook ewmacontrolchartandmachinelearningforanomalydetection |