Indonesia is currently experiencing the Double Burden Nutrition problem. This is a problem when significant proportion of population is endangered with the problem of malnutrition, but a significant other suffer from obesity or over-nutrition. People who are underweight, overweight, and obesity are included in the ten risks in terms of the global disease burden. In 2015 showing that the adult population in the world is overweight, with 38% of men and 40% of women. This study aims to detect over nutrition, undernutrition in obese sufferers to determine their nutritional status with 24 hours food recall method for assessing theur consumption and using data mining algorithms, namely: K-Nearest Neighbor, Naive Bayesian Classification, and Decision Tree for data mining calculations. From the results obtained with the help of RapidMiner tools, it can be concluded that not all obese sufferers are over nutritioned, they can also experience undernutrition or normal nutrition. From the results of this study, the best accuracy for calculating nutritional status is using the Naive Bayesian Classification algorithm with an accuracy rate of 100%, then the Decision Tree with an accuray rate of 86,67% and the last is K-Nearest Neighbor with an accuracy rate of 73,33%.