Muhammad Furqan Javed E-mail : mfurqanjaved@gmail.com Phone : +923336853091 Website: https://www.linkedin.com/in/furqanjaved Professional Experience CarePRN Sep 2022 — Dec 2022 Data Scientist Utilized advanced querying, visualization and analytics tools to analyze and process complex data sets. Built company A/B testing frameworks to test model quality. Developed intricate algorithms based on deep-dive statistical analysis and predictive data modeling. Identified, measured and recommended improvement strategies for KPIs across business areas. Created and implemented new forecasting models to increase company productivity. Compiled, cleaned and manipulated data for proper handling. Tested and validated models for accuracy of predictions in outcomes of interest. . Modeled predictions with Decision Tree classifier. Implemented randomized sampling techniques for optimized surveys. Integrated information from multiple data sources, solved common transformation problems and resolved data cleansing and quality issues. Mined internal and external sources and joined disparate, non-normalized data sets. Quantum Medical Billing & IT Technologies Nov 2021 — Aug 2022 Data analyst Identified and documented detailed business rules and use cases based on requirements analysis. Identified, analyzed and interpreted trends or patterns in complex data sets. Designed and implemented tools to support strategic value chain optimization initiatives. Used statistical methods to analyze data and generate useful business reports. Upheld security and confidentiality of documents and data within area of responsibility. Implemented code solutions to answer analytic questions and test and assess new methods. Analyzed transactions to build logical business intelligence model for real-time reporting needs. Documented effective and replicable methods for extracting data and organizing data sources. Personal Projects Ransomware detection using KNN Early detection of ransomware using KNN For our experiments we used ransomware data provided by University of New Brunswick The samples are divided into 3 categories, Trojan, Executable and Email Macros Used Euclidean distance the measure the distance between data points The model performed best when we used the value of K=3, for any higher and lower value of k, The accuracy tends to decrease. I used the trained KNN model with K=3 to classify new malware samples as trojans, executables, or email macros, based on the relevant input data. Proposed Ensemble Framework for Classification of Spam/Phishing Email Muhammad Furqan Javed 1 Used Ensemble technique to detect the spam/phishing emails Dataset used for this Scientific study was publicly available. The csv file consists of 5200 rows. The words were further processed by using count vectorizer which turned them into vectors. Combined their predictions such as by weighting their predictions based on their performance on the training data. By using the stacking ensemble method we decreased the variance of prediction of base models and improve the overall accuracy by reducing the overfitting. Successfully achieved the accuracy of 98% Flight ticket price prediction Obtained a dataset of historical flight ticket prices and relevant information (such as departure and arrival cities, dates, etc.). Cleaned and prepared the data for model training by handling missing values, scaling numerical data, and encoding categorical data. I Split the preprocessed data into training and testing sets. Trained a machine learning model, such as a linear regression or decision tree model, on the training data. Evaluated the performance of the trained model on the testing data. Used the trained model to make predictions on new flight ticket prices, given the relevant input data. Technical Skills Programming: Python (Pandas, NumPy, SciPy, Sci-kit learn, TensorFlow) and SQL Machine Learning: Linear Regression, Logistic Regression, Decision Trees, Support Vector Machines (SVM), Random Forest, and Neural Networks Data Visualization: Matplotlib, Seaborn, Excel, Microsoft Power BI Tools: GIT, Jupyter, and Markdown SQL Data base, Google Collab and IBM Watson Studio Exploratory data analysis, Data Visualization, Neural Networks Education Master's in Artificial Intelligence Feb 2022 — Present Air University Islamabad Bachelor's in Electrical (electronics) engineering Sep 2016 — July 2020 Comsats University Abbottabad Certifications IBM Data science Google Data analytics Science and Analytics Intro - IBM Data Science Fundamentals - IBM Machine Learning with Python - IBM SQL for Data Science - Coursera Python (University of Michigan - Coursera) Data Analysis with Python - freeCodeCamp Scientific Computing with Python - freeCodeCamp Machine Learning - Datacamp Muhammad Furqan Javed 2