Obfuscated computer virus detection using machine learning algorithm
Also available in printed version
| 第一著者: | |
|---|---|
| その他の著者: | |
| フォーマット: | Bachelor thesis |
| 言語: | 英語 |
| 出版事項: |
Universiti Teknologi Malaysia
2025
|
| 主題: | |
| オンライン･アクセス: | https://utmik.utm.my/handle/123456789/52582 |
| Abstract | Abstract here |
| _version_ | 1854975068660039680 |
|---|---|
| author | Tan, Hui Xin |
| author2 | Ismahani Ismail, supervisor |
| author_facet | Ismahani Ismail, supervisor Tan, Hui Xin |
| author_sort | Tan, Hui Xin |
| description | Also available in printed version |
| format | Bachelor thesis |
| id | utm-123456789-52582 |
| institution | Universiti Teknologi Malaysia |
| language | English |
| publishDate | 2025 |
| publisher | Universiti Teknologi Malaysia |
| record_format | dspace |
| record_pdf | Abstract |
| spelling | utm-123456789-525822025-08-20T23:51:12Z Obfuscated computer virus detection using machine learning algorithm Tan, Hui Xin Ismahani Ismail, supervisor Electrical engineering Also available in printed version Signature based detection method has been used for computer virus detection for many years but it fails to detect new obfuscated computer virus where signatures are not available. This constantly evolving computer virus not only affects the daily life of computer users, but also cause significant threat to information security of organizations and even government. It is no longer effective to store each signature into the database as the computer virus may vary its signature for each generation which will then require a large memory to store the data. This research proposes an alternative approach to the traditional signature based detection method and investigates the use of machine learning technique for computer virus detection. In this work, datasets for both virus and normal files are analyzed and processed to find out the most significant features and generate a suitable classifier model that can correctly classify the unknown files. Text string features are used in this work as it is informative and potentially only use small amount of memory space. The results obtained from the machine learning approach are then compared with the signature based approach. Using WEKA tool, unknown files which do not contain signatures can be correctly classified with 100% accuracy using SMO classifier model. Thus, it is believed that computer virus detection rate can be improved through machine learning approach which will help to strengthen the current computer virus defense zulaihi UTM 67 p. Project Paper (Sarjana Muda Kejuruteraan (Elektrik - Elektronik)) - Universiti Teknologi Malaysia, 2018 2025-03-14T07:29:04Z 2025-03-14T07:29:04Z 2018 Bachelor thesis https://utmik.utm.my/handle/123456789/52582 vital:115515 valet-20181017-11364 ENG Closed Access UTM Complete Unpublished Completion application/pdf Universiti Teknologi Malaysia |
| spellingShingle | Electrical engineering Tan, Hui Xin Obfuscated computer virus detection using machine learning algorithm |
| thesis_level | Other |
| title | Obfuscated computer virus detection using machine learning algorithm |
| title_full | Obfuscated computer virus detection using machine learning algorithm |
| title_fullStr | Obfuscated computer virus detection using machine learning algorithm |
| title_full_unstemmed | Obfuscated computer virus detection using machine learning algorithm |
| title_short | Obfuscated computer virus detection using machine learning algorithm |
| title_sort | obfuscated computer virus detection using machine learning algorithm |
| topic | Electrical engineering |
| url | https://utmik.utm.my/handle/123456789/52582 |
| work_keys_str_mv | AT tanhuixin obfuscatedcomputervirusdetectionusingmachinelearningalgorithm |