Aspiring Data Analyst skilled in transforming raw data into powerful insights using Excel, SQL, Power BI, Tableau, and Python.

👋 Hi, I'm P. V. Dharani

<aside> 💡

Aspiring Data Analyst | MCA Graduate

MCA graduate and passionate Data Analyst focused on transforming data into meaningful insights that support smarter decisions.

I enjoy working across the entire analytics workflow —

data cleaning → exploration → visualization → reporting → automation.

</aside>

🛠 Technical Skills ****

<aside> 💡

• Excel (Formulas, Pivot Tables, Dashboards)
• SQL (Joins, CTEs, Aggregations, Filters)
• Power BI (DAX, Data Modeling, KPI creation)
• Tableau (Basics: charts, filters, dashboards)
• Python (Pandas, NumPy)
• Machine Learning (Basics)
• Git & GitHub </aside>

🌟 Soft Skills

<aside> 💡

• Problem Solving
• Communication
• Attention to Detail
• Time Management </aside>

📂 Projects

<aside> 💡

1️⃣ Insurance Analytics Dashboard

📝 Project Overview

Designed and developed an end-to-end Insurance Analytics Dashboard to analyze key business metrics such as Premiums, Renewals, Cross-sell, and Achievement vs Targets.

The dashboard helps management quickly identify performance gaps, trends, and opportunities for growth.

🛠 Tools Used

• 🟩 Excel → Data cleaning, formulas, pivot tables
• 🟦 SQL → Joins, aggregations, data extraction
• 🟨 Power BI → KPI dashboards, DAX calculations, modeling
• 🔵 Tableau → Visual storytelling & interactive charts

📊 My Role & Responsibilities

• ✔ Cleaned and transformed raw insurance data using Excel & SQL
• ✔ Built interactive dashboards in Power BI and Tableau
• ✔ Created essential KPIs:
  • Total Premium
  • Renewal Rate
  • Cross-sell %
  • Achievement % vs Target
• ✔ Added filters for Region, Product, Channel, Month
• ✔ Designed executive-level summary visuals

🔍 Key Insights Identified

• 📌 High-performing regions contributed majority of premium revenue
• 📌 Some regions/products showed low renewal rates, indicating retention issues
• 📌 Cross-sell opportunities were identified in segments with high engagement
• 📌 Achievement vs Target analysis revealed underperforming months

🚀 Business Impact

The dashboard enabled teams to:

• Make data-driven decisions faster
• Improve renewal strategies
• Monitor KPIs in real-time
• Identify revenue growth opportunities
• Strengthen cross-sell and retention planning

🔗 Links & Attachments

• 🖼 Dashboard Screenshots :https://github.com/pillidharani27-alt/Insurance_Analytics-Dashboard/tree/main/Screenshots
• 🐙 GitHub Repository :https://github.com/yourusername/Insurance-Analytics

2️⃣ ML-Based Spam Comments Detection On YouTube (ML Project)

📝 Project Overview

Built a machine learning model that detects whether a YouTube comment is Spam or Not Spam using NLP preprocessing and multiple classification algorithms.

The system was deployed as a Flask web application where users can enter a comment and get real-time predictions.

🛠 Tools & Technologies

• 🐍 Python — pandas, numpy
• 🔤 NLP Techniques — text cleaning, tokenization, TF-IDF
• 🤖 Machine Learning Models — Naive Bayes, Decision Tree, MLP, AdaBoost
• 🌐 Flask — Web application integration
• 🐙 GitHub — Version control and project hosting

📊 My Role & Responsibilities

• ✔ Performed text cleaning: lowercasing, stop word removal, punctuation removal
• ✔ Converted text into numerical features using TF-IDF vectorization
• ✔ Trained and compared multiple ML models
• ✔ Selected the best-performing model based on accuracy & performance metrics
• ✔ Built a Flask UI where the user enters a comment → model predicts spam/not spam
• ✔ Added screenshots and deployed the app locally

🔍 Key Insights

• 📌 Naive Bayes performed well for text classification
• 📌 MLP & AdaBoost improved prediction stability for noisy comments
• 📌 TF-IDF features helped reduce overfitting and improved accuracy
• 📌 Most spam comments contained repetitive keywords and promotional patterns

🚀 Project Outcome

The final application can:

• Detect spam comments instantly
• Provide meaningful classification for content moderation
• Demonstrate end-to-end ML workflow (EDA → Model → Deployment)

🔗 Links & Attachments

• 🖼 Flask App Screenshots : https://github.com/pillidharani27-alt/YouTube-Spam-Detection-ML/tree/main/Screenshots
• 🐙 GitHub Repo Link : https://github.com/pillidharani27-alt/YouTube-Spam-Detection-ML
• 🧪 Sample test comments : https://github.com/pillidharani27-alt/YouTube-Spam-Detection-ML/tree/main/Data set

3️⃣Python Mini Projects (Python Project)

📝 Project Overview

Developed a collection of Python mini projects aimed at strengthening core Python concepts, logical thinking, and problem-solving skills.

</aside>

🎓 Education

<aside> 💡

</aside>