Hello I'm
Umar Farooq
Research Data Scientist
- imumarfarooq@gmail.com
- +92 303 4869420
- Lahore, Pakistan
I, Umar Farooq, currently working as a Remote Computer Vision Researcher at Dutch Autonomous Mobility (DAMBV) Netherlands, I am entrusted with the responsibility of addressing complex issues related to autonomous shuttles and developing robust, scalable software solutions. My research primarily focuses on decision-making methods for autonomous mobility and early-stage disease diagnosis, treatment planning, and monitoring. I am honored to serve as a Data Science Coordinator for the United Nations Development Program (UNDP) Pakistan, in collaboration with the Prime Minister National Health Program. Our mission is to provide swift and dignified access to medical healthcare to underprivileged citizens across Pakistan. In addition to my work with the UNDP, I am also closely working with the University of Education, Data Science Healthcare group. We are currently working on Brain Tumor classification using transfer learning and have published a paper on the Global Scientific Journal. This project has allowed me to further develop my research skills and contribute to the field of healthcare. My passion lies in utilizing my skills and expertise to make a positive impact in the world. I am committed to making a positive impact in the world through my work, and I am excited to continue pushing the boundaries of what is possible in this exciting and rapidly evolving field.
Deep learning has been trendy and intriguing in recent years in machine learning. DL is the most accurate, supervised, time, and cost-effective way in terms of ML. There is no end to the amount of knowledge you may get with deep learning.
Read MoreWe proposed a Software-Defined Future Access Network architecture that can help to replace the hardware-based bandwidth distribution component into software- based bandwidth distribution component. The bandwidth distribution component permits pluggable bandwidth distribution algorithms among multiple ONUs adaptive to traffic scenarios and network operational conditions.
Read MoreBrain tumor classification is a crucial aspect of medical applications that rely on computer-aided diagnosis. Misdiagnosing the type of brain tumor can hinder the patient's response to treatment and reduce their chances of survival. This paper presents a novel approach to classifying brain cancers in MRI scans using traditional CNN techniques and transfer learning methods.
Read More
The Iris dataset was used in R.A. Fisher's classic 1936 paper, The Use of Multiple Measurements in Taxonomic Problems, and can also be found on the UCI Machine Learning Repository.It includes three iris species with 50 samples each as well as some properties about each flower. One flower species is linearly separable from the other two, but the other two are not linearly separable from each other.
View Details
In this project, I'm applying Data Engineering techniques to analyze the data from Figure Eight to build a classification model for an API that classify the Disaster message into one of the given 36 categories and contact the particular category company to help these peoples.In data, Directory contains the dataset of real Disaster messages amd categories. You will also get the update data from Figure Eight website. I use these datasets to perform ETL operations to clean the data that can easily feed into the Machine Learning model and get better results. In Data Engineering, the main task is to clean the data and make a pipeline to use multiple algorithms at a time and get better accuracy.
View Details
For this project, I design a recommendation system in which you will analyze the interactions that users have with articles on the IBM Watson Studio platform, and make recommendations to them about new articles you think they will like. Below you can see an example of what the dashboard could look like displaying articles on the IBM Watson Platform.Though the above dashboard is just showing the newest articles, you could imagine having a recommendation board available here that shows the articles that are most pertinent to a specific user.
View Details
We proposed a Software-Defined Future Access Network architecture that can help to replace the hardware-based bandwidth distribution component into software- based bandwidth distribution component. The bandwidth distribution component permits pluggable bandwidth distribution algorithms among multiple ONUs adaptive to traffic scenarios and network operational conditions. In our proposed SDN architecture, the controller helps to control the flow of data traffic by segregating the traffic is divided into two types; one for high priority (HP) data traffic and other for low priority (LP) data traffic.
View Details
Brain tumor classification is a crucial aspect of medical applications that rely on computer-aided diagnosis. Misdiagnosing the type of brain tumor can hinder the patient's response to treatment and reduce their chances of survival. This paper presents a novel approach to classifying brain cancers in MRI scans using traditional CNN techniques and transfer learning methods. The proposed method can effectively classify the type of tumor and assist doctors in treating the patient according to their diagnosis. To achieve this, we employed transfer learning techniques using traditional CNN models such as MobileNetV2, Inception ResNet50, and VGG16 networks.
View Details
How to Build a Flower Species Classification MobileNetV2 Model using TPU?
Read More
What’s the breed type of Dog? How can we identify the nature of the Dog? How can we find which breed type of Dog loves ear scratches?
Read More
In this blog-post, I will go through the process of finding those factors that affect the survival rate prediction in the titanic disaster. I’ve to use the famous titanic dataset that available on Kaggle.
Read MoreDuring undergraduage studies, I participated in the PASTIC intra-university project competition, presented "Simulation-based real-time object detection and segmentation on an autonomous vehicle," and won the runner-up award.
During undergraduage studies, I participated in the IEEE Hackathon Competition at FAST University and presented a "Multi-tasking RoboCar" prototype that can perform actions such as line follower and obstacle detection with wireless operation and was awarded the third-place prize.
As the head of the JAZBA community, my responsibility was to plan and organize academic and recreational events at the university's computer science department. It offers a range of indoor technical and non-technical activities that strengthen the reputation of the institution and empower students.