Intro to Data Science & Analytics
Understand how raw facts transform into organizational intelligence. Discover structured vs unstructured data structures, analytics categories, predictive algorithms, decision support models, and recommendation pipelines.
Defining Data & Formats
**Data** can be formally defined as *facts and statistics collected together for reference or analysis*, or as *information in digital form that can be transmitted or processed further*. Digitized information flows primarily in two organizational formats:
Structured Data
Information organized inside a predefined schema with rigid schemas, clear categories, and mathematical relationships. Frequently located in relational databases (SQL) or clean spreadsheet tables (CSV).
data.frame(...)
Unstructured Data
Information that lack predefined organizational matrices or relational keys. Includes voice logs, raw audio wavs, video logs, PDFs, raw texts, images, and unstructured API feeds.
raw_data_stream
Interactive Lab: Data Format Classifier
Classify the data elements correctly to understand structured vs unstructured models.
The 4 Pillars of Data Analytics
Exploratory Historical Diagnostics
Examines historical raw databases to summarize past events, standardizing measurements through summary tables, graphs, and simple aggregations.
Historical crop yield reports across Philippine agricultural cooperatives.
Module 2: Introduction to R & RStudio Workspace
Explore the core design parameters of the R Ecosystem. Learn how the RStudio integrated environment operates, understand variables naming criteria, mathematical operators, and fundamental atomic types.
The 4 Quadrants of the RStudio Workspace
Source / Script Editor
This panel hosts your multi-line R scripts. Code written here can be saved, edited over time, and executed sequentially in the console.
Variables Naming & Operators
Test standard R calculations. Set variables X and Y to see output remainder or quotients:
> 17 %% 5
[1] 2
Determine the evaluation class of standard R variable assignments:
Module 3: Data Exploration (EDA)
Discover Day 2 strategies for core dataset analysis. Master the absolute mathematical differences between Univariate statistical measures and Bivariate correlation architectures.
Univariate Analysis & Central Tendency
Interactive Lab: Univariate Statistical Calculator
Module 4: Data Manipulation & dplyr Ecosystem
Manipulate continuous variables and categories to match desired analysis criteria. Learn to subset columns using select(), filter rows using filter(), arrange elements, and apply piping operators %>%.
Foundational dplyr Core Verbs
Interactive Sandbox: dplyr Core Verbs Calculator
Choose a dplyr verb below to manipulate a mock dataset of rice farmers and view the compiled output:
| ID | Name | Location | Yield (Bags) |
|---|
# Showing complete unmanipulated dataset
Module 5: Data Cleaning Clinic
Examine advanced methods to clean raw, erroneous databases. Understand how to locate missing values with is.na(), handle NA with statistical averages, and remove duplicated rows.
Techniques for Spotting and Resolving Errors
Interactive Lab: Dirty Data Clinic
This table contains duplicate records, missing yields (NAs), and trailing spaces. Clean it up:
| Name | Region | Precipitation (mm) | Duplicates status |
|---|
> df_cleaned <- raw_dataset
Module 6: Data Wrangling & tidyr
Understand the key principles of tidy data using the tidyr package. Explore how to pivot wider or longer, drop NA structures using drop_na(), and separate or combine string columns.
Tidy Data Principles & Reshaping
Interactive Sandbox: Pivot Reshaper (Wide vs Long)
# Datatable in standard Wide layout.
Module 7: Data Visualization & ggplot2
Explore the declarative Grammar of Graphics implemented by the ggplot2 library. Understand how to construct layers using the + operator.
Layer-by-Layer Chart Builder
Interactive Sandbox: ggplot2 Layer Stack Builder
ggplot(data = mpg)
Module 8: Advanced Analytics & ML
Discover advanced analytics topics including simple regression modeling, unsupervised algorithms, and k-NN classifier architectures. Learn to evaluate predictions with confusion validation matrices.
Machine Learning & Simple Regression
Interactive Lab: Linear Regression Slope Fitter
Adjust the slope slider below to fit a linear regression equation predicting crop yield, minimizing the residual Mean Squared Error (MSE):
Mean Squared Error (MSE) measures prediction variance. Try to minimize this score:
MSE: 12.5lm(Yield ~ Fertilizer_Qty, data = df)