Food Waste Reduction Assistant — Deep Learning Project

Overview

A deep learning project built for MSc Big Data & Analytics at Griffith College Dublin. Two-part system combining computer vision and NLP to assist with food waste reduction.

Results

Model	Task	Accuracy
MobileNetV2 (Transfer Learning)	Food image classification	84.3%
LSTM	News text classification (AG News)	90.5%
DistilBERT	News text classification (AG News)	92.0%

Part 1 — Image Classification

• Dataset: Food-11 (11 food categories)
• Model: MobileNetV2 with transfer learning and fine-tuning
• Framework: TensorFlow / Keras

Part 2 — Text Classification

• Dataset: AG News (4 categories)
• Models: LSTM and DistilBERT
• Framework: TensorFlow / Keras / HuggingFace

Tech Stack

Python · TensorFlow · Keras · scikit-learn · Pandas · NumPy · Matplotlib · JupyterLab

Topics Covered

Transfer learning · CNNs · RNNs · LSTMs · Transformers · Data augmentation · Confusion matrices · Training curves