Sultan Asiri image

Sultan Asiri

I am a Ph.D. student in the computer science department at the University of Alabama. I am a member of Dr. Xiao's lab (W4Net). I am lecturing at the King Khalid University, Saudi Arabia.

Research Interst

  • Phishing detection
  • Advertisement attacks
  • Natural Language Processing
  • Cybercrime

News

  • I received the Computer Science Department's Outstanding Graduate research award for the 2024 academic year.
  • One paper was accepted Computer & Security Elsevier Journal 2024
  • I passed my Ph.D. defense on March,5 2024. I am now a Ph.D. in Computer Science.
  • One paper was accepted in ACMSE 2024 conference
  • I joined the 2024 ACM Southeast Conference (ACMSE 2024) as a committee program member
  • In December 2023, I was awarded to attend the Neural Information Processing Systems (NeurIPS) conference in New Orleans, Louisiana, USA
  • One paper was published in MDPI Electronics journal 2023
  • I passed my Ph.D. proposal exam on Apr,20 2023. I am now a Ph.D. candidate in Computer Science.
  • One paper was published in ACMSE 2023 conference
  • One paper was published in IEEE Access Journal 2023
  • One paper was published in ACMSE 2022 conference
  • One paper was published in MDPI Electronics Journal 2022
  • I passed my Ph.D. qualifying exam on Sep,12 2021.
  • One paper was published in the 27th ACM SIGKDD Conference 2021

Resume:

Education

  • Bachelor of Computer Science
    • 2014
    • King Khalid University, KSA
  • Master of Computer and Information Science
    • 2029
    • Gannon University, USA
  • Philosophy of Computer Science
    • 2024
    • The University of Alabama, USA

Experience

  • Web developer
    • 2014 - 2016
    • Private Company, KSA
  • Lecturer
    • 2016 - Present
    • King Khalid University, KSA
  • Member of The High Performance Data Analytics and Computing (HiPDAC) Lab
    • 2019-2020
    • The University of Alabama, USA
  • Member of Geospatial data analysis lab
    • 2020-2021
    • The University of Alabama, USA
  • Member of W4NET lab with Dr. Yang Xiao
    • 2021- Present
    • The University of Alabama, USA

Research Intrests

  • Phishing detection
  • Advertisement attacks
  • Natural Language Processing
  • Cybercrime

Publications

Conference

  • Conti Ransomware Development Evaluation.
    • Saleh Alzahrani, Yang Xiao, and Sultan Asiri
    • April 2023
    • New York, NY, USA
    • (READ)
  • A low-power, machine learning-based optical communications system for disaster relief
    • Cary Xiao, Sultan Asiri
    • April 2022
    • New York, NY, USA
    • (READ)
  • Weakly Supervised Spatial Deep Learning based on Imperfect Vector Labels with Registration Errors.
    • Zhe Jiang, Wenchong He, Marcus Kirby, Sultan Asiri, Da Yan
    • August 2021
    • New York, NY, USA
    • (READ)

Journal

  • PhishingRTDS A real-time detection system for phishing attacks using a Deep Learning model
    • Sultan Asiri, Yang Xiao, Saleh Alzahrani, Tieshan Li
    • June 2024
    • Computer & Security (Elsevier)
    • (READ)
  • PhishTransformer A Novel Approach to Detect Phishing Attacks Using URL Collection and Transformer
    • Sultan Asiri, Yang Xiao, Tieshan Li
    • January 2023
    • MDPI
    • (READ)
  • A Survey of Intelligent Detection Designs of HTML URL Phishing Attacks.
    • Sultan Asiri, Yang Xiao, Saleh Alzahrani, Shuhui Li, Tieshan Li
    • January 2023
    • IEEE
    • (READ)
  • Traffic Sign Based Point Cloud Data Registration with Roadside LiDARs in Complex Traffic Environments.
    • Zheyuan Zhang, Jianying Zheng, Yanyun Tao, Yang Xiao, Shumei Yu, Sultan Asiri, Jiacheng Li, Tieshan Li
    • January 2022
    • MDPI
    • (READ)

Projects

Phishing Detection

Phishing attacks are a type of cybercrime that use social engineering to trick users into revealing sensitive information, such as personal identity or financial data. Attackers can design fake web pages to mimic legitimate sites, or send fraudulent emails or social media messages that appear to come from a trusted source. The rise of social media has made it easier for attackers to reach a larger pool of potential victims. According to the Anti-Phishing Working Group (APWG), there were over 250,000 phishing attacks reported in January 2021, a record high. Business compromise attacks also increased by 56% from the last quarter of 2020 to the first quarter of 2021. Financial institutions and social media sites are the most commonly targeted websites. Phishing attacks can have a devastating impact on victims. Attackers can use stolen information to commit identity theft, make unauthorized purchases, or even take control of victims' financial accounts. Phishing attacks can also damage businesses by disrupting operations, stealing customer data, or causing repetitional damage. Our goal is to develop a real-time phishing detection system to protect users from these attacks. Our system will use a variety of techniques, including machine learning and natural language processing, to identify and block phishing emails, websites, and social media messages.
  • Date: Present
  • PhD Research
  • Advisor: Dr. Yang Xiao
  • The University of Alabama

Location Extraction from Social Media Posts

In 2017 Hurricane Harvey ruined multiple areas in Houston, Texas. Many homes were destroyed, or the flood prison people in their homes. Therefore, people started to contact the rescue to save them. However,due to the large number of people who needed saving, the call center was busy. Therefore, the people started to use social media to announce their needs for help so the volunteer organization could identify their location to rescue them. So, using Natural language processing to identify the location of people who needs helps artificially using social media post. Our goal was to design a model to analyze the social media tweets and extract the location.
  • Date: 2020
  • PhD Research
  • Advisor: Dr. Zhe Jiang
  • The University of Alabama

Bird Species Detection

Computer vision tasks have recently grown and become one of the most challenging tasks in deep learning. One of these tasks is instance segmentation, which is considered a difficult task in computer vision because of the ability to generate different mask colors on each object in the image. Mask-RCNN Proposed using three networks such as Region Proposed network to locate the object and three CNN networks, one for classification, one for extract feature, and another for generating the mask. These networks together make Mask R-CNN robust because it makes the image pass to different steps; each step got more in-depth in the image. Mask R-CNN showed an excellent result with the COCO dataset. So, we trained it on a fine-grained classification dataset such as birds’ species. It is a very challenging dataset because it contains several species, that each one of these species has very similar details from another. We trained Mask R-CNN on 100 species from CUB-200-2011 dataset
  • Date: 2019
  • Master Degree research
  • Advisor: Stephen T. Frezza
  • Gannon University