SciScholar ChatBot using Llama2

An AI-powered research assistant designed to help researchers, students, and academics interact withscholarly documents. This sophisticated chatbot leverages Meta's Llama 2 model and CTransformers toextract relevant content from PDFs and provide contextually accurate answers to research-related queries.

Key Features:

• AI-driven question answering based on research papers
• PDF content extraction for scholarly insights
• Customizable prompt engineering for better responses
• Local and cloud deployment options
• Integration with Pinecone vector database for efficient semantic search
• Flask web interface for user-friendly interaction

UNICEF Data Analytics Assessment

A comprehensive data analytics project developed for UNICEF's Data and Analytics technical evaluation, focusing on maternal health coverage disparities analysis. This project demonstrates advanced statisticalanalysis capabilities and professional data visualization skills.

Project Scope:

• Analysis of maternal health coverage disparities between on-track and off-track countries
• Population-weighted analysis using UN World Population Prospects 2022 data
• Multi-source data integration from UNICEF Global Data Flow and country classification systems
• Professional reporting with R Markdown and publication-ready visualizations

• 17.4 percentage point gap in ANC4 coverage between country groups
• 23.7 percentage point gap in skilled birth attendance coverage
• Systematic differences in health system capacity demonstrated

Plant Seedling Classification

A computer vision project aimed at modernizing agriculture through automated plant classification. Thismachine learning solution helps identify different plant seedlings to reduce manual inspection work inagricultural settings.

Objectives: Develop an automated system to classify plant seedlings, addressing the urgent need formodernization in agriculture and reducing manual labor requirements. Tech Stack: Python, Jupyter Notebook, Computer Vision, Machine Learning

Used Car Price Prediction

A comprehensive data science project that explores and visualizes used car datasets to build predictivemodels for pricing. This project demonstrates end-to-end machine learning workflow from exploratorydata analysis to model deployment.

Key Features:

• Comprehensive exploratory data analysis and visualization
• Linear regression model implementation for price prediction
• Business insights and recommendations generation
• Statistical analysis and feature engineering

Travel Package Purchase Prediction

A machine learning project focused on predicting customer behavior in the travel industry. This projectanalyzes customer data to predict likelihood of travel package purchases, enabling better marketingstrategies and customer targeting.

Applications

• Customer segmentation and targeting
• Marketing optimization strategies
• Predictive analytics for business decision making

US Airline Tweets Sentiment Analysis

A comprehensive natural language processing project that analyzes sentiment in airline customer tweets.This project uses Twitter data scraped in February 2015, containing 14,640 tweets about six major USairlines, to predict customer sentiment and identify pain points in airline services.

Key Features:

• Sentiment classification (positive, negative, neutral) using multiple ML algorithms
• Analysis of most tweeted airlines and reasons behind negative sentiment
• WordCloud visualization of most frequent words in tweets
• Text preprocessing and feature extraction for NLP tasks
• Confidence level analysis for sentiment predictions

• 14,640 tweet instances with 15 features
• Six major US airlines coverage
• Detailed negative reason categorization
• Geographic and temporal tweet metadata

Credit Card Churn Prediction

A critical business analytics project for Thera Bank to predict and prevent customer churn in credit cardservices. This project addresses the bank's steep decline in credit card users by building predictive modelsto identify at-risk customers and provide actionable insights for retention strategies.

Business Problem:

• Significant revenue loss from declining credit card usage
• Need to identify customers likely to close their accounts
• Requirement for data-driven retention strategies

• Classification model to predict customer churn probability
• Comprehensive exploratory data analysis and visualization
• Feature engineering and model optimization techniques
• Business insights and recommendations for customer retention
• Analysis of customer demographics, transaction patterns, and account characteristics

• Customer demographics (age, gender, education, income)
• Account information (card type, relationship duration, credit limit)
• Transaction behavior (amounts, counts, seasonal changes)
• Customer engagement metrics (contacts, inactive months)

Time Series Analysis & Forecasting

A comprehensive educational project covering advanced time series analysis and forecasting techniquesusing machine learning models. This project demonstrates expertise in temporal data analysis, trendidentification, and predictive modeling for time-dependent datasets.

Course Content:

• Fundamentals of time series analysis and decomposition
• Advanced forecasting methodologies and algorithms
• Integration of machine learning techniques with time series data
• Practical implementations of forecasting models
• Feature engineering for temporal data

• Time series decomposition (trend, seasonality, residuals)
• ARIMA and seasonal ARIMA modeling
• Machine learning approaches to forecasting
• Model evaluation and validation techniques
• Hyperparameter optimization for time series models

• Comprehensive learning materials with YouTube video tutorials
• Hands-on coding examples and datasets
• Progressive curriculum from basics to advanced concepts
• Real-world forecasting applications and case studies

Predictive Modeling for Under-5 Malnutrition

Advanced statistical modeling project focusing on predicting malnutrition among children under 5 yearsof age. This research applies sophisticated predictive analytics to identify risk factors and patterns inchildhood malnutrition for policy intervention and healthcare planning.

Research Focus:

• Child health outcomes and nutritional status prediction
• Risk factor identification and analysis
• Statistical modeling for public health applications
• Evidence-based intervention strategies

Under-5 Mortality Predictive Modeling

Comprehensive epidemiological study using advanced statistical techniques to predict under-5 mortalityrates. This research contributes to understanding childhood mortality patterns and supports evidence-based policy making for child survival interventions.

Research Application:

• Mortality risk assessment and prediction
• Epidemiological pattern analysis
• Public health policy support
• Child survival program evaluation

Spatial Analysis - Texas Geographic Research

Advanced geospatial analysis projects examining demographic and geographic patterns in Texas usingzip code-level data. These studies demonstrate expertise in spatial statistics, GIS analysis, and geographicinformation systems for demographic research.

Key Features:

• Shapefile processing and spatial data manipulation
• Geographic pattern analysis and visualization
• Demographic mapping and spatial statistics
• GIS-based research methodologies

Spatial Count Data Modeling with GLM

Advanced statistical modeling project applying Generalized Linear Models (GLM) to spatial count data.This research demonstrates expertise in handling complex spatial datasets and applying appropriatestatistical techniques for geographic analysis.

Statical Methods:

• Generalized Linear Models for count data
• Spatial autocorrelation analysis
• Model diagnostics and validation
• Advanced regression techniques

Spatially Autoregressive Models Research

Cutting-edge research in spatial econometrics and geographic analysis using spatially autoregressivemodels. These projects demonstrate advanced understanding of spatial dependencies and geographicmodeling techniques.

Research Applications:

• Spatial dependency modeling
• Voronoi polygon analysis
• Geographic pattern recognition
• Spatial econometric analysis

Ethiopia Demographic and Health Survey Analysis

Comprehensive demographic research using Ethiopia's Demographic and Health Survey (DHS) data withadvanced spatial analysis techniques. This research demonstrates expertise in processing large-scale demographic datasets and applying GIS methodologies.

Research Scope:

• DHS data processing and analysis
• Point pattern analysis and spatial statistics
• Demographic indicator mapping
• Health outcome spatial distribution

Cox Regression and Survival Analysis

Advanced survival analysis research applying Cox regression and frailty models to demographic data. Thisstudy demonstrates sophisticated statistical modeling techniques for time-to-event analysis indemographic and health research.

Research Scope:

• Cox proportional hazards modeling
• Frailty models for heterogeneity
• Survival curve estimation
• Hazard ratio analysis

Population Demographics Analysis

Detailed demographic analysis focusing on population change and ethnic composition in major Texasmetropolitan areas. These studies utilize American Community Survey (ACS) data for comprehensivepopulation analysis.

Research Scope:

• Population change mapping and analysis
• Ethnic demographic patterns
• Metropolitan area demographic studies
• ACS data processing and visualization

Women's Health and Nutrition Research

Critical research on women's health outcomes focusing on undernutrition patterns and determinants. Thiswork contributes to understanding gender-specific health challenges in developing countries.

Research Applications:

• Women's nutritional status assessment
• Health determinants analysis
• Gender-focused health research
• Public health intervention planning

Shiny Interactive Applications

A collection of interactive web applications built with R Shiny framework. This repository demonstratesexpertise in creating dynamic, user-friendly data visualization and analysis tools for web deployment.

Features:

• Interactive dashboards and data visualization
• Real-time data processing and analysis
• User-friendly web interfaces for complex data analysis
• Multiple example applications showcasing different Shiny capabilities

Shiny Gallery Applications

An extended collection of Shiny applications demonstrating advanced interactive visualization techniquesand dashboard development. This project showcases the ability to create publication-ready interactiveapplications.

Applications:

• Advanced data visualization techniques
• Interactive statistical analysis tools
• Professional dashboard development
• Multi-page application architecture

Statistical Analysis Toolkit

A comprehensive collection of statistical analysis projects demonstrating expertise in applied statistics,hypothesis testing, and statistical modeling for research and business applications.

Key Components:

• Advanced statistical modeling techniques
• Hypothesis testing and experimental design
• Bayesian analysis and inference
• Time series analysis and forecasting

Demographic Data Analysis

Specialized projects focusing on demographic analysis, population studies, and social science research.This work demonstrates expertise in handling large-scale demographic datasets and conducting policy-relevant research.

Research Area:

• Population dynamics and demographic transitions
• Social determinants of health outcomes
• Migration patterns and urbanization studies
• Policy impact assessment and evaluation

Health Data Analytics

A portfolio of healthcare analytics projects focusing on public health outcomes, disease surveillance, andhealth system performance evaluation.

Project Scope:

• Disease pattern analysis and epidemiological modeling
• Health system performance indicators
• Clinical data analysis and outcomes research
• Public health surveillance systems

Educational Research Projects

Projects focused on educational outcomes research, student performance analysis, and institutionaleffectiveness studies in higher education settings.

Research Focuss:

• Student success predictive modeling
• Educational outcome evaluation
• Institutional research and analytics
• Academic program assessment

Consulting and Client Projects

A collection of professional consulting projects demonstrating ability to work with diverse stakeholdersand deliver actionable insights for business and organizational decision-making.

Project Types:

• Business intelligence and analytics solutions
• Market research and customer analytics
• Organizational performance evaluation
• Strategic planning and policy analysis

• Programming Languages: Python, R, JavaScript, HTML, CSS
• Machine Learning: TensorFlow, PyTorch,Scikit-learn, LangChain
• Data Analysis: Pandas, NumPy, dplyr, ggplot2
• Databases: Pinecone, SQL
• WebFrameworks: Flask, Shiny
• Cloud & Deployment: Local and cloud deployment strategies
• Visualization: Matplotlib, Seaborn, ggplot2, interactive dashboards

• Natural Language Processing and Large Language Models
• Computer Vision applications in agriculture and healthcare
• Statistical analysis for social impact and policy research
• AI-powered research tools and academic applications
• Data-driven solutions for international development