| 摘要: |
| 目的 针对饮食不均衡和搭配失当而造成的饮食问题给人们身心健康带来不良影响,尤其是饮食对慢性病
研究的发展等问题,为了设计和开发一个能够根据用户个人状况、喜好、口味等因素,提出向用户推荐符合其身体
需求的营养膳食的智能系统。 方法 搭建了一个基于人工智能知识库的营养膳食推荐系统,利用专家验证的膳食的
明确数据集,结合基于推理决策支持系统( RDSS) 和营养计划( NP) 运用 NAct 本体论提供高度准确的饮食计划,跨
越 10 个用户组,包括健康的受试者和有健康状况的参与者。 结果 该系统的有效性通过广泛的实验进行评估,评估
涉及合成数据,包括生成 3 000 个虚拟用户档案和他们的每周膳食计划。 结果显示,在大多数用户类别中,推荐适
当成分的精确度和召回率都很高,而膳食计划生成器对所有营养素的推荐达到了 94%的总推荐精确度。 结论 基
于人工智能知识库的营养膳食推荐系统可以根据用户的身体状况、喜好、饮食禁忌等方面进行个性化推荐。 这样
的个性化推荐能够更好地满足用户的需求,从而提高推荐的准确度。 专家知识库包含了广泛的营养学和健康知
识,这些知识可以帮助系统识别出最适合用户的膳食方案。 这些方案不仅可以提供充足的营养,还可以避免与用
户的健康状况不兼容的食物或成分。 |
| 关键词: 人工智能 营养膳食 营养计划 推荐系统 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Research on Nutritional Dietary Recommendation System Based on Artificial Intelligence Knowledge Base |
|
ZHANG Yuelin,WANG Chuangjian
|
|
1. Key Laboratory of Metallurgical Equipment and Control Ministry of Education Wuhan University of Science and
Technology Wuhan 430081 China
2. Hubei Provincial Key Laboratory of Mechanical Transmission and Manufacturing Engineering Wuhan University of
Science and Technology Wuhan 430081 China
|
| Abstract: |
| Addressing the adverse effects of dietary problems caused by imbalanced nutrition and poor food
combinations on people?? s physical and mental well-being particularly concerning research on the relationship between
diet and chronic diseases an intelligent system capable of recommending sound and nutritionally balanced meals tailored
to users?? conditions preferences and tastes was designed and developed. Methods A nutrition meal recommendation
system based on an artificial intelligence knowledge base was constructed. It utilized a definitive dataset of diets validated
by experts combined with a reasoning decision support system RDSS and a nutritional plan NP employing NAct
ontology to provide highly accurate dietary plans. It spanned 10 user groups including healthy subjects and participants
with health problems. Results The effectiveness of the system was assessed through extensive experiments which involved synthetic data including the generation of 3 000 virtual user profiles and their weekly dietary plans. The results indicated
high precision and recall rates for recommending appropriate ingredients across most user categories with the dietary plan
generator achieving an overall recommendation accuracy of 94% for all nutrients. Conclusion The nutrition meal
recommendation system based on an artificial intelligence knowledge base can personalize recommendations based on
users?? physical conditions preferences dietary restrictions and other aspects. Such personalized recommendations can
better meet users ?? needs thereby enhancing the accuracy of the recommendations. The expert knowledge base
encompasses extensive nutrition and health knowledge which can assist the system in identifying the most suitable dietary
plans for users. These plans not only provide ample nutrition but also avoid foods or ingredients incompatible with users??
health conditions. |
| Key words: artificial intelligence nutritional diet nutrition plan recommendation system |
