Thai Ingredient Replacement Recommendation using Taxonomy

Cooking preparation is linked with ingredient avail-ability. However, there is often the case that we could not prepare all the listed ingredients in the recipe. It might be because some ingredients are not available in the kitchen or personal preferences. In this case, users will want to use another ingredient for replacement. Therefore, recommending other ingredients that are similar to an exchanged ingredient and suitable with the remained ingredient in the recipe for substitution is necessary. In this work, we present a new approach for improving the performance of alternative ingredient recommendations for Thai recipes. The proposed approach is based on the combination of the use of ingredient taxonomy, the characteristics of ingredients, the Density-Based Spatial Clustering of Application with Noise method (DBSCAN), and similarity function. Our proposed approach is evaluated on the Thai recipe domain and compared with our prior research on the alternative ingredient recommender system, including through the cooking and tasting experiment. The evaluation results show that our approach further improves the performance of alternative ingredient recommendations. It increases user satisfaction with recommendation lists in terms of appropriateness and compatibility. Moreover, some alternative ingredients can generate a new dish. This study can enhance both beginners and experts can cook Thai food conveniently and creatively by using alternativeingredient.