seyibuilds/auspify-ml-internship

How seyibuilds/auspify-ml-internship works

A set of four standalone Netflix-catalogue machine-learning tasks: recommendation, content-type prediction, segmentation, and audience-band analytics. The user runs a task script (or the documented all-tasks runner); each loads and cleans the catalogue through shared feature engineering, then produces task-specific results. The README documents model comparisons, recommendation evaluation, cluster profiles, automated insights, figures, and reports. Arrows reflect excerpted calls or explicit README documentation; unsampled implementation details are not inferred.

Main entry points: task1_recommendation.py Recommendation Task · task2_content_type.py Type Prediction Task · task4_segmentation.py Segmentation Task · common.py Load and Clean

Components 21

Shared Data Prep

Netflix Catalogue data/Dataset.csv
Load and Clean src/common.py
Shared Features src/common.py

Recommendations

Recommendation Task src/task1_recommendation.py
Weighted Content Text src/task1_recommendation.py
Similarity and Ranking src/task1_recommendation.py
Recommendation Evaluation src/task1_recommendation.py

Type Prediction

Type Prediction Task src/task2_content_type.py
Leakage-safe Features src/task2_content_type.py
Model Comparison src/task2_content_type.py
Feature Ablation src/task2_content_type.py

Content Segmentation

Segmentation Task src/task4_segmentation.py
Cluster Features src/task4_segmentation.py
Cluster Selection src/task4_segmentation.py
Cluster Profiles src/task4_segmentation.py

Success Analytics

Analytics Task src/task6_analytics_engine.py
Advanced Features src/task6_analytics_engine.py
Audience Models src/task6_analytics_engine.py
Insights and Dashboard src/task6_analytics_engine.py

Other components

Analyst
Reports and Figures outputs

Connections 27

Analyst → Recommendation Task: runs task README.md
Analyst → Type Prediction Task: runs task README.md
Analyst → Segmentation Task: runs task README.md
Analyst → Analytics Task: runs task README.md
Netflix Catalogue → Load and Clean: loads README.md
Load and Clean → Shared Features: engineers src/common.py
Recommendation Task → Load and Clean: loads via src/task1_recommendation.py
Type Prediction Task → Load and Clean: loads via src/task2_content_type.py
Segmentation Task → Load and Clean: loads via src/task4_segmentation.py
Analytics Task → Load and Clean: loads via src/task6_analytics_engine.py
Recommendation Task → Weighted Content Text: builds src/task1_recommendation.py
Weighted Content Text → Similarity and Ranking: vectorizes for src/task1_recommendation.py
Recommendation Task → Similarity and Ranking: requests recommendations src/task1_recommendation.py
Recommendation Task → Recommendation Evaluation: evaluates README.md
Recommendation Task → Reports and Figures: writes results README.md
Type Prediction Task → Leakage-safe Features: builds src/task2_content_type.py
Type Prediction Task → Model Comparison: trains and compares src/task2_content_type.py
Type Prediction Task → Feature Ablation: runs ablation src/task2_content_type.py
Type Prediction Task → Reports and Figures: writes results README.md
Segmentation Task → Cluster Features: builds src/task4_segmentation.py
Segmentation Task → Cluster Selection: selects cluster count src/task4_segmentation.py
Segmentation Task → Cluster Profiles: profiles clusters src/task4_segmentation.py
Segmentation Task → Reports and Figures: writes reports src/task4_segmentation.py
Analytics Task → Advanced Features: engineers src/task6_analytics_engine.py
Analytics Task → Audience Models: trains and compares src/task6_analytics_engine.py
Analytics Task → Insights and Dashboard: generates src/task6_analytics_engine.py
Insights and Dashboard → Reports and Figures: writes report and dashboard README.md

Drawn by GitDiagram on November 15, 2026 from the repository's file tree, README and source files. Read as Markdown ·