🎯 Aspiring Data Analyst | AI/ML Enthusiast
📊 Solving Real-World Business Problems Using Data
I am a Data Analytics and Machine Learning fresher with strong hands-on experience in solving real-world business problems using data.
While I do not have formal industry experience yet, I have independently built multiple end-to-end analytics projects simulating real business environments — including fintech transaction analysis, airline performance dashboards, retail intelligence systems, and predictive machine learning models.
My focus is not just building models, but understanding business problems and delivering data-driven insights.
Python | SQL | Excel
Pandas | NumPy | Matplotlib | Seaborn | Scikit-Learn
Power BI | Tableau | Excel Dashboards
- Linear Regression
- Logistic Regression
- Decision Trees
- Random Forest
- K-Means Clustering
- Time Series Forecasting
- Model Evaluation & Hyperparameter Tuning
Data Cleaning
Exploratory Data Analysis (EDA)
Feature Engineering
Statistical Analysis
Business Insight Generation
(Python + Power BI)
Problem: Digital payment platforms face high failure rates during peak hours and network instability.
Solution:
- Generated realistic synthetic transaction dataset
- Performed full data cleaning & EDA in Python
- Built interactive Power BI dashboard using DAX
- Identified:
- Peak-hour failure spikes
- Network-based system risks
- Seasonal load impact
- Payment method reliability gaps
Outcome: Demonstrated how data can improve platform reliability and reduce transaction failures.
(SQL)
Problem: Understanding revenue drivers and customer ordering behavior.
Solution:
- Built 13 business-driven SQL queries
- Used JOINs, subqueries, aggregations
- Identified top-selling products and revenue trends
- Analyzed order patterns and size preferences
Outcome: Converted raw transactional data into actionable revenue insights.
(Power BI)
Problem: Analyzing operational performance and passenger trends.
Solution:
- Cleaned and transformed data
- Built KPI-based dashboard
- Created DAX measures
- Identified route-level performance gaps
Outcome: Presented insights in executive-level visual storytelling format.
(Machine Learning)
Problem: Predicting employee attrition using historical HR data.
Solution:
- Applied Logistic Regression
- Performed feature engineering
- Evaluated using Precision, Recall, F1-score
- Interpreted model for business impact
Outcome: Showcased predictive analytics for workforce retention strategy.
- House Price Prediction
- Advertising ROI Analysis
- Student Performance Prediction
- Customer Segmentation (K-Means)
- Retail Sales Forecasting
- Google Reviews Clustering
- Public Transportation Optimization
These projects demonstrate practical application of regression, classification, clustering, and forecasting techniques.
✔ Business-problem-first approach
✔ Clean and structured data pipelines
✔ Strong visualization and storytelling
✔ End-to-end project execution
✔ Focus on practical implementation
- Advanced Machine Learning (XGBoost, Model Optimization)
- Real-time analytics dashboards
- LLM-powered data assistants
- Failure prediction systems
📧 Email: prathamsoni1128@gmail.com
💻 Portfolio: https://pratham-soni-portfolio.lovable.app
🔗 LinkedIn: https://www.linkedin.com/in/pratham-soni-600787268
⭐ Open to entry-level Data Analyst / Business Analyst / ML roles.