soniya patil

soniya patil

Data AI Analyst
India
Kannada, Marathi, English, Hindi

About Me

Detail-oriented Data Scientist with strong expertise in Python, SQL, and Machine Learning. Experienced in building predictive models, performing exploratory data analysis, and developing scalable data pipelines for data …

Experience

Blink Password Authentication System

<p>Designed secure authentication system using computer vision and ML (KNN)<br>
Added multi-layer security with OTP and spoof detection<br>
Built backend APIs using Flask and integrated MySQL database<br>
Implemented data validation, processing pipelines, and user authentication flow</p>

Loan Approval Prediction using Machine Learning

<p>Built and compared ML models (Logistic Regression, KNN, Decision Tree, Gradient Boosting) on 2,000+ loan records<br>
Performed feature engineering (DTI ratio, Loan Sanction Amount) to improve model performance<br>
Evaluated models using Precision, Recall, F1-score, achieving ~80% accuracy with Gradient Boosting<br>
Demonstrated improved performance of ensemble models for accurate loan approval prediction</p>

Password Authentication System

Blink

Designed secure authentication system using computer vision and ML (KNN), Added multi-layer security with OTP and spoof detection, Built backend APIs using Flask and integrated MySQL database, Implemented data validation, processing pipelines, and user authentication flow

Plant Disease Prediction

Developed CNN-based model using TensorFlow for image classification (38 disease categories), Performed image preprocessing and feature extraction using OpenCV, Built Flask-based web application for real-time predictions, Optimized model performance and improved prediction accuracy

Machine Learning

Loan Approval Prediction

Built and compared ML models (Logistic Regression, KNN, Decision Tree, Gradient Boosting) on 2,000+ loan records, Performed feature engineering (DTI ratio, Loan Sanction Amount) to improve model performance, Evaluated models using Precision, Recall, F1-score, achieving ~80% accuracy with Gradient Boosting, Demonstrated improved performance of ensemble models for accurate loan approval prediction

Certifications

NPTL DATA MINING

IIT · Mumbai, India · 2026

Skills

Python Active Listening Adaptability NumPy Time Management Power BI Tableau Machine Learning & Analytics Regression Classification Clustering Model Evaluation Predictive Programming SQL Data Structures & Algorithms (DSA) OOP Modeling Data Engineering & Processing Data Wrangling (ETL) Data Cleaning Feature Engineering Data Preprocessing Pandas Scikit-learn Matplotlib Seaborn Advanced Excel Statistics & Deployment EDA Probability Hypothesis Testing Flask (API Development) NLP Deep Learning (CNN) Soft Skill Self-Motivation OOP Predictive Modeling Data Wrangling ETL Flask Deep Learning CNN TensorFlow OpenCV MySQL KNN Logistic Regression Decision Tree Gradient Boosting
Report this Profile?