Through the project, we achieved a 7.5% ratio of expected losses to the total financed amount, it divided into 4 parts:
1. Data preparation: we cleaned the data and created columns from useful, close to each other data.
2. We made a PD model from useful data.
3. Monitoring: At this stage we compared the Actual and Probable results, at this stage we received the data of 2015 (load_data_2015)
4. LGD, EAD, and EL: In the last step, we made LGD, EAD models and put all the models into one DataFrame and found EL through them.
This project contains SQL queries for cleaning and transforming a dataset. The goal of this project is to demonstrate the skills and knowledge of cleaning and transforming data using SQL.
It can successfully predict 93% of the customer's repurchase, which product he will buy,
which similar products he will divide, and how many he will accept, using Machine Learning and
Deep Learning Algorithms through a Neural network.
This holds all of my Tableau Dashboards.
The Resume Parser is a Python project on GitHub that parses resumes and extracts data like name, contact, skills, education, and work experience.
It uses natural language processing and machine learning to analyze different resume formats, and including PDF files.
It helps recruiters, HR professionals, and job seekers to automate resume analysis and screening.
This portfolio project on GitHub shows the author's time series analysis and forecasting skills in Python. The project uses libraries like NumPy, Pandas, Matplotlib, and Prophet to forecast retail sales from historical data.
It involves data preprocessing, feature engineering, visualization, and forecasting with ARIMA and Prophet.
It reveals the steps of time series analysis and the author's Python and data analysis expertise.
This project shows how to use Python libraries like BeautifulSoup and Requests to scrape product data from Amazon.
It can automatically extract product data from Amazon 86,400 times a day.
It explains how to get product name, price, rating, and reviews for a certain category. It also cleans and visualizes the data to give insights.
The project shows web scraping skills and Python data analysis.
The response is derived from the user’s text. A short meaning of the text is:
This project shows how to make a chatbot for programmers using ChatGPT API. The chatbot runs in a Jupyter Notebook and uses a loop to keep talking until the user says “bye”. The chatbot sends the user’s input to the API and gets back a response.
The chatbot can help programmers with their questions and can be changed for different needs. The project shows how machine learning can make smart chatbots.
How do data workers make money, which continent earns more money, are they satisfied with the money they are being paid, what programming language do they use, what is their level of data complexity,
what is their life-work balance, and similar questions. If you want to find an answer, I recommend you to refer to this dashboard.
The model increased house price prediction to 97% accuracy. Based on a lot of data obtained from the popular home-selling website.
CRiSP-DM methodology during the project: Business Understanding, Data Understanding, Data Preparation, Modelling, and Evaluation, and using Machine Learning algorithms.
Success Tableau Dashboard made about Covid 19.
You can see the top 5 countries regarding illness, death, recovery, and vaccination through it.
Leveraged expertise in SQL query development and database management to design and implement complex queries, ensuring accuracy and completeness of data analysis.
Overall, the project provides a good example of how SQL can be used to analyze and visualize COVID-19 data, and may be useful for anyone interested in data exploration or epidemiology.
Which gender rides a bicycle more often, which continent, and whether they are married or not,
how much it depends on the level of education (bachelor etc).
We can find answers to these and similar questions through this dashboard.
This is a Python project that uses data science and machine learning to predict employee absenteeism.
It shows how to clean, analyze, and model data using various libraries and algorithms. It showcases the author’s data skills and insights.