Data Science Simplified : )


undefined Data Science is the process of gathering, analyzing, and visualizing large voluminous amounts of data to gain some insights or knowledge about it to make better decisions. If needed, machine learning algorithms ( clustering, Neural networks, etc) and NLP are also used. The data used here can be structured ( 6 digit pin code, dates, credit card numbers) or unstructured ( audio files, videos, email messages, etc). Data science helps us to find and express solutions to business problems. Hence, we can say that Data science enables the creation of data products.
Some of the real-world applications are :
The recommended movies and products displayed to us in the online sites ( Netflix, Amazon, etc) are implemented with the help of Data Science. It is also used to predict the demand for particular brands based on customer preferences and shopping behavior. It is widely used in the shipments industry to calculated the optimized routes etc.
undefinedThe Data Science process: It typically consists of six steps. They are 1) Setting the research goal, 2) Retrieving the data, 3) Data preparation, 4) Data exploration, 5) Data modeling and 6) Data visualization.
Setting the research goal: Data science helps us to find and express solutions to business problems. It is usually applied in the context of organizations. A project charter is prepared that contains information about the research – data and resources needed, benefits, timetable, and deliverables.
Retrieving the data: With the evolution of the Internet and smart technologies, data is grown as Big data. Sometimes, the data is extracted from the web using web scraping methods.
Data preparation: The collected data ( .csv, .pdf, or .txt formats) is cleansed to remove the outliers (data that does not fall within the range or boundary), bad data, etc. The database system used for loading the data is MySQL or PostgreSQL.
Data exploration: Statistical & predictive modeling methods and probability distributions are used to explore the patterns in data. This step is often called Exploratory data analysis ( EDA, in short). PYTHON or R programming is extensively used in Data Science. Jupyter, Spyder, or Pycharm integrated development environment can be used for writing the coding in Python language. It offers interactive coding ( My goodness, what an amazing feature it is !!!) with Jupyter Notebooks which enable intermixing of code, module outputs, plots, and charts, all within one seamless notebook.
Data modeling or Model building: Either the predictive modeling techniques or Machine learning algorithms can be used to build a model. It involves selecting the variables for the model, model execution, and diagnostics.
Data visualization: For midsized data, the standard libraries supported by python is used for presenting the data distribution in a visual form. Tableau, Microsoft Power BI can also be used for visualizing them.
We shall discuss the amazing Python language features in significant detail sometime later.