## Superstore Sales Analysis and Visualization in R This project presents an exploratory data analysis and visualization of the Superstore retail dataset using R. The dataset contains transactional sales records including product categories, customer segments, shipping modes, order dates, and financial metrics such as sales, profit, and discounts. The primary objective of this analysis is to understand business performance patterns and identify key drivers of revenue and profitability. Data preprocessing was carried out using **dplyr** and **lubridate**, which involved cleaning the dataset, converting date formats, handling missing values, and aggregating sales metrics across time and categories. Several analytical questions were explored, including: * How sales and profit vary over time * Which product categories generate the highest revenue * The relationship between sales and profit * The impact of shipping mode on order distribution * Customer segment purchasing behavior * Identification of loss-making sub-categories * Seasonal sales trends across months Visualization was performed using **ggplot2** to present insights through bar charts, scatter plots, and trend summaries. These visualizations highlight patterns such as strong year-end sales peaks, dominance of technology products in revenue contribution, and profitability challenges within certain furniture-related sub-categories. This project demonstrates practical skills in data cleaning, transformation, statistical summarization, and visual storytelling using R. It reflects an applied data analytics workflow from raw data inspection to business insight generation, supporting informed decision-making in retail operations. The work serves as a foundational portfolio piece showcasing competency in data manipulation, visualization, and analytical reasoning within the R programming environment.

Simple dashboard

Tugas Analisis Multivariat Pertemuan 1

This study, developed in R software, proposes a regional model of populated anthromes for applications in environmental management and socioecological analysis. Based on global anthromes mapping guidelines, census data were explored, mined, merged, and plotted to identify population patterns. Static and interactive maps were produced, followed by validation and uncertainty assessments to ensure quality. Results indicate that the regional model closely reflects local population realities compared to global mapping, with sufficient data to characterize demographic distribution. Statistical analyses confirmed the robustness of the approach, demonstrating how Metrology, Human Ecology, and Geocomputation integrate to structure socioecological models and validate geospatial information. The framework can be replicated in other territorial contexts for anthromes-based management.