Skip to content
View znadhiac's full-sized avatar

Block or report znadhiac

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
znadhiac/README.md

Hi, I'm Zulfi Nadhia Cahyani 👋

Data Analyst with an MSc in Environmental Engineering, specializing in transforming real-world data into actionable business insights using SQL, Python, Excel, and Tableau. Experienced in end-to-end analytics, including data cleaning, exploratory analysis, machine learning, and dashboarding, to support data-driven decision-making across various business functions.


Skills

  • SQL: Joins, Aggregations, CTEs, Window Functions
  • Python: Pandas, NumPy, Matplotlib, Seaborn, Plotly, Scikit-learn
  • Data Analysis: Data Cleaning, Exploratory Data Analysis (EDA)
  • Visualization: Tableau, Excel (Pivot Tables, VLOOKUP/HLOOKUP), Dashboards
  • Machine Learning: Classification, Regression, Feature Engineering, Model Evaluation, SHAP
  • Tools: Jupyter Notebook, Git, GitHub

SQL Python Excel Tableau Scikit-Learn Jupyter GitHub


Featured Projects

1. E-commerce Customer Churn Analysis and Prediction

🔗 GitHub Repository | Tableau Dashboard | Presentation

  • Tools: Python (Pandas, NumPy), Scikit-learn, CatBoost, Tableau
  • Description: Developed a machine learning classification model to predict customer churn using demographic, behavioral, and transactional data.
  • Key Insights:
    • High churn risk is driven by low engagement, inactivity, and customer complaints
    • CatBoost achieved the best performance with 95% recall and 0.90 F2-score
    • SHAP analysis revealed key behavioral patterns behind customer disengagement
  • Business Value: Enables early churn detection, targeted retention strategies, and more efficient marketing spend.

2. NYC Green Taxi Demand and Operations Analysis

🔗 GitHub Repository | Tableau Dashboard | Presentation

  • Tools: Python (Pandas, NumPy), Tableau
  • Description: Analyzed NYC Green Taxi trip data from January 2023 to uncover demand patterns, fare drivers, and passenger behavior for operational optimization.
  • Key Insights:
    • Peak demand occurs during weekday commuting hours (7–9 AM, 3–6 PM)
    • Manhattan accounts for over 60% of pickups, with strong zone-level concentration
    • Trip distance is the primary fare driver (r ≈ 0.86), while duration has minimal impact
    • Most trips are single-passenger (87%) and predominantly paid via card (65%)
  • Business Value: Enables data-driven fleet allocation, targeted pricing strategies, and service optimization based on demand patterns and customer behavior.

3. Olist Brazilian E-commerce Sales Performance Analysis

🔗 GitHub Repository | Tableau Dashboard

  • Tools: SQL, Python (Pandas, NumPy), Tableau
  • Description: Conducted SQL-based analysis of 2018 Olist e-commerce data to evaluate sales trends, customer behavior, delivery efficiency, and customer satisfaction.
  • Key Insights:
    • Revenue peaks in April ($965K) and May ($974K), driven by higher average order value
    • Revenue is concentrated among a small segment of high-value customers
    • São Paulo dominates total revenue, while low-volume states show high AOV potential
    • Approximately 4,900 late deliveries contribute to lower customer satisfaction
  • Business Value: Supports targeted marketing, high-value customer retention, regional expansion strategies, and delivery performance optimization.

4. Daegu Apartment Price Prediction and Optimization

🔗 GitHub Repository | Tableau Dashboard | Presentation

  • Tools: Python (Pandas, NumPy), Scikit-learn, XGBoost, Tableau
  • Description: Developed and benchmarked regression models to predict apartment prices in Daegu using transaction data from 1978 to 2015, including feature engineering, preprocessing, and hyperparameter tuning.
  • Key Insights:
    • Apartment size, location accessibility, and nearby facilities are the strongest drivers of price
    • XGBoost achieved the best performance (MAE: ₩35.1M, MAPE: 17.5%, R²: 0.803), outperforming baseline models
  • Business Value: Enables data-driven pricing decisions, reducing mispricing risk by up to 36–47% and supporting faster, more accurate property sales.

5. HR Employee Attrition and Performance Analysis

🔗 GitHub Repository

  • Tools: SQL, Python (Pandas, NumPy)
  • Description: Analyzed HR data to uncover patterns in employee attrition, performance, satisfaction, compensation, and career progression to support data-driven retention strategies.
  • Key Insights:
    • Attrition rate is 16.1%, with over 90% of exits coming from employees aged 18–35
    • High performers show the highest attrition rates (~27%), indicating risk of losing top talent
    • Attrition is concentrated in Sales, HR, and early-career employees, with lower-salary groups contributing over 90% of total exits
  • Business Value: Enables targeted retention strategies through early-career support, compensation optimization, and proactive identification of high-risk employee segments.

6. Boutique Hotel Performance Analysis

🔗 GitHub Repository

  • Tools: SQL, Python (Pandas, NumPy)
  • Description: Analyzed Boutique Hotel booking and revenue data to evaluate customer demographics, room utilization, and seasonal demand patterns for operational and pricing optimization.
  • Key Insights:
    • Guests aged 56+ account for ~54% of bookings, with domestic travelers (~55%) as the primary market
    • Single Rooms are fully occupied (100%) while premium rooms (Suites, Family) generate higher revenue per booking but remain underutilized
    • Demand peaks in June and declines in September, highlighting clear seasonal revenue opportunities
    • Credit cards dominate payments (~50%), with cash still widely used (~30%)
  • Business Value: Supports pricing optimization, improved room utilization, targeted marketing strategies, and data-driven seasonal revenue management.

Connect with Me

Pinned Loading

  1. ecommerce-churn-prediction ecommerce-churn-prediction Public

    A machine learning–based churn prediction model for an e-commerce platform, using features from five key behavioral domains: demographics, engagement, transaction behavior, platform preference, and…

    Jupyter Notebook

  2. daegu-apartment-price-prediction daegu-apartment-price-prediction Public

    A machine learning model for predicting optimal apartment listing prices in Daegu, South Korea, using transaction data and real estate features from 1978 to 2015.

    Jupyter Notebook

  3. nyc-green-taxi-analysis nyc-green-taxi-analysis Public

    A data-driven analysis of NYC Green Taxi trips focusing on demand patterns over time and location, fare-related factors, and passenger behavior. The insights aim to support service improvements and…

    Jupyter Notebook

  4. olist-ecommerce-sales-performance olist-ecommerce-sales-performance Public

    SQL-based analysis of Olist Brazilian e-commerce dataset to uncover sales, customer behavior, delivery, and satisfaction insights.

    Jupyter Notebook

  5. boutique_hotel_performance_analysis boutique_hotel_performance_analysis Public

    SQL-based analysis of the Boutique Hotel in Turkey dataset to uncover booking patterns, room performance, customer behavior, and revenue insights.

    Jupyter Notebook

  6. hr-employee-attrition-analysis hr-employee-attrition-analysis Public

    SQL-based analysis of HR Employee Attrition and Performance dataset to uncover workforce, attrition, performance, and satisfaction insights.

    Jupyter Notebook