Skip to content

dalida-dikici/TGSS-2024-Smart-Analysis-Lab

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

2 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸ”¬ TGSS 2024: AI-Powered Smart Analysis Lab

Transforming Raw Social Data into Actionable Strategic Insights

Python Streamlit Machine Learning

🎯 Project Overview

This project is a Decision Support System designed to analyze the Turkish General Social Survey (TGSS-2024). It bridges the gap between complex statistical data and executive decision-making by using Machine Learning to identify the "Root Causes" of social trends.

Why this matters: In a world of "thorny business challenges," simply seeing data isn't enough. This tool identifies the North Star metrics of social indicators and provides synthesized recommendations.

πŸš€ Key Features (Business & Technical)

  • Predictive Modeling (Root Cause Analysis): Utilizes Random Forest algorithms to determine feature importanceβ€”identifying exactly which variables (age, income, education, etc.) most significantly impact social outcomes.
  • Automated Executive Insights: A custom-built "Expert Commentary" module that translates statistical correlations into human-readable business language.
  • Dynamic Data Pipeline: Seamlessly handles SPSS (.sav) metadata, mapping complex survey labels to intuitive interactive dashboards.
  • Impact Simulation: Allows users to visualize how shifts in specific demographics could influence overall social satisfaction metrics.

πŸ› οΈ Tech Stack

  • Engine: Python 3.x
  • Analysis: Pandas, NumPy, Scikit-Learn (Random Forest Regressor/Classifier)
  • Frontend: Streamlit (Interactive Web UI)
  • Visualization: Plotly Express for high-fidelity interactive charting
  • Data Source: Turkish General Social Survey (TGSS-2024)

πŸ“Š Business Logic & Methodology

  1. Extraction: Cleaning and preprocessing high-dimensional survey data.
  2. Modeling: Training a Random Forest model to rank "Predictor Importance."
  3. Synthesis: Zooming out from raw coefficients to develop compelling, synthesized recommendations for policy-makers and stakeholders.

πŸŽ“ Acknowledgments

I was introduced to the TGSS-2024 dataset during the Data Analysis School, organized by the Council of Higher Education (YΓ–K) and the Institute of Population and Social Research. This project aims to bring a modern, AI-driven perspective to the valuable data provided by these institutions.

🎬 Demo & Documentation


πŸ“© Let's Connect: I am a Data Scientist focused on turning complex datasets into strategic advantages. For consultancy or collaboration, feel free to reach out via https://www.linkedin.com/in/dalida-dikici/ or DM.

About

πŸ”¬ AI-powered Decision Support System for TGSS-2024 data. Designed for root cause analysis and executive-ready social insights using Random Forest & Streamlit. πŸš€

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages