22 Years of Experience in Computer science and 4+ years as a Data Scientist, broad-based experience in building data-intensive applications, and overcoming complex architectural and scalability issues in diverse industries.
Having expertise in data storage structures, data mining, and data cleansing.
Translating numbers and facts to inform strategic business decisions.
Analyzing sales figures, market research, logistics, or transport data.
Creating and following processes to keep data confidential. Coming up with solutions to costly business problems.
Data Science Department, University of the Punjab, Lahore
Sep 2019 - Present
Research / Teaching Post Graduate Students, working on multiple projects of Data Science. Consultancy as Data Scientist to multiple companies. Managing multiple projects as Project Manager.
PhD Computer Engineering
University of Engineering and Technology, Taxila, Pakistan 2010 - 2015
Project Management Professional (PMP)®
Project Management Institute
The PMP is the gold standard of project management certification. Recognized and demanded by organizations worldwide, the PMP validates your competence to perform in the role of a project manager, leading and directing projects and teams.
Lean Six Sigma Green Belt
Lean Six Sigma Green Belt is a professional who is well versed in the core to advanced elements of Lean Six Sigma Methodology, who leads improvement projects and serves as a part of more complex improvement projects
PRINCE2 Agile Practitioner takes the knowledge acquired at the Foundation level and applies it to the workplace, using real-world management examples.
An Efficient SMOTE-Based Deep Learning Model for Heart Attack Prediction
Cardiac disease treatments are often subjected to the acquisition and analysis of a vast quantity of digital cardiac data. These data can be utilized for various beneficial purposes. The proposed research work presents a cost-effective solution to predict heart attack with high accuracy and reliability by using uses a synthetic minority oversampling technique (SMOTE) to handle given imbalance data.
Extracting Software Change Requests from Mobile App Reviews
The mobile apps have thousands of reviews which are widely acknowledged as a valuable resource for the community involved in the development of mobile apps. We contend that these reviews can be used to generate software change request documents for improving mobile apps. A pre-requisite for generating such a document is the identification of Software Change Requests (SCR) from the user reviews.
Manual processing of this large number of reviews to identify SCRs is a resource-intensive task....
Blind Image Deblurring Using Laplacian of Gaussian (LoG) Based Image Prior
International Journal of Innovations in Science & Technology
It is possible to deconvolve a blurred image into its original form without any knowledge of the actual image or the process that leads it to be blurred, known as a point spread function. Two phases are involved in producing a blurred image: convolution and deconvolution of the PSF from the blurred image. Video conferencing, diagnostic imaging, and celestial imaging all require this blind deconvolution, but it is difficult to calculate the PSF before the operation.
Ensemble-classifiers-assisted detection of cerebral microbleeds in brain MRI
Cerebral Microbleeds (CMBs) are considered an essential indicator in the diagnosis of critical cerebrovascular diseases such as ischemic stroke and dementia. The framework consists of three phases: brain extraction, extraction of initial candidates based on threshold and size-based filtering, and feature extraction and classification of CMBs from other healthy tissues in order to remove false positives using Support Vector Machine, Quadratic Discriminant Analysis and ensemble classifiers.
Mispronunciation detection using deep convolutional neural network and transfer learning-based model
Computer-assisted language learning (CALL) systems provide an automated framework to identify mispronunciation and give useful feedback. this research investigates the use of the deep convolutional neural network for mispronunciation detection of Arabic phonemes. This model uses convolutional neural network features (CNN_Features)-based technique and a transfer learning-based technique to detect mispronunciation detection.
A hybrid doctor-recommender system is proposed, by combining different recommendation approaches; content base, collaborative and demographic filtering to effectively tackle the issue of doctor recommendation. The proposed system addresses the issue of personalization by analyzing a patient's interest in selecting a doctor. It uses a novel adaptive algorithm to construct a doctor's ranking function.
Multi-class Alzheimer's disease classification using image and clinical features
Cardiac disease treatments are often subjected to the acquisition and analysis of a vast quantity of digital cardiac data. These data’s utilization becomes more important when dealing with critical diseases like a heart attack where patient life is often at stake. The research work presents a cost-effective solution to predict heart attacks with high accuracy and reliability by predicting heart attack via various machine learning algorithms without the involvement of any feature engineering.