Halal Food Prediction Using the Similarity Graph Algorithms

Halal food is food allowed by Islamic sharia. On the contrary, haram food is not permitted, such as alcohol, pork, blood, carrion, and meat not slaughtered according to sharia. Based on Article 39 of Law Number 33 of 2014 concerning halal product guarantees, halal certificates are issued by the Halal Product Guarantee Agency (BPJPH) in Indonesia. The halal certification guarantees that food has a composition containing halal ingredients. However, many food products still do not have a halal certificate. Therefore, it is necessary to estimate the halal status of food products that are still not certified. In this work, we predict the halal status of food using graph similarity algorithms. In this case, we acquire products from the Indomaret website. The product data contains the product name, the composition of the food product, and the manufacturer. Moreover, we crawl the halal food database on Halal MUI website. Both datasets are merged into a single dataset based on the products name. Then, the similarity algorithms such as Jaccard similarity, Approximate Nearest Neighbor, Adamic Adar and Preferential Attachment are performed amongst products in the dataset. F-measure evaluate the accuracy of each algorithm.