BACKGROUND
Major Depressive Disorder is primarily treated with medication, but antidepressants have side effects that affect the patient's body and lifestyle. Therefore, alternative methods for treating depression are applying nutrients to help treat symptoms of depression better.
OBJECTIVE
This research presents the design of a food menu recommendation system for depressed patients with malnutrition. The system focuses on the ability to infer food menus for depressive patients with physical ailments that affect their nutritional status.
METHODS
The menu inference rules have been designed using ontology based on descriptive logic and using SPARQL to query the menus. The system was evaluated by randomly selecting 10 case studies based on different physical ailment conditions of depressed patients. Then, menu results from each case were sent to seven nutrition experts to assess the consistency of the menus with patients' conditions.
RESULTS
The evaluation of the overall model performance applied standard metrics, which was F-measure. The results obtained from this experiment showed that the designed system could recommend food menus consistent with patients' physical and mental illnesses with an average F-measure of 0.93.
CONCLUSIONS
The experimental results showed that the designed system could recommend a menu consistent with the patient's physical and mental illnesses, with an overall performance. The system could be applied to new food menus whose suitability for depression symptoms was unknown.