本研究旨在根據衛生福利部國民健康署之「學校午餐食物內容及營養基準」(以下簡稱學校午餐基準)並透過大數據分析以及機器學習方法，建置一套自動化菜單規劃系統，以降低學校午餐菜單開立者在規劃菜單的作業時間，並且滿足營養均衡攝取、菜色多樣性以及食材成本的考量。自動化菜單規劃系統包含：菜單組合(menu combination)與菜單優化(menu optimization)兩個演算法。菜單組合演算法是透過基於內容過濾(content-based filtering, CBF)方法篩選合適的食譜來組合菜單，根據菜單開立者在設計菜單時的考量因素做為系統組合菜單時挑選食譜的依據，透過食譜特徵萃取演算法將食譜資料中的顏色、口味、食材等特徵轉化成數值資訊，並計算各食譜之間在各特徵上之相似度以排除相似的食譜，提高菜單中的菜色多樣性。此外，自動化菜單規劃系統會根據菜單開立者設定的菜單成本範圍來規劃菜單。菜單優化流程是將菜單組合演算法產出的菜單透過多目標優化的方法，調整菜單所用食材的重量，使菜單在營養成份分布符合學校午餐基準，以及在食材總和成本符合設定之範圍。最後將優化後滿足營養均衡攝取、菜色多樣性以及食材成本之菜單儲存至雲端系統，供菜單開立者參考與選用。本研究利用菜單設計滿意度問卷收集學校午餐菜單開立者對於自動化菜單規劃系統所設計之菜單在菜色多樣性和食材成本方面的滿意度回饋，問卷評分使用李克特量表(Likert scale)之七點量表進行衡量。透過模擬菜單開立者使用自動化菜單規劃系統開菜以及菜單設計滿意度問卷，其分析結果顯示：(1)自動化菜單規劃系統可滿足在菜單設計時的營養成份以及成本上的考量，於營養成份上的每日平均誤差，除了未設定的乳品類在各學校級別包含國小、國中以及高中上有2.75、2.74以及2.75份的誤差之外，其他六大類食物份量之誤差皆為0；在菜單食材成本與設定的成本目標值方面，各個學級的平均誤差為0.85、0.34以及0.23元，(2) 以自動化菜單規劃系統模擬開立一周的菜單20次需的平均時間及標準差如下：小學: 8.098±1.602秒、國中: 8.104±1.847秒、高中: 6.711±1.711秒，(3)菜單開立者對於自動化菜單規劃系統所設計之菜單在菜色多樣性以及成本上的滿意度在7分中各項評分介於5.6分至5.8分之間。根據上述結果，本研究所開發之自動化菜單規劃系統可設計符合營養均衡攝取、菜色多樣性以及食材成本之菜單。

This study aims to develop an automated menu planning system, rooted in the guidelines of the "School Lunch Food Content and Nutritional Standards" provided by the Health Promotion Administration of the Ministry of Health and Welfare, hereafter referred to as the School Lunch Standards. The system harnesses the power of big data analysis and machine learning techniques to streamline the menu planning process in schools. Its primary objective is to reduce the time and effort required by menu planners while ensuring the fulfillment of key considerations, including balanced nutritional content, menu diversity, and ingredient cost control. The automated menu planning system is comprised of two essential algorithms: the menu combination algorithm and the menu optimization algorithm. In the menu combination algorithm, a content-based filtering (CBF) approach is employed to curate suitable recipes for menu construction. These recipe selections are based on factors considered by menu planners during the menu design phase. An algorithm for feature extraction from recipes is used to extract attributes such as color, flavor, and ingredients into numerical data. This allows for the calculation of similarities between recipes across multiple features, resulting in the exclusion of overly similar choices and thereby enhancing menu diversity. Moreover, the automated menu planning system takes into account the cost constraints established by menu planners when designing menus. In the menu optimization algorithm, the quantities of ingredients used in the menu, as determined by the menu combination algorithm, are fine-tuned through a multi-objective optimization approach. This adjustment aims to ensure that the menu adheres to the nutritional guidelines set forth in the School Lunch Standards while staying within the defined budget for ingredients. The optimized menus, which successfully meet the requirements of balanced nutrition, menu variety, and ingredient cost control, are stored in a cloud-based system for menu planners' easy access and selection. To evaluate the effectiveness of the system, satisfaction surveys were conducted among school lunch menu planners. These surveys, using a seven-point Likert scale, gathered feedback regarding menu variety and ingredient cost for menus generated by the automated system. The analysis of survey responses yielded the following key findings: 1. The automated menu planning system effectively aligns with considerations of nutritional content and cost during menu design. Daily average errors in nutritional content, except for dairy products, remained at 2.75 portions for primary schools, 2.74 for junior high schools, and 2.75 for high schools. Errors in the other six major food categories were negligible. Regarding menu ingredient cost compared to the set cost target, the average errors for each school level were 0.85, 0.34, and 0.23 NT dollars, respectively. 2. When simulating the creation of a week's worth of menus 20 times using the automated menu planning system, the average time required and standard deviation were as follows: primary schools: 8.098 ± 1.602 seconds, junior high schools: 8.104 ± 1.847 seconds, high schools: 6.711 ± 1.711 seconds. 3. Menu planners expressed a high level of satisfaction with the menus designed by the automated system, particularly in terms of menu variety and cost. Ratings on the seven-point scale consistently fell between 5.6 and 5.8. In conclusion, this study demonstrates that the developed automated menu planning system effectively designs menus that fulfill the criteria of balanced nutrition, menu diversity, and ingredient cost control, streamlining the menu planning process for schools.

[1] A. Felfernig, M. Jeran, G. Ninaus, F. Reinfrank, S. Reiterer, and M. Stettinger, “Basic Approaches in Recommendation Systems,” in Recommendation Systems in Software Engineering, pp. 15–37, 2013.
[2] A. Gunawardana and G. Shani, “A survey of accuracy evaluation metrics of recommendation tasks,” Journal of Machine Learning Research, vol. 10, pp. 2935–2962, Dec 2009.
[3] A. Parasuraman, V. A. Zeithaml, and L. L. Berry, “SERVQUAL: A multiple-item scale for measuring consumer perceptions of service quality,” Journal of Retailing, vol. 64, no. 1, pp. 12–40, 1988.
[4] B. Patel, P. Desai and U. Panchal, "Methods of recommender system: A review," in 2017 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS), Coimbatore, India, 2017, pp. 1-4.
[5] D. Goldberg, D. Nichols, B. M. Oki, and D. Terry, “Using collaborative filtering to weave an information tapestry,” Communications of the ACM, vol. 35, no. 12, pp. 61–70, Dec. 1992.
[6] E. Segredo, G. Miranda, J. M. Ramos, C. León, and C. Rodríguez-León, “SCHOOLTHY: Automatic Menu Planner for Healthy and Balanced School Meals,” IEEE Access, vol. 8, pp. 113200–113218, 2020.
[7] F. Ricci, L. Rokach, B. Shapira, and P. B. Kantor, Eds., Recommender Systems Handbook, Boston, MA: Springer US, 2011.
[8] F.F. Kuo, C.T. Li, M.K. Shan, and S.Y. Lee, “Intelligent menu planning: recommending set of recipes by ingredients,” in Proceedings of the ACM multimedia 2012 workshop on Multimedia for cooking and eating activities, Nara, Japan, 2012, pp.1-6.
[9] J. Bobadilla, F. Ortega, A. Hernando, and A. Gutiérrez, “Recommender systems survey,” Knowledge-Based Systems, vol. 46, pp. 109–132, Jul. 2013.
[10] J. L. Balintfy, “Menu planning by computer,” Commun. ACM, vol. 7, no. 4, pp. 255–259, Apr. 1964.
[11] J.B. Schafer, J. Konstan, J. Riedl, "E-commerce recommendation applications", Data Mining and Knowledge Discovery, vol. 5, pp.115-153, 2001.
[12] K. M. Morrison, S. Shin, M. Tarnopolsky, and V. H. Taylor, “Association of depression & health related quality of life with body composition in children and youth with obesity,” Journal of Affective Disorders, vol. 172, pp. 18–23, Feb. 2015.
[13] L. Wang, Q. Li, N. Li, G. Dong, and Y. Yang, “Substructure similarity measurement in chinese recipes,” in Proceedings of the 17th international conference on World Wide Web, Beijing China, 2008, pp.979–988.
[14] M. Elena. Renda and U. Straccia, “A personalized collaborative Digital Library environment: a model and an application,” Information Processing & Management, vol. 41, pp. 5–21, Jan 2005.
[15] M. Loesdau, S. Chabrier, and A. Gabillon, “Hue and Saturation in the RGB Color Space,” in 2014 International Conference on Image and Signal Processing (ICISP), Cherbourg, France, 2014, pp. 203–212.
[16] Nadamoto, S. Hanai, and H. Nanba, “Clustering for Similar Recipes in User-Generated Recipe Sites Based on Main Ingredients and Main Seasoning,” in 2016 19th International Conference on Network-Based Information Systems (NBiS), Ostrava, Czech Republic, 2016, pp. 336-341.
[17] P. B. Thorat, R. M. Goudar, and S. Barve, “Survey on Collaborative Filtering, Content-based Filtering and Hybrid Recommendation System,” International Journal of Computer Applications, vol. 110, no. 4, pp. 31–36, Jan. 2015.
[18] P. Jaccard, “The Distribution of The Flora in The Alpine Zone,” New Phytologist, vol. 11, no. 2, pp. 37–50, Feb. 1912
[19] Q. Li, W. Chen, and L. Yu, “Community-based recipe recommendation and adaptation in peer-to-peer networks,” in Proceedings of the 4th International Conference on Uniquitous Information Management and Communication (ICUIMC '10), Suwon, Korea, 2010, pp.1-6.
[20] R. Likert, “A Technique for the Measurement of Attitudes”, Archives of Psychology, vol. 140, pp. 1–55, 1932.
[21] R. P. C. Moreira, E. Wanner, F. V. C. Martins, and J. F. M. Sarubbi, “An Evolutionary Mono-Objective Approach for Solving the Menu Planning Problem,” in 2018 IEEE Congress on Evolutionary Computation (CEC), Rio de Janeiro, Brazil, 2018, pp. 1–8.
[22] S. Tamaru, H. Taki, R. Usuki, and T. Nakanishi, “Recipe Recommendation Method by Similarity Measure with Food Image Recognition,” in Proceedings of the 6th International Conference on Information System and Data Mining, Silicon Valley CA, USA, 2022, pp.81–88.
[23] S. Tan, J. Bu, C. Chen, B. Xu, C. Wang, and X. He, “Using rich social media information for music recommendation via hypergraph model,” ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 7, pp. 1–22, Oct 2011.
[24] T. N. Trang Tran, M. Atas, A. Felfernig, and M. Stettinger, “An overview of recommender systems in the healthy food domain,” Journal of Intelligent Information Systems, vol. 50, pp. 501–526, Jun 2017.
[25] U. M. Umoren, B. A. Ojokoh, and O. S. Ijarotimi, “Addressing Malnutrition in School-Aged Children with a Diet Recommender System,” International Journal of Innovative Science and Research Technology, vol. 7, no. 9, 2022.
[26] W. H. Organization, Nutrition action in schools: a review of evidence related to the nutrition-friendly schools initiative, World Health Organization, 2020.
[27] W. Min, B.-K. Bao, S. Mei, Y. Zhu, Y. Rui, and S. Jiang, “You Are What You Eat: Exploring Rich Recipe Information for Cross-Region Food Analysis,” IEEE Transactions on Multimedia, vol. 20, no. 4, pp. 950–964, Apr. 2018.
[28] Y. Chung, “Finding food entity relationships using user-generated data in recipe service,” in Proceedings of the 21st ACM international conference on Information and knowledge management, Maui Hawaii, USA, 2012, pp.2611–2614.
[29] Y. Pinxteren, G. Geleijnse, and P. Kamsteeg, “Deriving a recipe similarity measure for recommending healthful meals,” in Proceedings of the 16th international conference on Intelligent user interfaces, Palo Alto CA, USA, 2011, pp. 105–114.
[30] Z. Huang, W. Chung, and H. Chen, “A graph model for E-commerce recommender systems,” Journal of the American Society for Information Science and Technology, vol. 55, no. 3, pp. 259–274, 2004.
[31] 沈柏毅,“基於多目標優化之學校午餐智慧開菜系統,” M.S. Department of Electrical Engineering, National Cheng Kung University, Tainan, Taiwan, 2021.
[32] 張中榕, "以IPGA模式探討國民小學學校午餐服務品質之研究-以臺中市為例," M.S. thesis, Department of Techonolgy Management, National Chung Hua University, Hsinchu, Taiwan, 2016.
[33] 紹建華、劉敬賢、李冠嘉, 烹飪實務教戰守策, 生活家, Taiwan, Taichung, 2003.
[34] 曾文燦、高鋼輝、林后儀、張桂秋、廖成文、陳正忠、徐近平、曾瓊慧、李怡君, 中餐烹調原理與實務, 第一版, 華格納, Taiwan, Taichung, 2009.
[35] 曾詩耕,“以多目標遺傳演算法設計個人適性化菜單,” M.S. thesis, Department of Computer Science and Information Engineering, National Chung Hua University, Hsinchu, Taiwan, 2012.
[36] 黃韶顏、蔡蘭芳, 團體供膳—盤餐菜單設計, 第二版, 華香園, Taiwan, Taipei, 1993.
[37] 黃建誠, "學校午餐公辦公營及外訂餐盒之滿意度 差異研究—以桃園市國民小學為例," M.S. thesis, Department of Information Management & Graduate School of Information Management, Vanung University, Taoyuan, Taiwan, 2018.
[38] 楊景昭, 中華廚藝：理論與實務, 華杏, Taiwan, Taipei, 2005.
[39] 楊淑君, "國小學校午餐服務品質與滿意度之研究－以臺南市為例," M.S. thesis, Department of Food and Beverage Management, Far East University, Tainan, Taiwan, 2021.
[40] 賈永喆, "國民小學學校午餐之滿意度與重要度分析—以桃園縣國民小學為例," M.S. thesis, Department of Applied Mathematics, National Hsinchu University of Education, Hsinchu, Taiwan, 2007.
[41] 蔡永富, "國民小學學校午餐服務品質滿意度之調查研究," M.S. thesis, Department of Food Science and Technology, Tajen University, Pingtung, Taiwan, 2013.
[42] “「108年度新學校午餐創新推動計畫」智慧化校園餐飲服務平臺：規劃與建置_期末報告書,” 2019.
[43] “「111年度新學校午餐創新推動計畫」智慧化校園餐飲服務平臺(含食登平臺2.0)：教育與訓練_期末報告書,” 2022.
[44] “MariaDB foundation,” MariaDB.org, 13-Nov-2019. [Online]. Available: https://mariadb.org/. [Accessed: 20-Sep-2023].
[45] “The PHP framework for web artisans,” Laravel. [Online]. Available: https://laravel.com/. [Accessed: 20-Sep-2023].
[46] “Vue.js - the progressive javascript framework: Vue.js,” Vue.js - The Progressive JavaScript Framework | Vue.js. [Online]. Available: https://vuejs.org/. [Accessed: 20-Sep-2023].