Social engineering attack classifications on social media using deep learning
In defense-in-depth, humans have always been the weakest link in cybersecurity. However, unlike common threats, social engineering poses vulnerabilities not directly quantifiable in penetration testing. Most skilled social engineers trick users into giving up information voluntarily through attacks...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tech Science Press
2023
|
| Subjects: | |
| Online Access: | http://eprints.utm.my/106321/1/NurHalizaAbdulWahab2023_SocialEngineeringAttackClassificationsonSocialMedia.pdf |
| _version_ | 1811132315236040704 |
|---|---|
| author | Aun, Yichiet Gan, Ming-Lee Abdul Wahab, Nur Haliza Guan, Goh Hock |
| author_facet | Aun, Yichiet Gan, Ming-Lee Abdul Wahab, Nur Haliza Guan, Goh Hock |
| author_sort | Aun, Yichiet |
| collection | ePrints |
| description | In defense-in-depth, humans have always been the weakest link in cybersecurity. However, unlike common threats, social engineering poses vulnerabilities not directly quantifiable in penetration testing. Most skilled social engineers trick users into giving up information voluntarily through attacks like phishing and adware. Social Engineering (SE) in social media is structurally similar to regular posts but contains malicious intrinsic meaning within the sentence semantic. In this paper, a novel SE model is trained using a Recurrent Neural Network Long Short Term Memory (RNN-LSTM) to identify well-disguised SE threats in social media posts. We use a custom dataset crawled from hundreds of corporate and personal Facebook posts. First, the social engineering attack detection pipeline (SEAD) is designed to filter out social posts with malicious intents using domain heuristics. Next, each social media post is tokenized into sentences and then analyzed with a sentiment analyzer before being labelled as an anomaly or normal training data. Then, we train an RNN-LSTM model to detect five types of social engineering attacks that potentially contain signs of information gathering. The experimental result showed that the Social Engineering Attack (SEA) model achieves 0.84 in classification precision and 0.81 in recall compared to the ground truth labeled by network experts. The experimental results showed that the semantics and linguistics similarities are an effective indicator for early detection of SEA. |
| first_indexed | 2024-09-24T00:01:35Z |
| format | Article |
| id | utm.eprints-106321 |
| institution | Universiti Teknologi Malaysia - ePrints |
| language | English |
| last_indexed | 2024-09-24T00:01:35Z |
| publishDate | 2023 |
| publisher | Tech Science Press |
| record_format | dspace |
| spelling | utm.eprints-1063212024-06-29T05:57:31Z http://eprints.utm.my/106321/ Social engineering attack classifications on social media using deep learning Aun, Yichiet Gan, Ming-Lee Abdul Wahab, Nur Haliza Guan, Goh Hock QA75 Electronic computers. Computer science In defense-in-depth, humans have always been the weakest link in cybersecurity. However, unlike common threats, social engineering poses vulnerabilities not directly quantifiable in penetration testing. Most skilled social engineers trick users into giving up information voluntarily through attacks like phishing and adware. Social Engineering (SE) in social media is structurally similar to regular posts but contains malicious intrinsic meaning within the sentence semantic. In this paper, a novel SE model is trained using a Recurrent Neural Network Long Short Term Memory (RNN-LSTM) to identify well-disguised SE threats in social media posts. We use a custom dataset crawled from hundreds of corporate and personal Facebook posts. First, the social engineering attack detection pipeline (SEAD) is designed to filter out social posts with malicious intents using domain heuristics. Next, each social media post is tokenized into sentences and then analyzed with a sentiment analyzer before being labelled as an anomaly or normal training data. Then, we train an RNN-LSTM model to detect five types of social engineering attacks that potentially contain signs of information gathering. The experimental result showed that the Social Engineering Attack (SEA) model achieves 0.84 in classification precision and 0.81 in recall compared to the ground truth labeled by network experts. The experimental results showed that the semantics and linguistics similarities are an effective indicator for early detection of SEA. Tech Science Press 2023 Article PeerReviewed application/pdf en http://eprints.utm.my/106321/1/NurHalizaAbdulWahab2023_SocialEngineeringAttackClassificationsonSocialMedia.pdf Aun, Yichiet and Gan, Ming-Lee and Abdul Wahab, Nur Haliza and Guan, Goh Hock (2023) Social engineering attack classifications on social media using deep learning. Computers, Materials and Continua, 74 (3). pp. 4917-4931. ISSN 1546-2218 http://dx.doi.org/10.32604/cmc.2023.032373 DOI : 10.32604/cmc.2023.032373 |
| spellingShingle | QA75 Electronic computers. Computer science Aun, Yichiet Gan, Ming-Lee Abdul Wahab, Nur Haliza Guan, Goh Hock Social engineering attack classifications on social media using deep learning |
| title | Social engineering attack classifications on social media using deep learning |
| title_full | Social engineering attack classifications on social media using deep learning |
| title_fullStr | Social engineering attack classifications on social media using deep learning |
| title_full_unstemmed | Social engineering attack classifications on social media using deep learning |
| title_short | Social engineering attack classifications on social media using deep learning |
| title_sort | social engineering attack classifications on social media using deep learning |
| topic | QA75 Electronic computers. Computer science |
| url | http://eprints.utm.my/106321/1/NurHalizaAbdulWahab2023_SocialEngineeringAttackClassificationsonSocialMedia.pdf |
| work_keys_str_mv | AT aunyichiet socialengineeringattackclassificationsonsocialmediausingdeeplearning AT ganminglee socialengineeringattackclassificationsonsocialmediausingdeeplearning AT abdulwahabnurhaliza socialengineeringattackclassificationsonsocialmediausingdeeplearning AT guangohhock socialengineeringattackclassificationsonsocialmediausingdeeplearning |