Not-so-quick-start Guide

Project by Categories

==========================

Focuses on the technique to obtain or “scrape” data that are located anywhere on the world wide web
The data can be both “structured” or “unstructured”
Strutured data are data that are stored in the form of traditional row-column database
Unstructured data on the other hand are basically everything else than the traditional row-column database. This can include - but not limited to - presentation slides, movie ratings, a picture on a website, or a report of a pandemic on WHO website
Compared to structured data, unstructured are a bit more complicated, either to scrape them, or to clean them

Focuses on the technique of transforming of messy or un-tidy data into “tidy” data, which is a term revolutionized by Hadley Wickham in his popular journal article
Messy data can be both structured and unstructured data
Tidy data are a lot more easier to analysis, as it has the formal structure where;
1. Each variable forms a column
2. Each observation forms a row
3. Each type of observational unit forms a table

Focuses on the technique of exploring the data to answer questions and visualizing the results
EDA allows us to validate data and unravel the relationships between variables
It is common to use summary statistics and regression models in EDA

Focuses on the technique of “supervised” and “unsupervised” learning
Supervised learning is done when we expect an output from a particular dataset, say predicting how many people will buy a particular service or product
Unsupervised learning on the other hand focuses on the natural structure of the dataset, say given a large dataset of people, can we categorize them into such and such clusters depending on their natural characteristics?

Focuses on the technique of visualizing the findings from the data in order to communicate the findings clearly and efficiently
The audience can be both from those who are well-versed in the field of data science and those who are not familiar with the field
Beautiful are not the main goal in visualizations; it is comprehensibility and accuracy. Beautiful is a plus

A category for projects that are too simple to be broken down into the above categories
Each project are completed in a single post

=================================

Each project will have a name that comprises of three(3) parts
First part consist of the actual name of the dataset that are utilized
Second part consist of the name of the project’s category. It is separated from the first part with “-“
Third part consist of the name of the programming language that are utilized. It is enclosed in brackets
For example, consider a project titled “Parkinson’s Disease - Machine Learning (RStudio)”. “Parkinson’s Disease” is the first part, “Machine Learning” is the second part, and “(RStudio)” is the name of the programming language
Keep the naming convention in mind as you browse the website. It makes navigating the projects a lot easier