Python is an easy to use programming language and one of its capabilities is machine learning.
I've used open to public databases and competitions to fit my machine learning algorithm and reach acceptable answers to real business questions. Projects are in form of regression, classification, association, and clustering.
This project is about predicting house price regarding former sold houses information. Most important part of this project is feature engineering which leads to more accurate model. (Dataset: https://www.kaggle.com/c/house-prices-advanced-regression-techniques)
Learn more
It is a fraud detection in credit card transactions. As the fraud records are less than total transactions, database was unbalanced and main part of this project is to deal with it. (Dataset: https://www.kaggle.com/datasets/mlg-ulb/creditcardfraud?datasetId=310&sortBy=voteCount)
Learn more
This is a market basket analysis and I've used apriori algorithm to suggest which items have more probability to sale together. Key part of this project was to unstack products and group them by each customer. (Dataset: https://www.kaggle.com/datasets/heeraldedhia/groceries-dataset?select=Groceries_dataset.csv)
Learn more
This is a wholesale dataset which I used clustering algorithm to have a better understanding of customers behaviour. Important part of this project was to find the optimal count of clusters. (Dataset: https://archive.ics.uci.edu/ml/datasets/wholesale+customers)
Learn more
In this project, input/output values of different German sectors are analyzed. Plotly was used to build final report. This report shows the sectors which are very dependent on import.
View final report
In this project, input/output values of different Dutch sectors are analyzed. Plotly was used to build final report. This report shows the sectors which are very dependent on import.
View final report