Slides for IA 310 - Mapping World Cities - at Lafayette College

The primary objective of this project was to explore emerging companies and trends in the EdTech industry, utilizing HolonIQ’s comprehensive list of startups and innovators. This initiative sought to provide a detailed understanding of the current market landscape by categorizing the startups based on their specific focus areas. Additionally, the project aimed to identify potential clients, providers, partners, competitors, and emerging markets. The analysis was carefully structured to examine these companies across different geographic regions and industry verticals. By synthesizing this data, the project ultimately offered strategic insights into Edunext’s position within the competitive landscape, providing actionable recommendations for enhancing market positioning and fostering growth.

regresi linear

In this project, I utilized the well-known Iris dataset to demonstrate how data can be effectively analyzed and presented using R Markdown. While the Iris dataset is often considered introductory, I chose it specifically because many are familiar with it, allowing the focus to be on the clarity and aesthetics of the analysis rather than the data itself. This project highlights how to go beyond basic analysis by creating beautiful, interactive notebooks that enhance both the presentation and discovery of insights. Through clear visualizations such as histograms, box plots, and scatter plots, paired with statistical tests like the Shapiro-Wilk normality test, I illustrate how R Markdown can be a powerful tool not only for performing data analysis but also for presenting it in a professional and polished manner.

Hồi quy Logistic

This project investigates the use of clustering techniques to analyze and categorize financial data for bankruptcy prediction. The primary aim is to uncover inherent groupings within the data, specifically identifying clusters that represent entities at risk of bankruptcy versus those that are financially stable. The project involved preprocessing and analyzing financial datasets, applying various clustering algorithms to uncover patterns, and evaluating the effectiveness of these methods in distinguishing between different financial states. By exploring and validating clusters, the project seeks to understand the underlying financial characteristics associated with bankruptcy and assess how well clustering can segregate bankrupt and non-bankrupt entities. The results demonstrate that the clustering methods successfully identified two distinct clusters, reflecting the financial dichotomy of bankruptcy versus stability, and achieved high classification accuracy. This highlights the potential of clustering techniques as valuable tools for financial analysis and risk assessment.

conjunto de análises de investimento no Dolar e no Real