Halal Food Prediction Using Long Short-Term Memory Neural Networks

This research investigates the application of Long Short-Term Memory (LSTM) neural networks to predict the halal status of food products based on the ingredient lists. It utilized a comprehensive data set containing 39,787 food products, with 21,826 labeled as halal and 17,961 labeled as haram, the study aimed to develop an efficient and accurate classification model. Each product's data includes a list of ingredients and a corresponding halal or haram label. The LSTM model was trained and evaluated on this data set, achieving an impressive performance metrics. For class 0 (halal), the model attained a precision, recall, and F1-score of 0.99, with a support of 21,826 instances. Similarly, for class 1 (haram), the precision, recall, and F1-score were also 0.99, with a support of 17,961 instances. These results indicate that the model can accurately predict the halal status with high reliability. The overall accuracy of the model stands at 99%, underscoring its robustness and efficacy in handling both classes. This research demonstrates the potential of LSTM neural networks in automating the halal status identification, offering a scalable solution to address the growing demand for halal products in the market.