Ashraf Kamal


Ph.D. (JMI, New Delhi)

Machine Learning Engineer

PayPal logo with overlapping blue P letters and the word PayPal in blue text, representing a professional and trustworthy financial technology environment
iD 0000-0002-8344-3792

Latest Updates

September 2025: Idea got approved for filing in USPTO. August 2025: US Patent Issued (Granted). July 2025: Presented the accepted Paper at ICCCNT 2025, IIT Indore, MP, India. June 2025: US Patent Issued (Granted). May 2025: Two US Patents Published. April 2025: Two US Patents filed in USPTO. October 2024: Presented the accepted Paper at ICDPN 2024, Czech Republic, Europe. June 2024: Presented two accepted Papers at ICDAM 2024, London Metropolitan University, London, UK. July 2023: Presented the accepted Paper at ICCCNT 2023, IIT Delhi, New Delhi, India.

About

Dr. Ashraf Kamal has earned Doctor of Philosophy (PhD) - Computer Science from Jamia Millia Islamia (A Central University with A++ grade by the NAAC), New Delhi, India. PhD was funded by Ministry of Electronics & Information Technology, Government of India under prestigious Visvesvaraya PhD scheme. Research work is substantiated by more than 20 research publications and 5 US patents (2 Granted; 3 Published). Research papers are published in reputed SCI-indexed journals, including two in IEEE/ACM Transactions and CORE ranking conferences. Research interest lies in the intersection of Computational Linguistics, GenAI, ML, DL, and NLP. Research areas include the detection of figurative language like sarcasm & irony, abusive data, fake news, rumor, fintech, etc., in user-generated content via machine and deep learning techniques.


Working in Privacy ML Team at PayPal. Experience in Machine Learning, Deep Learning, Text Mining, Natural Language Processing, Generative AI, IR, RAG, and Agentic AI. Currently, involved in develop novel theoretical ideas with practical insights, publish technical research papers, patent filing, and tech excellence activities on emerging technologies in Fintech. Besides that, involved in several GenAI related projects.


Proficient in predictive modelling, statistical analysis tools and techniques, exploratory data analysis, data processing, and data mining algorithms, as well as scripting language, including Python. Capable in leveraging advance analytics to drive strategic decision-making and efficient in ideation and development of highly adaptive, robust, and diverse applications to translate business functionalities into substantial deliverables.

Citation Metrics & Research Impact

20+
Total Publication
550+
Citations
11
Filed Patents
2
Granted Patents
5
Published Patents
10
h-index
11
i10-index
6
Journal Papers
11
Conference Papers
1
Book Chapter

Professional Experience

2023 - Present
Machine Learning Engineer, Chennai, Tamil Nadu (Remote)
PayPal
2021 - 2023
Data Scientist
ACL Digital, Bengaluru, Karnataka (Remote)
2015 - 2021
Doctoral Researcher (PhD - Computer Science)
Jamia Millia Islamia, New Delhi
2012 - 2013
Software Engineer
Altudo (Formerly, eDynamic Softech Solutions), Gurgaon, Haryana
2011 - 2012
Software Developer
VAS Data Services (YEPME.com), Gurgaon, Haryana
2010 - 2011
Software Engineer
Oglacs Software Pvt. Ltd., New Delhi

Education

Ph.D - Computer Science
Jamia Millia Islamia
2015 - 2021
Master of Computer Applications
Jamia Millia Islamia
2007 - 2010
B.Sc (Honours) Physics
Jamia Millia Islamia
2003 - 2006

Ph.D. Thesis

Supervisor: Prof. Jahiruddin
Co-supervisor Prof. Muhammad Abulaish
University: Department of Computer Science, Jamia Millia Islamia (A Central University), New Delhi, India
Award Year: 2021
Summary:

The research presented in thesis is an attempt to propose a unified data mining approach for detecting different FL categories, mainly in Twitter, which is a rich source of textual data containing FL. The unified approach aims to develop Machine Learning (ML) and Deep Learning (DL)-based classification techniques for detecting four different FL categories - sarcasm, humor, irony, and satire in Twitter. These FL categories are extensively found on the Web; and over the years, they have become a pervasive phenomenon mainly in short texts like tweets. This thesis also presents a fine-grained analysis of three FL categories, such as sarcasm, humor, and irony, and proposes classification techniques for computational detection of self-deprecating sarcasm, self-deprecating humor, and situational irony, respectively.

Projects

Sarcasm Detection

Developed deep learning models for detecting sarcasm and irony in social media texts, including code-mixed and multilingual data.

Privacy ML

Worked on privacy-aware machine learning solutions for fintech applications at PayPal, focusing on secure data handling and compliance.

Fake News Detection

Built NLP pipelines for identifying fake news and misinformation in online content using transformer-based models.

Abusive Language Detection

Created robust classifiers for detecting hate speech and abusive language in user-generated content, leveraging capsule networks and attention mechanisms.

Financial Sentiment Analysis

Led projects on sentiment and stance detection in financial texts, enabling better risk assessment and market analysis.

AI Skills

  • Innovation: Patent filing (USPTO)
  • Research: Conferences (IEEE, Springer, ACM), Journals (SCIE)
  • Data Science: EDA, Feature engineering, Visualization, Statistical analysis
  • Machine Learning: Supervised, Unsupervised, Reinforcement Learning
  • Natural Language Processing: NLP, Text mining, Information retrieval
  • Generative AI: LLMs, RAG, Prompt engineering, Agentic AI

Technical Skills

  • Python: NumPy, Pandas, Scikit-learn, Matplotlib, Seaborn
  • Model Deployment: Flask, FastAPI, Streamlit, Docker
  • Cloud Platforms: AWS, GCP, Azure ML
  • Collaboration Tools: Git, Jupyter, VS Code

Research Interests

🤖 Machine Learning

Deep learning architectures and optimization

🗣️ NLP

Language models and sentiment analysis

👁️ Computer Vision

Object detection and segmentation

🧠 Explainable AI

Interpretable ML models

⚖️ AI Ethics

Fairness and bias mitigation

Patents

Granted Patents

P1
Systems and Methods for Leveraging Embedded Services
Inventors: Padmapriya Mohankumar, Vishal K. Singh, and Ashraf Kamal
US Patent Application Number : US18/219,308
Grant: August 2025
Link PDF
P2
Systems and Methods for Leveraging Embedded Services
Inventors: Padmapriya Mohankumar, Vishal K. Singh, and Ashraf Kamal
US Patent Application Number : US18/219,352
Grant: June 2025
Link PDF

Published Patents

P3
Systems and methods for Early Fraud Detection in Deferred Transaction Services
Inventors: Padmapriya Mohankumar, Ashraf Kamal , and Vishal K. Singh
US Patent Application Number : US18/521,115
Published: May 2025
Link PDF
P4
Systems and methods for Early Fraud Detection in Deferred Transaction Services
Inventors: Padmapriya Mohankumar, Ashraf Kamal, and Vishal K. Singh
US Patent Application Number : US18/521,212
Published: May 2025
Link PDF
P5
Digital Verification of Users Based On Real-time Video Stream
Inventors: Vishal K. Singh, Padmapriya Mohankumar, and Ashraf Kamal
US Patent Application Number : US18/059,212
Published: May 2024
Link PDF

Filed Patents

P6
System and Method for Managing Threats in Open Banking Services
Inventors: Ashraf Kamal, Padmapriya Mohankumar, and Vishal K. Singh
US Patent Application Number : US18/XXX,XXX (TBA)
Filed: June 2024
P7
System and Method for Copyright and Data Protection in AI Models
Inventors: Padmapriya Mohankumar, Vishal K. Singh, and Ashraf Kamal
US Patent Application Number : US18/XXX,XXX (TBA)
Filed: September 2024
P8
System and Method for Interpretability of Complex Black-Box Models
Inventors: Vishal K. Singh, Padmapriya Mohankumar, and Ashraf Kamal
US Patent Application Number : US18/XXX,XXX (TBA)
Filed: April 2025
P9
System and Method to Avoid Cross-Channel Risks in Omnichannel Services
Inventors: Padmapriya Mohankumar, Ashraf Kamal, and Vishal K. Singh
US Patent Application Number : US18/XXX,XXX (TBA)
Filed: April 2025
P10
Privacy-Enforcing Smart Contracts with Compliance Automation for Data Sharing
Inventors: Padmapriya Mohankumar, Vishal K. Singh, and Ashraf Kamal
US Patent Application Number : US18/XXX,XXX (TBA)
Filed: May 2025
P11
System and Method to Control Threats in AI-based Multi-Agents
Inventors: Padmapriya Mohankumar, Ashraf Kamal, and Vishal K. Singh
US Patent Application Number : US18/XXX,XXX (TBA)
Filed: July 2025

Publications

Journal Papers

  • J1
    Contextualized Satire Detection in Short Texts Using Deep Learning Techniques
    Ashraf Kamal, Muhammad Abulaish, and Jahiruddin
    Journal of Web Engineering (JWE), 23(21), pp. 27-5, 2024
    | (Indexing: SCIE; Impact factor: 1.0)
    Link PDF BibTeX
  • J2
    BiCapsHate: Attention to the Linguistic Context of Hate via Bidirectional Capsules and Hatebase
    Ashraf Kamal Tarique Anwar and Vineet K. Sejwal and Mohd. Fazil
    IEEE Transactions on Computational Social Systems (TCSS), 11(2), pp. 1781 - 1792, 2024
    | (Indexing: SCIE; Impact factor: 4.9)
    Link PDF BibTeX
  • J3
    BiCHAT: BiLSTM with Deep CNN and Hierarchical Attention for Hate Speech Detection
    Shakir Khan and Mohd Fazil and Vineet K. Sejwal and Mohammed A. Alshara and Reemiah M. Alotaibi and Ashraf Kamal and Abdul Rauf Baig
    Cognitive Computation (COGN), 34(7), pp. 4335-4344, 2022
    | (Indexing: SCIE; Impact factor: 4.3)
    Link PDF BibTeX
  • J4
    HCovBi-Caps: Hate Speech Detection Using Convolutional and Bi-Directional Gated Recurrent Unit With Capsule Network
    Shakir Khan and Ashraf Kamal and Mohd Fazil and Mohammed A. Alshara and Vineet K. Sejwal and Reemiah M. Alotaibi
    IEEE Access, 10, pp. 7881-7894, 2022
    | (Indexing: SCIE; Impact factor: 3.6)
    Link PDF BibTeX
  • J5
    CAT-BiGRU: Convolution and Attention with Bi-directional Gated Recurrent Unit for Self-Deprecating Sarcasm Detection
    Ashraf Kamal and Muhammad Abulaish
    Cognitive Computation (COGN), 14, pp. 91-109, 2021
    | (Indexing: SCIE; Impact factor: 4.3)
    Link PDF BibTeX
  • J6
    A Survey of Figurative Language and its Computational Detection in Online Social Networks
    Muhammad Abulaish, Ashraf Kamal, and Mohammed J. Zaki
    ACM Transactions on the Web (TWEB), 14(1), pp. 1-52, 2020
    | (Indexing: SCIE; Impact factor: 4.1)
    Link PDF BibTeX

Conferences

  • C1
    Deep Reinforcement Learning Approach for Customer Churn Prediction Using Multi AI Agents
    Ashraf Kamal and Padmapriya Mohankuamr and Vishal K. Singh
    In Proceedings of the 16th IEEE International Conference on Computing Communication and Networking Technologies (ICCCNT) IIT Indore, India, July 7-11, 2025, pp. 1-6
    Link PDF BibTeX
  • C2
    Cyber Threat Detection by Leveraging Contextual and Semantic Knowledge
    Vishal K. Singh and Padmapriya Mohankuamr and Ashraf Kamal
    In Proceedings of the International Conference on Data-Processing and Networks (ICDPN) London, India, June 14-15, 2024, pp. 363-372
    Link PDF BibTeX
  • C3
    CoMFinSe-MusCaAt: Code-Mixed Financial Sentiment Classification via Multi-scale Context-Aware Attention on Low-Resource Language Settings
    Ashraf Kamal and Padmapriya Mohankuamr and Vishal K. Singh
    In Proceedings of the 5th International Conference on Data Analytics and Management (ICDAM) London, India, June 14-15, 2024, pp. 383-392
    Link PDF BibTeX
  • C4
    Financial Fact-Check Via Multi-modal Embedded Representation and Attention-Fused Network
    Padmapriya Mohankuamr and Vishal K. Singh and Ashraf Kamal
    In Proceedings of the 5th International Conference on Data Analytics and Management (ICDAM) London, India, June 14-15, 2024, pp. 35-43
    Link PDF BibTeX
  • C5
    Financial Misinformation Detection via RoBERTa and Multi-channel Networks
    Ashraf Kamal and Padmapriya Mohankuamr and Vishal K. Singh
    In Proceedings of the 10th International Conference on Pattern Recognition and Machine Intelligence (PReMi) ISI Kolkata, India, Dec 12-15, 2023, pp. 646-653
    Link PDF BibTeX
  • C6
    Fin-STance: A Novel Deep Learning-Based Multi-Task Model for Detecting Financial Stance and Sentiment
    Vishal K. Singh and Padmapriya Mohankuamr and Ashraf Kamal
    In Proceedings of the 14th IEEE International Conference on Computing Communication and Networking Technologies (ICCCNT) IIT Delhi, India, July 6-8, 2023, pp. 1-6
    Link PDF BibTeX
  • C7
    Financial Fake News Detection via Context-Aware Embedding and Sequential Representation using Cross-Joint Networks
    Padmapriya Mohankuamr and Ashraf Kamal and Vishal K. Singh and Amrish Satish
    In Proceedings of the 15th International Conference on COMmunication Systems & NETworkS (COMSNETS) Bengaluru, India, January 3-8, 2023, pp. 780-784
    Link PDF BibTeX
  • C8
    IMFinE: An Integrated BERT-CNN-BiGRU Model for Mental Health Detection in Financial Context on Textual Data
    Ashraf Kamal and Padmapriya Mohankuamr and Vishal K. Singh
    In Proceedings of the 19th International Conference on Natural Language Processing (ICON), IIIT Delhi, New Delhi, India, December 15-18, 2022, pp. 139-148
    Link PDF BibTeX
  • C9
    An LSTM-Based Deep Learning Approach for Detecting Self-Deprecating Sarcasm in Textual Data
    Ashraf Kamal and Muhammad Abulaish
    In Proceedings of the 16th International Conference on Natural Language Processing (ICON), LTRC, IIIT Hyderabad, India, December 18-21, 2019, pp. 201-210
    Link PDF BibTeX
  • C10
    Self-Deprecating Humor Detection: A Machine Learning Approach
    Ashraf Kamal and Muhammad Abulaish
    In Proceedings of the 16th International Conference of the Pacific Association for Computational Linguistics (PACLING), Hanoi City, Vietnam, Communications in Computer and Information Science, vol. 1215. Springer, October 11-13, 2019, pp. 483--494
    Link PDF BibTeX
  • C11
    Self-Deprecating Sarcasm Detection: An Amalgamation of Rule-Based and Machine Learning Approach
    Muhammad Abulaish and Ashraf Kamal
    In Proceedings of the IEEE/WIC/ACM International Conference on Web Intelligence (WI), Santiago, Chile, Dec. 3-6, 2018, pp. 574-579
    Link PDF BibTeX

Book Chapters

  • B1
    Generative Adversarial Networks in Protein and Ligand Structure Generation: A Case Study
    Syed Aslah A. Faizi, Nripendra K, Singh, Ashraf Kamal, and Khalid Raza
    Khalid Raza and Debmalya Barh and Deepak Singh and Naeem Ahmad (Eds.), Deep Learning Applications in Translational Bioinformatics, pp. 231-248, 2014
    Link PDF BibTeX

Awards

  • Visvesvaraya PhD Fellowship, Ministry of Electronics & IT (MeitY), Government of India - (2015 - 2020)
  • Qualified UGC NET (Computer Science) - 2014
  • IEEE, Student Member - (2015 - 2020)
  • ACM (Student Member) - (2015 - 2020)
  • Merit-cum-Means Scholarship, Minitry of Minorities Affairs, Government of India - (2008 - 2010)

Service

PC Member and Reviewer: Conferences
  • ICICCT - 2025
  • ICON - 2023
  • ICONIP - 2022
  • ICON - 2021
  • ICON - 2020
Reviewer: Journals (IEEE, Elsevier, Springer, etc.)
  • IEEE Transactions on Computation Social Systems (TCSS)
  • Cluster Computing
  • Cognitive Computation
  • Cognitive Neurodynamics
  • Discover Applied Sciences
  • Discover Artificial Intelligence
  • International Journal of Data Science and Analytics
  • IEEE Access
  • Neurocomputing
  • PeerJ
  • Knowledge and Information Systems
  • Language Resources and Evaluation
  • Peer-to-Peer Networking and Applications
  • Scientific Reports
  • Signal, Image and Video Processing
  • Social Network Analysis and Mining
  • The Journal of Supercomputing
Reviewer: Books
  • Deep Learning in Genetics & Genomics, Elsevier (2024)
  • Artificial Intelligence and Autoimmune Diseases, Springer (2024)

Media

IEEE Spectrum
February - 2023
Featured in IEEE Spectrum
The paper entitled: "BiCapsHate: Attention to the Context of Hate via Bidirectional Capsules and Hatebase" published in IEEE Transactions on Computational Social Systems is featured in reputed IEEE Spectrum. The article discusses how AI can help detect and mitigate online hate speech and highlights the challenges and advancements in this critical area of research.
Read Article

Contact Me


Feel free to reach out for collaborations, research opportunities, or academic discussions


Open Contact Form
↑