A multi-tool data analytics portfolio covering SQL, Python, Excel, Big Data (Hive), and BI dashboarding (Tableau/Power BI–style visuals). The repo is organized as a flat collection of standalone projects rather than nested folders, with each file (or small group of files) representing a self-contained exercise in a particular tool.
SQL - Data Cleaning.sql — Cleans a Nashville housing dataset: standardizes date formats, fills missing property addresses via self-joins, splits address fields into street/city/state using a custom SPLIT_STR function, normalizes Y/N flags into Yes/No, removes duplicates with a windowed CTE, and drops unused columns.
SQL - Data Exploration.sql — Explores global COVID-19 case and vaccination data: death/infection rates by country and continent, global daily totals, and rolling vaccination percentages using window functions, CTEs, a temporary table, and a view for downstream visualization.
Instagram Clone SQL - Database & Inserting Data.sql — Builds a relational schema for an Instagram-style app (users, photos, comments, likes, follows, tags, photo_tags) and seeds it with data.
Instagram Clone SQL - Exploratory Data Analysis.sql — Eighteen analysis questions against that schema (most active users, top hashtags by usage and likes, registration trends by day of week, regex-based username filtering, engagement and comment-rate breakdowns) plus four stored procedures for popular tags, engaged users, comment counts, and per-user activity lookup.
PostgreSQL-BI-CHALLENGE — A three-part business intelligence case study on a construction bidding platform (projects, bid packages, bid requests, subcontractors): data quality checks (duplicates, nulls), analysis of bid request volume, conversion rates, and decline reasons by subcontractor type, and a closing segmentation of subcontractors by country.
Python - Movie Industry EDA Project.ipynb — An exploratory analysis of the film industry using pandas, matplotlib, and seaborn, framed as a business case for a company entering the movie business. It works through eight questions — optimal production budget, most profitable genres, best release timing, highest-value actors/directors, Oscar-winning budget thresholds, the effect of rating/runtime, profitability benchmarks against major studios, and which studio to model best practices on — and closes with a numbered set of actionable recommendations.
Hadoop(Hive) - NYC Yellow Taxi Case Study.txt — HiveQL queries against NYC Yellow Taxi trip data: table creation and load, total trips and revenue, toll/tip share of revenue, average fare by payment type, highest-revenue hour of day, and average trip distance.
| File | Sheets / focus |
|---|---|
| Excel - LOOKUP, INDEX, MATCH, SUMIFS.xlsx | A customer quoting workbook using LOOKUP/INDEX/MATCH against an international price list and a SUMIFS-driven discount matrix. |
| Excel - Pivot Tables, Pivot Chart, Slicers.xlsx | Shipping data summarized with pivot tables, a pivot chart, and slicers, with per-rep breakdown sheets. |
| Excel - Sales Performance Dashboard.xlsx | A sales dataset rolled up into trend, region, product, segmentation, and state views, combined into a single dashboard sheet. |
| Excel - Scenario Manager, Solver (Data Modeling).xlsx | A project costing model with Scenario Manager and Solver used for what-if analysis and optimization. |
Screenshots of each workbook's output are in visuals/excel/.
Static exports of dashboards built in Tableau/Power BI-style tools, in visuals/ (with two duplicated at the repo root):
- Instagram Clone Data Analysis Dashboard — visual layer on top of the Instagram Clone SQL data: most-liked users, most active posters, sign-ins by day, and most-used hashtags.
- SuperStore Weekly series (
TopDownDashboard,BottomUpDashboard,KPIDashboard,Q&ADashboard,GeoChart) — five linked dashboards over the same retail superstore dataset, each built from a different design approach (top-down KPI view, bottom-up exploratory view, executive KPI scorecard, guided Q&A view, and a geographic customer map). - Retail Pricing Analytics — profit and pricing-curve analysis to find the price point that maximizes profit across product lines.
- London Bus Safety — incident counts by borough, type, route, and season for London buses (2015–2018).
- E-Commerce Retail — email campaign revenue, top cities by sales, and traffic-source breakdown for an online store.
- World Bank CO2 Emission — CO2 emissions and per-capita emissions by country over time, using World Bank data.
- Work From Home — a Makeover Monday-style breakdown of why people prefer remote work.
- Municipality Data Analysis Dashboard and Grover Data Analyst Dashboard — case-study dashboards on social media post volume by category/source and on order/revenue trends, respectively.
- SQL: MySQL-flavored SQL (window functions, CTEs, views, stored procedures, regex), PostgreSQL
- Python: pandas, numpy, matplotlib, seaborn
- Big Data: Apache Hive (HiveQL)
- Spreadsheets: Excel (LOOKUP/INDEX/MATCH, SUMIFS, PivotTables, Slicers, Scenario Manager, Solver)
- BI/Visualization: Tableau-style dashboarding
- The
.sqlfiles are scripts meant to be read or run against a database with the referenced tables already loaded (e.g.nashville_housing,coviddeaths_csv,covidvacinations_csv) — the underlying datasets aren't included in this repo. - The Hive case study is a plain-text walkthrough of queries and tasks rather than a runnable script.
- Some dashboards in
visuals/are practice/case-study exercises (credited to their original designer where the image itself carries that credit) rather than dashboards built from scratch on original data; they're included here as part of the BI skills on display.