Allergen detection model and Ingredient similarity analysis

Keywords: Allergen, detection

Background

This study employs artificial neural networks (ANN) to build an effective allergen detection model and analyze ingredient similarity. The model accurately identifies allergens in food products based on their ingredients, while the ingredient similarity analysis identifies similar ingredients. The research aims to enhance food safety and facilitate the development of safer options for individuals with allergies. Choosing the most relevant features or representations of the ingredient information that capture the necessary characteristics for allergen detection and ingredient similarity analysis can be complex. Identifying informative features that effectively differentiate allergen-containing products and measure ingredient similarities is important. To develop a reliable allergen detection model using ANN that accurately identifies allergens in food products based on ingredient information. Additionally, to conduct ingredient similarity analysis using ANN to identify ingredients that share similar characteristics. Sklearn library is used to import data and preprocess the data while the Keras library is used for building and training deep learning models, making it a popular choice for CNN implementations. While training the model 20 epoch runs and training loss is 0.0925 and accuracy is 0.9654. whereas test loss is 0.1522 and test accuracy is 0.9250.

Aim & Objectives

Aim
To detect the Allergen Model using DL Techniques
Objective

Develop an ANN model to detect allergens in food products, ensuring accurate and reliable allergen identification.
Create an ANN system that analyzes ingredient data to provide similarity analysis, aiding in recipe development.
Evaluate the ANN's allergen detection accuracy and ingredient similarity predictions, supporting safe and informed consumer choices.

Research Methodology

ANN Technique used

Software & Tools

Jupyter, Google Collab

— Preview truncated. The complete document (full chapters, references, diagrams, and appendices) is shared with clients as part of project delivery. —