Halal Restaurant Integration Using Bidirectional Recurrent Neural Networks

Salsa Putri Islammia, Nur Aini Rakhmawati


Indonesia has the largest Muslim population in the world, so it is mandatory to consume halal food. However, there are still many websites that do not provide information on halal restaurants, such as Google Map. Data integration is needed to generate broader information and ensure the suitability of halal restaurants in several different data sources, such as the Indonesia Halal Product Assurance Agency (BPJPH) and Google Maps. The two datasets were cleaned during the preprocessing stage and then labelled using the Jaccard index. Finally, the Bidirectional Recurrent Neural Networks (BRNN) model was built using deepmatcher and evaluated using the F1-score metric. The integration of the two data produces 155 rows of matching data pairs.


bidirectional recurrent neural networks; halal; data integration; restaurant

Full Text:


DOI: https://doi.org/10.18461/ijfsd.v15i1.I5

ISSN 1869-6945


This work is licensed under a Creative Commons License