MD RASHIDUNNABI

Completed a year-long rotation focusing on data-driven decision-making:

• AI Gen ACW Time Analysis: Led AI-generated call log analysis using Python, Snowflake, and Plotly, reducing After-Call Work (ACW) time by 30% through outlier detection and data exploration.
• April Call Trend Analysis: Analyzed call volume fluctuations using Python, Plotly, and Snowflake to identify key factors affecting 'Claims' and 'Renewals', improving engagement strategies by 15%.
• Call and Email Volume Analysis: Analyzed call/email trends with SQL and Python, performing exploratory data analysis to reveal a 20% increase in contact rate trends relative to membership growth.
• Third-Party Claim Detection: Developed a TensorFlow-based machine learning model, using Seaborn for visualization and NLP to analyze Claim_Notes, enhancing model accuracy by 25% through A/B testing.

Completed a 5-month project in environmental analytics, improving air quality forecasting with advanced machine learning and collaborative strategies.
1. Advanced Predictive Modeling: Led data analysis at five monitoring stations using Python and R, refining predictive blending and hybrid models to accurately forecast air pollutant levels.
2. Model Development with Advanced Algorithms: Integrated Random Forests, XGBoost, and CatBoost into complex models, significantly improving forecast reliability and accuracy.
3. Data Visualization Expertise: Crafted detailed visualizations with Plotly to showcase data patterns and model predictions, facilitating clear communication with stakeholders.
4. Collaborative Project Leadership: Key contributor to the BEHAV-I-AIR project, collaborating with Cork City Council and UCC's Applied Psychology department to utilize forecasts in reducing urban pollution impacts.

Participated in advanced research projects using deep learning and statistical methods to tackle healthcare challenges:

• Dengue Disease Spread Prediction (ML/AI): Used Python, Scikit-learn, and Hadoop for data preprocessing and analysis, enhancing precision of public health decisions.
• Brain Tumor Detection (Deep Learning/AI): Developed models with Python, TensorFlow, and Keras, employing Unet and ResNet50 architectures with extensive data augmentation, improving MRI-based tumor detection accuracy.
• Plant Disease Detection (AI/Deep Learning): Applied TensorFlow and Python for Convolutional Neural Networks and image processing, effectively classifying plant diseases to support agricultural interventions.

Executed various projects using machine learning and data analytics across multiple domains:

• Fake News Detection (NLP/AI): Implemented a Python and TensorFlow-based NLP model to detect and classify fake news, significantly reducing misinformation spread.
• Toxic Comment Classification (NLP/AI): Developed a system using NLTK and TensorFlow for preprocessing and identifying toxic comments, enhancing social media moderation.
• COVID-19 Detection with X-Ray Images (Healthcare AI): Utilized transfer learning and CNNs in TensorFlow and Keras to enhance the speed and accuracy of COVID-19 diagnostics.
• Smart Traffic Monitoring System (AI/Traffic Monitoring): Built an ANN model with Python and TensorFlow for vehicle detection and verification, improving traffic flow efficiency and reducing congestion.

Provided freelance data analysis services, enhancing client success across various industries:

• Market Segmentation Analysis: Utilized Python and K-means clustering via Scikit-learn to identify customer segments, increasing conversion rates by 25%.
• Sales Forecasting Model: Employed Python's ARIMA model for time series analysis, forecasting sales trends to reduce inventory costs by 15%.
• Customer Churn Prediction: Developed a logistic regression model with Python and TensorFlow, analyzing user data to decrease churn by 20%.
• Operational Efficiency Analysis: Performed regression analysis to boost manufacturing efficiency by 30%, visualizing improvements in Power BI for stakeholders.

As a Data Science Intern at Flip Robo Technologies, applied data science and machine learning skills to deliver actionable insights and improve operations:

• Data Analysis: Processed complex datasets to support business decisions using Python, R, SQL, and various visualization tools.
• Model Development: Built and refined machine learning and deep learning models, employing A/B testing to enhance accuracy and reliability.
• Team Collaboration: Worked with product managers, engineers, and data scientists to align on business needs and deliver data-driven solutions that meet project and company objectives.