Numbers Dont Lie

Learning how to approach the project

Today I learned aboout the Data Science Lifecycle. For me doing things in basketball I think it is all about the numbers. I think learning this lifecycle will help me determine the performance of most players, coaches, trainers, athletic directors, exectuvies, and even the league commissioners. This has been a dream of mine since around 2014 without even realizing it.

8 years ago I had a dream about how I will use the game of basketball to teach math. Sense that dream I have developed or am in development on an app, a game, an online platform, and now I am learning data science so I can create metrics, models, and analyze performance in the hoops world. I am so happy I was persistent in going after this feat. This is one of the highlights of the journey. Never in my wildest dreams would I have thought the guy who hated school and just wanted to play ball would be excited about data science.

I saw this quote on my morning workout this morning and I think it is relevant for this moment. “So many of our dreams at first seem impossible, then they seem improbable, and then, when we summon the will, they soon become inevitable.” Christopher Reeves

I will get into my journey in another post but for now lets focus on the work at hand.

I was told by a data scientist that the Data Science Lifecycle is the most important work in data science. So I will go through each one. I will also say what each one means in my own words below the brief description.

1. Business Understanding
   • Ask relevant questions and define objectives for the problem that needs to be tackled.
     • Ask the right questions for your project so that what you are looking for makes complete sense and then define each for the problem
2. Data Mining
   • Grab and scrape the data necessary for the project.
     • Go and find the right data and then depending on how much data there is go in and grab what is relevant.
3. Data Cleaning
   • Fix the inconsistencies within the data and handle the missing values.
     • Make sure the data is consistent and make sure I find something to put in the place for whats missing
4. Data Exploration
   • Form Hypotheses about your data through visualizing the data
     • I think this means that we are finding out the details of the relationship with the data and our project
     • I keep hearing visualize the data a lot. I think it may be important.
5. Feature Engineering
   • Select important features and construct more meaningful ones using the raw data that you have.
     • Start to get into detail about what data works so your project can be dope.
6. Predictive Modeling
   • Train machine learning models, evaluate their performance, and use them to make predictions.
     • I am somewhat confused on how data mining and machine learning work together.
7. Data Visualization
   • Communicate the findings with key stakeholders using plots and interactive visualizations
     • I think this is when you show other people what your data project consists of to get feeback before publishing it.

I have tried to simplify it even further after reading Alivias notes again.

Business Understanding

• Understanding how the business and the data work together

Data Mining

• Finding the data I need

Data Cleaning

• Giving the data a haircut

Data Exploration

• EDA is finding the foundational metrics about your population or dataset

Feature Engineering

• Find and make dope features

Predictive Modeling

• Make models that can help me expand

Data Visualization

• Finding out what metrics are important for the project through visualizing the problem and the solution to that problem