Portfolio Projects

About Me

I am a master's graduate with a passion for analyzing and interpreting large datasets to uncover insights and improve business performance.

I have experience working in both the index and software industries, where I honed my skills in data analysis and visualization. I am skilled in programming languages such as Python and SQL, with a strong background in Power BI.

I am deeply passionate about using data to drive decision-making and am always seeking new challenges and opportunities to learn and grow. My portfolio showcases my proficiency in Power BI and highlights my ability to effectively communicate complex data visually.

Project 1: HR Data Analysis using PowerBI

Link to view my report

Below is the HR data analysis report summary. In this overview, I've examined a Human Resources dataset provided by Zoomcharts - Drill Down Visuals for Power BI - Turn your reports into interactive experience, presenting key insights from the HR Data Analysis dashboard.

Data sheet is accessed on May $16^{th}$ 2024 and cloned to my project here.

Page 1

Page 2

1. Employee Diversity

The company's workforce displays diversity in gender and ethnicity. Females slightly outnumber males, comprising 518 individuals compared to 482 males. Asian employees constitute the largest ethnic group at 404, followed by Caucasian (271), Latino (251), and Black (74). Given that 64.3% of the workforce is based in the United States, this country appears to be a significant market focus for the company.

2. Age Distribution

A substantial portion of the company's workforce falls within the 40-49 age bracket, numbering 288 individuals. This is followed by the 50-59 (263) and 30-39 (237) age groups. Only around 12% of the workforce is aged 20-29, suggesting a mature and experienced employee base. This demographic composition may influence strategies related to training and succession planning.

3. Average Annual Salary

Gender-wise, average annual salaries are closely aligned, with males earning slightly more ($114K) than females ($112K). However, between 2017 and 2022, the average income of females increased to $114K, surpassing that of males, which decreased to $111K. Potential reasons for this shift include higher turnover among males, increased promotion opportunities for females, or salary adjustments during the period.

There's notable disparity in salary ranges for board members, with the highest average annual salary reaching $256,561, nearly four times higher than the lowest. This indicates significant pay variations based on factors such as experience, location, or performance. Additionally, system administrators tend to earn more than their network counterparts across all departments.

4. Retention Rate

The retention rate fluctuates over time but generally remains above 98%, suggesting high overall employee satisfaction or effective retention strategies. A notable decrease occurred in 2020-2021, likely influenced by the COVID-19 pandemic's impact on the labor market, followed by a strong rebound in 2022.

While higher-paid areas generally exhibit higher retention rates, this trend does not consistently apply across all job titles. This highlights the need for a more detailed workforce analysis to tailor HR strategies beyond monetary aspects.

Project 2: Amazon Electronic Products Sales using Python

This notebook demonstrates a comprehensive sales analysis of an Amazon electronics products dataset with dimensions 1.3M x 10. The solution is implemented in Python, leveraging Pandas for data manipulation and Matplotlib for visualization.

The analysis aims to address the following questions:

1. What are the categories of electronic products?
2. Find the Top-10 users that bought the most in each category.
3. Analyze sales in a certain year, grouped by categories.
4. Given a brand, identify the categories in which it has products.
5. Determine the categories with the highest market competition.

Approach

A logical and systematic approach is followed throughout the analysis, with each step accompanied by insights into the dataset. Key observations include:

• Timestamp Accuracy: Inconsistencies between the timestamp and year columns are observed, and thus decided to use the timestamp as the time reference for each purchase.
• Data Completeness: A significant proportion of rows (97%) contain NaN values in either the brand or user_attr columns, leading to the exclusion of these columns from some analyses due to their limited utility.

Dataset Details

The dataset encompasses Amazon electronics sales data spanning from 1999 to 2018. It is available on Kaggle.

Gained Insights

1. Categories of Electronic Products: Identified the unique categories of products available in the dataset.

2. Top-10 Users per Category: Determined the top 10 users who purchased the most in each category.

3. Yearly Sales Grouped by Categories: Analyzed and visualized sales data for specific years, grouped by product categories.

4. Brand-wise Product Categories: For each brand, identified the categories in which its products are listed.

5. Market Competition Analysis: Determined which categories have the highest market competition based on the number of unique brands.