The main objective of this data science personal project portfolio is to demonstrate my skills in solving business challenges through my knowledge and tools of Data Science.

_{Kidou Keiji Jiban}

_{Data Scientist}

I have been studying Machine Learning since 2018 in environmental engineering college, in my final paper i used a neural network to detect semantical segmentation of landslides scars.

I have mastery of all stages of developing a business solution using the concepts and tools of Data Science, from understanding the business to publishing the model in production using Clouds.

I have already developed solutions for important business problems such as insights trought data and sales forecast.

The details of each solution are described in the projects below.

Analytical Tools:

Data Collect and Storage: SQL, MySQL, PostgreSQL, SQL Server, SQLite. Flask-sqlachemy.

Data Processing and Analysis: Python.

Development: Git, Linux, Scrum, Docker, Flask.

Machine Learning Modeling: Classification, Regression, Clustering, Time Series.

Machine Learning Deployment: Heroku, Azure, AWS.

Building a Machine Learning Model to sales is a common and essential use in data science.

In this project, i developed a Machine Learning model able to forescat sales with Mean Absolute Percentage error accuracy(MAPE): 0.14 +/- 0.02. The performance of this model would increase revenue of R$5 millions according to the company's business model described in the problem definition.

Learning to rank (LTR) is a class of algorithmic techniques that apply supervised machine learning to solve ranking problems in search relevancy. In other words, it’s what orders query results.

In this project, i developed a model able to predict the probability of a customer purchasing vehicle insurance . In this way, ranking customers from the most likely to purchase to the least likely.

With 20.000 calls made to a new set of 127.000 customers, the perfomance of this model with Recall at 20000: 0.81 would increase annual revenue of R$2.4 millions according to the company's business model described in the problem definition.

Knowing how to analyze and explore data is critical knowledge for a data scientist.

In this project, through exploration, analysis and data visualization tools, I created insights for the company and in the end recommended the purchase of 20 properties resulting in revenue of R$9 millions according to the company's business model described in the problem definition.

On Going