Food Recommendation System Using Big Data Based on Scoring Taste Adjectives

It is a general practice to evaluate food taste based on sensory tests, however, this test method’s disadvantage is that a lot of cost and time is required and significant deviation is taken place depending on each evaluator as well. Food taste evaluation by utilizing SNS-based big data for supplementing this disadvantage is considered to be a new challenge and innovative method. The objective of this study is to suggest a system that evaluates and recommends the level of domestic food taste by not only clustering food preference using k-means algorithm after sorting out food-related tweet contents from typical twitter of SNS, followed by scoring taste adjectives being mainly used in daily life by using rough set and selecting food-related adjectives among the scored adjectives, but also exploring the level of salty, sour, savory, bitter and sweet tastes through perception map.