EMPOWERING PEOPLE THROUGH DATA
Khaled
Mahfouz
Data Scientist
Aerospace Engineer
Services
Deep-dive eda
There is an unexplained fluctuation in the diamond price for the same weight.
EDA or, exploratory data analysis is an approach to finding hidden patterns, trends, and observations within data.
data analytics
Focused on predicting future results through data, taking into account predictor variables that contribute to the target variable, or by predicting the likelihood of an event to happen like fraud detection & churn prediction.
Models that I currently apply:
-
Linear regression.
-
Ridge regression.
-
Logistic regression.
-
Decision Tree classifier.
-
Random Forest & Gradient Boosting.
-
ARIMA & GARCH forecasting models.
-
K-means segmentation model.
AUTOMATION
Reduce error, save time & costs!
Designing and implementing customized scripts tailored for business needs, If the task is routine, error-sensitive, and requires labor, then let scripts do it for you!
-
Mass certificate automation (image processing and editing)
-
Microsoft office & google apps automation
-
Email automation
Data collection
Utilizing all data gathering techniques from all kinds of sources. cleansing, reshaping, and detecting outliers.
Sources and techniques that I currently use:
-
Flat files (xlsx, csv, tsv .... etc).
-
Relational databases (PostgresSQL, MS SQL).
-
Web scraping/crawling (Beatuifulsoup).
-
API data collection.
Describe your image
Describe your image
Describe your image
A/B test
Make a data-driven decision using the power of hypothesis tests, where we can find out if results are due to chance or are they coming from changes that we've made!
Projects
Projects that I've been working on lately.
Medical Appointments No-Show Analysis
In this project, we perform a deep dive EDA to discover multiple facts about the Brazilian medical sector, from appointments rescheduling phenomenon to concentrations of an age-specific group for scheduling appointments at a specific time of day that might raise concerns for medical urgency or a disease that affects a certain age bracket. Performed using python and jupyter notebooks
E-Commerce website Conversion rates: A/B Test
This project is a research on conversion rates of an E-commerce website, where we perform an A/B test on a website click-through rate through control and experiment groups, to decide which page design brings more conversion or are both of them are the same. Testing
takes the guesswork out of website optimization and enables data-informed decisions that shift business conversations from "we think" to "we know".
Check-Send-Read-Remind "CSRR"
A scripted action-based semi-automated workflow tailored for e-mail marketing and managing email campaigns, CSRR stands for the four actions that it does, Check Data, send emails, Read Inbox & Remind unanswered. Recipients’’ email addresses are stored in an Excel file and assigned a message (thread) ID, they get validated to protect the sender's reputation, It sends personalized emails to all recipients, then reads the inbox marking those who replied in the Excel file, and reminds those who didn’t reply automatically.
Data Expo 2009: Airline on-time data Flight Data Analysis
From 1987 to 2008, we analyze the evolution of flight departure and arrival delays through 12 giga-byte flight records for all commercial flights within the USA, scatter charts depicting the creep of flights towards early arrivals and departures through the years. Conducted using python and jupyter notebooks.
Hello,
I'm Khaled!
A quality engineer working in home appliances industry, based in Egypt.
I focus on delivering impactful data-driven solutions and insights, that push the business forward.
I earned my B.Eng. in Aerospace Science and Aeronautical Engineering from Cairo University in 2018, An interdisciplinary field that made me versatile, ever-evolving & target-oriented.
I shifted my career to data science back in July 2021, and since then, I've been honing my skills, learning new tools and technologies to stay ahead of the curve.
I'm motivated to enhance everyday life by turning what might seem unsolvable into simple, obvious, effortless, and scalable as much as It can be.
