201033926 六、發明說明： 【發明所屬之技術領域】 5 10 15 參 20 本發明係關於一種飲食建議系統、提供一飲食建議的 方法及可攜式電子裝置。此飲食建議系統、提供一飲食建 議的方法及可攜式電子裝置係藉由模糊多屬性運 出一建議料理清單，使得使用者可得到一不但符合其飲食 偏好，更可達到其營養需求的建議料理建議清單。 【先前技術】 由於營養師需要參考患者每曰的飲食紀錄、此位患者 的個人狀況（如目前身體狀況、疾病史）以及同時參考國人營 養成分攝取量（dietary reference — DRIs)，才能提出適 合此位患者的飲食建議，造成營養師的工作負荷量極大。 所以’營養師對於-種能協助他們提出適#飲食建議之飲 食建議系統需求度頗高。 但疋’目冑業界所能提供的營養師的輔助系统僅能將 所有會使用到的資料電子化以及將營養師與患者之間的諮 詢互動行為資訊化。雖然這些f訊化流程可提升病患填寫 飲食紀錄的方便性 '可將營養師所需資料的資訊化、可節 省整，程序所花費的紙張等，但是此種習知的輔助系統仍 、一提么、任何飲建議的功能’營養師本身仍需親自將所 有資料I過通盤考慮才能提出適合此位患者的飲食建議， 營養師的工作負擔仍相當沈重。況且，由於營養師需要考 慮的因素已經太多，其所提出的飲食建議往往無法顧慮到 3 201033926 患者個人的飲食偏好，且往往充滿專業性的敘述，如建議 者夕攝取蛋白質2g,維生素匚5 mg這一類的敘述，造成 患者遵循營養纟巾所提出之飲食建制意願極低，根本無法 達到此營養建議所欲達到的效果。 因此，業界需要一種可提供一不但符合使用者之飲食 偏好，更可達到使用者之營養需求的建議料理清單的營養 建議系統。 【發明内容】 θ本發明之主要目的係在提供一種飲食建議系統，俾能 提供不但4合使用者之飲食偏好，更可達到使用者之營 養需求的建議料理清單。 法 本發明之另一目的係在提供一種提供一飲食建議的方 15 20 俾能提供一不但符合使用者之飲食偏好，更可達到使 用者之營養需求的建議料理清單。 本發明之又一目的係在提供一種可攜式電子裝置，俾 ^隨時提供―不但符合❹者之飲食偏好，更可達到使用 者之營養需求的建議料理清單。 本發明之飲食建議系統，包括：一資料輸入模組係 =-飲食紀錄及—基本資料；一資料儲存模組係儲存 複數個料理營養成份資料及一 養兩求資料；-決策分析 自此上 糊多屬性運算單元，此決策分析模組係 輸入模組接受此飲食紀錄及此基本資料且自此 耗轉受此㈣㈣養成份:㈣及此營養需求資 4 201033926 料，此模糊多屬性運算單元再依據此飲食紀錄、此基本資 料、此等料理營養成份資料及此營養需求資料運算出一建 議料理清單；以及一資料輸出模組，係自此決策分析模組 接受並輸出此建議料理清單。 5 本發明之提供一飲食建議的方法，包括下列步驟：提 供一飲食建議系統，係包括一資料輸入模組、一資料儲存 模組、一決策分析模組及一資料輸出模組，且此資料儲存 模組儲存複數個料理營養成份資料及一營養需求資料此 ® #策分析模組具有-模糊多屬性運算單元；輸入一飲食紀 10錄及一基本資料至此資料輸入模組；此決策分析模組自此 資料輸入模組接受此飲食紀錄及此基本資料，且自此資料 儲存模組接受此等料理營養成份資料及此營養需求資料， 此模糊多屬性運算單元再依據此飲食紀錄、此基本資料、 此等料理營養成份資料及此營養需求資料運算出一建議料 15 理清單；以及此資料輸出模組輸出此建議料理清單。 本發明之可攜式電子裝置，栽有一飲食建議系統，此 ❿ 飲食建議系統包括：一資料輸入模組，係接受一飲食紀錄 及一基本資料；一資料儲存模組，係儲存複數個料理營養 成份資料及一營養需求資料；一決策分析模組，係具有— 20模糊夕屬性運异單元，此決策分析模組係自此資料輸入模 組接受此飲食紀錄及此基本資料，且自此資料儲存模組接 受此等料理營養成份資料及此營養需求資料，此模糊多屬 性運算單元再依據此飲食紀錄、此基本資料'此等料理營 養成份資料及此營養需求資料運算出一建議料理清單：以 5 201033926 及一資料輸出模組，係自此決策分析模組接受並輪出此建 議料理清單。 由於本發明之飲食建議系統之決策分析模組具有一模 糊多屬性運算單元，且此模糊多屬性運算單元係依據模糊 5 理論中之「語言變數」分別將營養師對於每一道料理之營 養成分含量的判斷以及及營養師對於本發明之飲食建議系 統之使用者（患者）的營養需求建議量予定量化與數學化，即 所謂的解模糊化。如此，本發明之飲食建議系統在輸出建 ® 議料理清單時，可有效避免營養師對於每道料理之營養成 10 分含量的判斷所不可避免的模糊性敘述（如含量極多、含量 很多、含量略多、含量中等、含量略少、含量很少及含量 極少）以及對於本發明之飲食建議系統之使用者（患者）的營 養需求建議量所不可避免的模糊性敘述（如非常大量、很大 量、大量、多量、加量、增量、適量、減量、少量、較少 15量、微量、很微量、非常微量）對此份建議料理清單所造成 的不良影響。另-方面，由於本發明之飲食建議系統可藉 〇 由其輸入模組接受使用者之飲食紀錄（如偏好之料理資料） 及基本資料（如使用者之飲食紀錄資料、身高、體重、年齡、 身體狀況、疾病史以及營養師提供之營養需求建議），本發 20明之飲食建議系統之決策分析模組在運算出建議料理清單 時可將這些資料通盤納入考慮，使得使用者可得到一份不 但符合其飲食偏好，更可達到其營養需求的建議料理清單。 另一方面，由於本發明之提供一飲食建議的方法係應 用本發明之飲食建議系統而提供使用者（患者）一飲食建 6 201033926 議，所以本發明之提供一飲食建議的方法可提供使用者一 不但符合其飲食偏好，更可達到其營養需求的建議料理建 議清單。同樣地，因本發明之可攜式電子裝置係載有本發 明之飲食建議系統，且本發明之飲食建議系統運作於本發 5明之可攜式電子裝置中，所以本發明之可攜式電子裝置可 隨時隨地提供使用者一不但符合其飲食偏好，更可達到其 營養需求的建議料理建議清單。 本發明之飲食建議系統之輸入模組可接受任何類別之 _ I本資料’其較佳包括飲食建議系統之使用者之飲食紀錄 ίο資料、身尚、體重、年齡、身體狀況、疾病史以及營養師 提供之營養需求建議。本發明之飲食建議系統之模糊多屬 性運算單元可應用任何種類之演算法來運算出一建議料理 /月單，其較佳係應用理想解類似度偏好法（Technique f〇r201033926 VI. Description of the invention: [Technical field to which the invention pertains] 5 10 15 Reference 20 The present invention relates to a dietary recommendation system, a method for providing a dietary recommendation, and a portable electronic device. The dietary suggestion system, the method of providing a dietary recommendation, and the portable electronic device deliver a list of recommended dishes by obscuring multiple attributes, so that the user can obtain a suggestion that not only meets their dietary preferences but also meets their nutritional needs. A list of recommendations for cooking. [Prior Art] Because the dietitian needs to refer to the patient's daily diet record, the patient's personal condition (such as current physical condition, disease history) and the reference to the national reference (DRIs), it can be proposed. The dietary recommendations of the patients caused the dietist's workload to be extremely high. Therefore, the dietitian's demand for a diet recommendation system that can help them make appropriate diet recommendations is quite high. However, the nutrition system of the nutritionist that can be provided by the industry can only electronically use all the information that will be used and inform the interaction between the dietitian and the patient. Although these f-streaming processes can improve the convenience of patients filling out dietary records, they can inform the information needed by dietitians, save the whole, the paper spent on the procedures, etc., but this kind of conventional auxiliary system is still The function of any drink recommendation, 'The nutritionist himself still needs to personally consider all the information I to make a dietary recommendation suitable for this patient. The work load of the nutritionist is still quite heavy. Moreover, because dietitians need to consider too many factors, their dietary recommendations often fail to care about the individual's dietary preferences of 3, 2010,339, and often full of professional narratives, such as recommending protein intake 2g, vitamin 匚5 The narrative of mg, such as the patient's willingness to follow the nutritional formula, is extremely low, and it is impossible to achieve the desired effect of this nutritional recommendation. Therefore, there is a need in the industry for a nutritional recommendation system that provides a list of recommended dishes that not only meet the user's dietary preferences, but also meet the nutritional needs of the user. SUMMARY OF THE INVENTION The main object of the present invention is to provide a dietary suggestion system which can provide a list of recommended dishes that not only meet the dietary preferences of the user but also meet the nutritional needs of the user. Another object of the present invention is to provide a list of recommended dishes that provide a dietary recommendation and provide a dietary preference that not only meets the user's dietary preferences, but also meets the nutritional needs of the user. Still another object of the present invention is to provide a portable electronic device that provides a list of recommended dishes that not only meet the dietary preferences of the latter, but also meet the nutritional needs of the user. The dietary suggestion system of the present invention comprises: a data input module system - a dietary record and - basic data; a data storage module is used to store a plurality of food nutrition components and a feed and a request data; - decision analysis from then on The multi-attribute attribute calculation unit, the decision-making analysis module is an input module that accepts the dietary record and the basic data and is consumed by the (4) (4) nutrients: (4) and the nutritional requirement 4 201033926, the fuzzy multi-attribute operation unit Based on the dietary record, the basic information, the nutrient composition information of the cuisine and the nutritional requirement data, a list of recommended dishes is calculated; and a data output module receives and outputs the list of recommended dishes from the decision analysis module. 5 The method for providing a dietary recommendation of the present invention comprises the steps of: providing a dietary recommendation system comprising a data input module, a data storage module, a decision analysis module and a data output module, and the data The storage module stores a plurality of nutrient information and a nutritional requirement data. The ® analysis module has a fuzzy multi-attribute computing unit; input a food record 10 and a basic data to the data input module; the decision analysis module The group accepts the dietary record and the basic information from the data input module, and the data storage module accepts the nutrient composition information and the nutritional requirement data of the food, and the fuzzy multi-attribute computing unit is based on the dietary record. The information, the nutrient composition information of the dishes and the nutritional requirement data are calculated by a list of suggestions; and the data output module outputs the list of recommended dishes. The portable electronic device of the present invention has a dietary recommendation system. The food recommendation system includes: a data input module that accepts a dietary record and a basic data; and a data storage module that stores a plurality of food nutrition. Ingredients data and a nutritional requirement data; a decision analysis module has a -20 fuzzy eve attribute transfer unit, and the decision analysis module receives the food record and the basic data from the data input module, and the data is from this data. The storage module accepts the nutrient composition information of the cuisine and the nutritional requirement data, and the fuzzy multi-attribute computing unit calculates a list of recommended dishes based on the dietary record, the basic information 'the nutritional content of the cuisine and the nutritional requirement data: With 5 201033926 and a data output module, the decision analysis module accepts and rotates the list of recommended dishes. Since the decision analysis module of the dietary suggestion system of the present invention has a fuzzy multi-attribute operation unit, and the fuzzy multi-attribute operation unit respectively according to the "language variable" in the fuzzy 5 theory, the nutritionist content of the nutritionist for each dish is respectively The judgment and the nutritionist's recommended amount of nutritional requirements for the user (patient) of the dietary recommendation system of the present invention are quantified and mathematically, so-called defuzzification. In this way, the dietary suggestion system of the present invention can effectively avoid the inevitable ambiguity of the nutritionist's judgment on the content of the nutrient of each dish when the product list is exported (for example, the content is extremely high and the content is high, a slightly more ambiguous description of the nutritional requirements of the user (patient) of the dietary recommendation system of the present invention (slightly large, medium in content, slightly less in content, low in content, and very low in content) (eg, very large, very A large number, a large amount, a large amount, a large amount, an increase, an appropriate amount, a reduced amount, a small amount, a small amount, a small amount, a small amount, a very small amount, a very small amount) adverse effects caused by the list of recommended dishes. In addition, the dietary recommendation system of the present invention can accept the user's dietary records (such as preferred cooking information) and basic information (such as the user's dietary record data, height, weight, age, etc.) The physical condition, the history of the disease, and the nutritional needs of the dietitian.) The decision analysis module of the Diet Suggestion System of this section can take into account the information in the list of recommended dishes, so that users can get a copy. A list of recommended dishes that meet their dietary preferences and meet their nutritional needs. On the other hand, since the method for providing a dietary recommendation of the present invention is to provide a user (patient) to a dietary recommendation system using the dietary recommendation system of the present invention, the method for providing a dietary recommendation of the present invention can provide a user. A list of recommended dishes that not only meet their dietary preferences, but also meet their nutritional needs. Similarly, the portable electronic device of the present invention carries the dietary recommendation system of the present invention, and the dietary recommendation system of the present invention operates in the portable electronic device of the present invention, so the portable electronic device of the present invention The device provides a list of recommended recommendations for the user to meet their nutritional needs and meet their nutritional needs anytime, anywhere. The input module of the dietary suggestion system of the present invention can accept any type of information - it preferably includes the dietary record of the user of the dietary recommendation system. οοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοοο Suggested nutritional needs provided by the division. The fuzzy multi-function computing unit of the dietary suggestion system of the present invention can apply any kind of algorithm to calculate a recommended cooking/monthly order, and the preferred method is to apply the ideal solution similarity preference method (Technique f〇r
Order Preference by Similarity to Ideal Solution, TOPSIS)的 15 模糊邏輯演算法運算出此建議料理清單。 本發明之提供一飲食建議的方法所提供之飲食建議系 〇 統的輸入模組可接受任何類別之基本資料，其較佳包括飲 食建議系統之使用者之飲食紀錄資料、身高、體重、年齡、 身體狀況、疾病史以及營養師提供之營養需求建議。本發 20明之提供一飲食建議的方法所提供之飲食建議系統的模糊 多屬性運算單元可應用任何種類之演算法來運算出一建議 料理清單’其較佳係應用理想解類似度偏好法（丁〇1>8]^)的 模糊邏輯演算法運算出此建議料理清單。 7 201033926 本發明之可攜式電子裝置可具有任何類型之資料輸入 單元，其較佳為一指向裝置或一觸控面板❶本發明之可攜 式電子裝置可具有任何類型之資料輸出單元，其較佳為一 影像輸出埠、一平面顯示模組或一列印裝置。本發明之可 5攜式電子裝置所載有之飲食建議系統的模糊多屬性運算單 兀可應用任何種類之演算法來運算出一建議料理清單，其 較佳係應用理想解類似度偏好法（T0PSIS)的模糊邏輯演算 法運算出此建議料理清單。 1〇 【實施方式】 如圖1所示，本發明之飲食建議系統，包括：一資料輸 入模組11、一資料儲存模組12、一決策分析模組13及一資 料輸出模組14。其中，資料輸入模組丨丨可為一資料輸入網 頁，以接受一使用者之飲食紀錄及一基本資料。一般而言， 15飲食紀錄包括此位使用者（患者）當日的飲食紀錄，例如蛋炒 飯、烤雞腿、炒地瓜葉、鮮乳和彰化肉圓。另一方面，前 # 述之基本資料至少包括此位使用者之飲食紀錄資料（例如 前一餐所食用的料理）、身高、體重、年齡、身體狀況、疾 病史以及營養師提供之營養需求建議。在本實施例中，資 20 料輸入模組11係藉由一網際網路15與決策分析模組13相耦 合。 此外’資料儲存模組12係儲存複數個料理營養成份資 料及一營養需求資料。其中’這些料理營養成份資料係根 據行政院衛生署所公布之「台灣地區食品營養成分資料庫 8 201033926 中的「食品成分表」而建置。此資料庫共建置約五百道常 用料理，其内容包含每道料理之各項營養成分的含量。在 此，為了說明的方便，僅以五道料理做為例子（即蛋炒飯、 烤雞腿、炒地瓜葉、鮮乳及彰化肉丸）。另一方面，前述之 5 營養需求資料則參考行政院衛生署所公布之「營養成分攝 取量」（dietary reference intakes, DRIs)建置，其提供 了國人 在不同年齡、不同性別的情況下，每日各種營養成分的建 議攝取量。在本實施例中，資料儲存模組12係與決策分析 參 模組13相耦合》 10 請再參考圖1，本發明之飲食建議系統的決策分析模組 13具有一模糊多属性運算單元（fUzzy multiple attribute decision making unit，FMADM unit) 13 1。此外，決策分析模組13係自資 料輸入模組11接受前述之飲食紀錄及基本資料。此外，決 策分析模組13另自資料儲存模組12接受前述之料理營養成 15 份資料及營養需求資料。接著’模糊多屬性運算單元13便 依據所接受之飲食紀錄、基本資料、料理營養成份資料及 φ 營養需求資料，藉由其模糊多屬性運算單元131運算出一建 議料理清單。其中，在本實施例中，模糊多屬性運算單元 131係應用理想解類似度偏好法（TOPSIS)的模糊邏輯演算 20 法運算出前述之建議料理清單。 TOPSIS模糊邏輯演算法係由Hwang & Yoon所發展而 出，其基本觀念為先界定正理想解（Positive-ideal s〇lution) 與負理想解（negative-ideal solution)。其中’正理想解是各 個決策方案中各屬性間越接近目標期望的集合。反之’負 9 201033926 集各個決策方案中各屬性間越不符合目標期望的 找’透過距離（歐式距離）來衡量各個決策方案，以 正理想解最近而負理想解最遠的決策方案，而此 東案即為最符合前述之需求的決策方案。 5 10 15 至於本發明之飲食建議系統的決策分析模組_具之 、糊多屬ί·生運算單兀⑴應用前述之理想解類似度偏好法 (TOPSIS)而運算出前述之建議料理清 序敘述於下： 册 、首先，在本實施例中，僅選取五道料理與八種營養成 刀舉例說明’五道料理，它們分別為蛋炒飯、烤雞腿炒 地瓜葉、鮮乳和彰化肉圓(al〜a5) ’八種營養成分則分別為 熱量、蛋白f、脂肪、酷類、維生素E、維生素c、約和鐵 (gl 〜g8)。 步驟1 使用者，如營養師，先將前述之五種料理（蛋 炒飯、烤雞腿、炒地瓜葉、鮮乳和彰化肉圓）所分別具有之 八種營養成分（熱量、蛋白質、脂肪、醣類、維生素E、維 生素C、鈣和鐵）含量之模糊子集合，根據語意變數成分區 分七個模糊子集合（含量極多、含量很多含量略多、含量 中等、含量略少、含量很少及含量極少），即營養師對這五 道料理所分別具有之營養成分含量的主觀判斷，如圖2所 不。需注意的是，這些判斷係儲存於本發明之飲食建議系 統的資料儲存模組12中，且資料儲存模組12也已儲存了約 五百道常用料理的營養成份資料於其中。 20 201033926 接著藉由如圖3所示之營養成分含量的語意變數與模 糊數的對照表，圖2所示之表格可轉換為圖4所示之表格， 其中，營養師對於每道料理之每種營養成分含量的判斷均 對應轉換為模糊數。至於前述之七個模糊子集合所屬之模 5掏數便如圖5所示之營養成分含量三角模糊數所示。此外， 雖然常用的模蝴數有三角形、梯形及高斯分佈三種類型， 但因為三角圖形模糊數具有較高的準確性，本發明之飲食 建議系統係採用三角圖形做為模糊化工作。 ® ㈣2:建構出-決策矩陣。在本實關中，營養師提 1〇供之營養需求建議共可分為13種選項，而每一種選項所具 有之權重則如圖6所示。其中，「非常大量」與「非常微量」 具有相同的權重（7)，「很大量」與「很微量」具有相同的 權重（6)’「大量」與「微量」具有相同的權重（5)，「多量」 與「較少量」具有相同的權重（4)，「加量」與「少量」具 15有相同的權重（3)，「增量」與「減量」具有相同的權重（2)， 「適量」則具有單一的權重。此外，如圖7所示，在營養 ® 師所提供的營養需求建議中，所有八種營養成分之需求建 議（建議攝取情況）所對應之權重會再除以所有權重之總 和’以使調整後之各種營養成分之最終權重的總和約為1。 20 如圊8所示，其中al〜a5分別代表五種料理（蛋炒飯、 烤雞腿、炒地瓜葉、鮮乳和彰化肉圓），g丨〜g8則分別代表 八種營養成分（熱量、蛋白質、脂肪、醣類、維生素E、維 生素C、鈣和鐵）’至於屬性權重（營養師所提供的營養需求 建議）則如下所示： 11 201033926Order Preference by Similarity to Ideal Solution, TOPSIS) The 15 fuzzy logic algorithm computes a list of recommended dishes. The dietary recommendations provided by the method of providing a dietary recommendation of the present invention are generally applicable to any type of basic information, preferably including dietary records of users of the dietary recommendation system, height, weight, age, Physical condition, history of the disease, and recommendations for nutritional needs provided by the dietitian. The fuzzy multi-attribute arithmetic unit of the dietary suggestion system provided by the method of providing a dietary suggestion can apply any kind of algorithm to calculate a list of recommended dishes, and the preferred method is to apply the ideal solution similarity preference method. The fuzzy logic algorithm of 〇1>8]^) computes the list of recommended dishes. 7 201033926 The portable electronic device of the present invention can have any type of data input unit, preferably a pointing device or a touch panel. The portable electronic device of the present invention can have any type of data output unit. Preferably, it is an image output port, a flat display module or a printing device. The fuzzy multi-attribute operation unit of the dietary suggestion system contained in the portable electronic device of the present invention can apply any kind of algorithm to calculate a recommended cooking list, and the preferred method is to apply the ideal solution similarity preference method ( The fuzzy logic algorithm of T0PSIS) computes the list of recommended dishes. 1] Embodiments of the present invention include a data input module 11, a data storage module 12, a decision analysis module 13, and a data output module 14. The data input module can be a data input web page to accept a user's eating record and a basic data. In general, the 15 dietary record includes the dietary record of the user (patient) on the day, such as egg fried rice, roasted chicken legs, fried sweet potato leaves, fresh milk and Changhua meatballs. On the other hand, the basic information described in the previous section includes at least the dietary record information of the user (such as the food consumed in the previous meal), height, weight, age, physical condition, disease history, and nutritional needs of the dietitian. . In the present embodiment, the resource input module 11 is coupled to the decision analysis module 13 via an internet 15 . In addition, the data storage module 12 stores a plurality of nutritional ingredients and a nutritional requirement data. Among them, the nutritional content of these dishes was established according to the “Food Ingredients Table” in the Taiwan Food and Nutrition Database 8 201033926 published by the Department of Health of the Executive Yuan. This database has a total of about 500 regular dishes, which contain the nutritional content of each dish. Here, for the convenience of explanation, only five dishes are used as an example (ie, egg fried rice, roasted chicken legs, fried sweet potato leaves, fresh milk, and Changhua meatballs). On the other hand, the above-mentioned information on nutritional needs is based on the “dietary reference intakes” (DRIs) published by the Department of Health of the Executive Yuan. It provides information on the ages and genders of the Chinese people. Suggested intake of various nutrients for the day. In this embodiment, the data storage module 12 is coupled to the decision analysis module 13. 10 Referring again to FIG. 1, the decision analysis module 13 of the diet suggestion system of the present invention has a fuzzy multi-attribute operation unit (fUzzy). Multiple attribute decision making unit, FMADM unit) 13 1. In addition, the decision analysis module 13 receives the aforementioned dietary records and basic data from the data input module 11. In addition, the decision analysis module 13 receives the above-mentioned food nutrition into 15 pieces of information and nutritional demand data from the data storage module 12. Then, the fuzzy multi-attribute operation unit 13 calculates a list of recommended dishes by the fuzzy multi-attribute operation unit 131 based on the accepted dietary record, basic data, cooking nutrient composition data, and φ nutrient demand data. In the present embodiment, the fuzzy multi-attribute operation unit 131 calculates the aforementioned recommended cooking list by applying the fuzzy logic calculus 20 method of the ideal solution similarity preference method (TOPSIS). The TOPSIS fuzzy logic algorithm was developed by Hwang & Yoon. The basic idea is to define the positive-ideal s〇lution and the negative-ideal solution. The 'positive ideal solution' is the set of the attributes in each decision-making scheme that are closer to the target expectation. Conversely, 'negative 9 201033926 sets the distance between each attribute in each decision-making scheme that does not meet the target expectation's 'transmission distance (European distance) to measure each decision-making plan, and the most ideal decision-making solution with the most ideal and the most ideal negative solution. The East case is the decision-making plan that best meets the aforementioned needs. 5 10 15 As for the decision analysis module of the dietary suggestion system of the present invention, the above-mentioned recommended cooking order is calculated by applying the aforementioned ideal solution similarity preference method (TOPSIS). Described below: First, in this example, only five dishes and eight nutrients are selected to illustrate 'five dishes, which are egg fried rice, roasted chicken legs, fried sweet potato leaves, fresh milk and Changhua meatballs. (al~a5) 'The eight nutrients are calories, protein f, fat, cool, vitamin E, vitamin C, and iron (gl ~ g8). Step 1 Users, such as dietitians, firstly have the following five kinds of ingredients (eg, fried rice, roasted chicken legs, fried sweet potato leaves, fresh milk and Changhua meatball) with eight nutrients (calories, protein, fat, sugar). The fuzzy sub-sets of the content of vitamins, vitamin E, vitamin C, calcium and iron are divided into seven fuzzy sub-sets according to the semantic component (the content is very high, the content is much more, the content is medium, the content is slightly less, and the content is small and The content is very small), that is, the nutritionist's subjective judgment on the nutrient content of the five dishes, as shown in Figure 2. It should be noted that these judgments are stored in the data storage module 12 of the dietary recommendation system of the present invention, and the data storage module 12 has also stored nutrient information of about five hundred commonly used dishes. 20 201033926 Then, by the comparison table of the semantic variables and the fuzzy numbers of the nutrient content shown in FIG. 3, the table shown in FIG. 2 can be converted into the table shown in FIG. 4, wherein the dietitian for each dish is cooked. The judgment of the nutrient content is converted into a fuzzy number. As for the above-mentioned seven fuzzy sub-sets, the modulus of the moduli 5 is shown by the triangular fuzzy number of the nutrient content shown in FIG. In addition, although the commonly used modular butterfly numbers have three types of triangle, trapezoidal and Gaussian distribution, since the triangular graphic fuzzy number has high accuracy, the dietary suggestion system of the present invention uses a triangular figure as a blurring work. ® (4) 2: Construct a decision matrix. In this implementation, the nutritionist's recommendations for nutritional needs can be divided into 13 options, and each option has the same weight as shown in Figure 6. Among them, "very large amount" has the same weight as "very small amount" (7), "very large amount" has the same weight as "very small amount" (6) '"large amount" has the same weight as "micro" (5) "Multiple" has the same weight (4) as "Less", "Add" has the same weight as "15" (3), and "Increment" has the same weight as "Decrement" (2) ), "Moderate" has a single weight. In addition, as shown in Figure 7, in the nutritional needs recommendations provided by the Nutrition®, the weights of all eight nutrient requirements (recommended intake) will be divided by the sum of ownership weights to adjust The sum of the final weights of the various nutrients is about 1. 20 As shown in Figure 8, where al~a5 represents five kinds of dishes (eg fried rice, roasted chicken legs, fried sweet potato leaves, fresh milk and Changhua meatball), and g丨~g8 represents eight nutrients (calories, protein). , fat, sugar, vitamin E, vitamin C, calcium and iron) 'As for the weight of the attributes (recommended by the nutritionist's nutritional needs) are as follows: 11 201033926
Wj = [ 0.16, 0·16，0.14，0.14，0.14，0.03，0.14，0，11 ] 在本實施例中，營養師所提供的營養需求建議，表示 2圖7中，也就是wj，但是食物不能有負數（不能從人體中 拿出來，只能少吃，或不吃），所以必須利用轉換圖4來達 5到目的，圖4是依據圖7之建議量轉換成圖8，其中還要用到 圖3的語意模糊數。例如：以r熱量」來說，營養師建議應 攝取”很微量，’，但是蛋炒飯之熱量為（5/6山1)含量極多，就 應該轉換成（0,0，1/6)含量極少，來表示應該攝取極少，以此 ® 類推，若建議量為正則轉換成負，若建議量為負，則轉換 10成正。所以，在圊3中，若需求量是少的，就必須轉換，其 方式為在圖3中等級1和等級7互換，等級2和等級6互換，等 級3和等級5互換，等級4不變。 如圖9所示，加權後正規化決策矩陣Κ = [ίς·]可藉由下 列公式計算出： —► —► 15 匕=Ί. φ 接著，再將矩陣三角模糊數解模糊化。在本實施例中， 係採用重心法（center of area method)將矩陣三角模糊數解 模糊化。即對模掏數％而言，其解模糊化其公 式為v 3 。 步驟3:料w，y分別得到正理想解讀負理想解 ，，如下所示： 12 20 201033926 = {0.133,0.15,0.117,0.133’ 0.133’ 0.023,0.093,0· 103} 笊={0.01，0.027,0.023,0.067,0.007,0.003,0.047,0.04} 步驟4 :如圖1 〇所示’分別計算各決策方案（五道料理） 5 的正理想解S+與負理想解S-的距離，其中： =秦-V)2 , i = '拉(v"’Vy ) , i = l,2,3...m 步驟5 :如圖11所示，計算各決策方案％(五道料理）與 10 正理想解集合的相對接近值c,: C, = S* +S~ , i = \,2,3...m . Ο <c.+ < 1Wj = [ 0.16, 0·16, 0.14, 0.14, 0.14, 0.03, 0.14, 0, 11 ] In this embodiment, the nutritional needs of the nutritionist suggestion, 2, Figure 7, ie wj, but food Can not have negative numbers (can not be taken out of the human body, can only eat less, or not eat), so you must use the conversion map 4 to achieve 5 goals, Figure 4 is based on the proposed amount of Figure 7 converted into Figure 8, which also The semantic number of Figure 3 is used. For example, in terms of r calories, dietitians recommend that they should be “very small,” but the calories of egg fried rice are very high (5/6 Shan 1) and should be converted to (0,0,1/6). The content is very small, which means that it should be ingested very little. In this way, if the recommended quantity is positive, it will be converted into negative. If the recommended quantity is negative, the conversion will be 10%. Therefore, in 圊3, if the demand is small, it must be The conversion is in the form of level 1 and level 7 interchange in Figure 3, level 2 and level 6 are interchanged, level 3 and level 5 are interchanged, level 4 is unchanged. As shown in Figure 9, the weighted normalized decision matrix Κ = [ Σς·] can be calculated by the following formula: —► —► 15 匕=Ί. φ Next, the matrix triangular fuzzy number solution is blurred. In this embodiment, the center of area method is used. The triangular fuzzy number solution of the matrix is fuzzy. That is, for the modulus %%, its solution is vaguely v 3 . Step 3: The material w, y respectively obtains the positive ideal interpretation of the negative ideal solution, as follows: 12 20 201033926 = {0.133,0.15,0.117,0.133' 0.133' 0.023,0.093,0· 103} 笊={0.01,0.027,0 .023, 0.067, 0.007, 0.003, 0.047, 0.04} Step 4: Calculate the distance between the positive ideal solution S+ and the negative ideal solution S- for each decision scheme (five dishes), as shown in Figure 1 ,, where: = Qin-V) 2 , i = ' pull (v" 'Vy ), i = l, 2, 3...m Step 5: As shown in Figure 11, calculate each decision plan % (five dishes) and 10 The relative approximate value c of the set of positive ideal solutions, C, = S* + S~ , i = \, 2, 3...m . Ο <c.+ < 1
I 步㈣··進行決策方案排序。#是決策方案^為最佳選 15擇’則其^應該越接近卜反之’若決策方案巧為最差選 擇，則其C,·越接近0。此外，若同時有兩個或兩個以上的 決氣方案具有相同的ς+時’它們的順位將隨機排列。 13 201033926 因此’如圖12所*，依照計算出值大小將各 案依序排列，α4>α3>α2>α >α 音 、 5 忍即，在各決策方案（五请 料理鮮乳為最好的選擇，其餘各道料理依照順位為炒 地瓜葉、烤雞腿、彰化肉圓及蛋炒飯。圖12所示之列表即 為本發明之飲食建4系統的決策分析模組13所運算出之建 議料理清單，且每道材料並依照—建議順位依序由上 排列。 • 如此，前述之資料輸出模組Μ所輸出之建議料理清單 便包括前述之五個建議料理（蛋炒飯、烤雞腿炒地瓜葉、 10鮮乳和彰化肉圓），且這五個料理依照一建議順位排序（第一 順位為鮮乳，第二順位為炒地瓜葉，第三順位為烤雞腿， 第四順位為彰化肉圓，第五順位為蛋炒飯）。其中，資料輸 出模組14為一資料輸出網頁，且資料輸出模組14亦藉由網 際網路15與決策分析模組13相耦合，如圖1所示。 15 另一方面，如圖1所示，在本發明另一實施例中，本發 φ 明之飲食建議系統可選擇性地更包括一與決策分析模組13 相耗合之營養師管理介面16 ’且營養師管理介面16亦透過 網際網路15與決策分析模組13相耦合，以供一營養師或相 關專業人員至少執行下列之動作其中之一： 20 1.調整前述之儲存於資料儲存模組12之複數個料理 營養成分含量的語意變數（即改變各項營養成分的 權重），使得營養師可隨時改變其對於某道料理所 具之各種營養成分含量的判斷。 201033926 2.增加營#師對於一转料理所具之各種營養成分 含1的判斷，且將這些新增的資料儲存於資料儲存 模組12。 3·修改使用者先前所輸入之基本資料，如身高、體 5 重、年齡、身體狀況、疾病史以及營養師提供之營 養需求建議，使得營養師可再次確認使用者所輸入 之基本資料是否無誤。 4.在資料輸出模組14輸出前述之建議料理清單前，使 © 得營養師可確認此即將輸出之建議料理清單是否 1〇 合理並可加以修改。 因此，本發明之飲食建議系統可有效避免營養師對於 每道料理之營養成分含量的判斷所不可避免的模糊性敘述 以及對於本發明之飲食建議系統之使用者的營養需求建議 量所不可避免的模糊性敘述對此份建議料理清單的不良影 15響。另一方面，由於本發明之飲食建議系統可藉由其輸/ 模組接受使用者之飲食紀錄及基本資料，本發明之飲^建 • 議系統之決策分析模組在運算出建議料理清單時可將這此 資料通通納入考慮，使得使用者可得到一份不但符合=飲 食偏好’更可達到其營養需求的建議料理清單。 ' 20 如圖13所示，本發明之提供一飲食建議的方法，包括 下列步驟： 步驟A :提供一飲食建議系統，係包括一資料輪入模 組、一資料儲存模組、一決策分析模組及一資料輸出模組、 15 201033926 且此資料儲存模組儲存複數個料理營養成份資料及一營養 需求資料，此決策分析模組具有一模糊多屬性運算單元： 步驟B:輸入一飲食紀錄及一基本資料至此資料輸入模 組； 5 步驟C:此決策分析模組自此資料輸入模組接受此飲食 紀錄及此基本資料，且自此資料儲存模組接受此等料理營 養成份資料及此營養需求資料，此模糊多屬性運算單元再 依據此飲食紀錄、此基本資料、此等料理營養成份資料及 書 此營養需求資料運算出一建議料理清單；以及 10 步驟D:此資料輸出模組輸出此建議料理清單。 其中，由於前述之本發明之提供一飲食建議的方法所 提供之飲食建議系統即為本發明之飲食建議系統，其組成 與運作方式便如m至圖12所示。因此，此飲食建議系統之 各組成模組（即資料輸入模組、資料儲存模組決策分析模 Μ組及資料輸出模組）的形式與輕合方式、此飲食建議系統接 受輸入之各項資料的種類、此飲食建議系統根據這些輸入 ^ "^龍糊㈣存於資㈣存餘崎鍾運算的詳細流 程、以及最後輸出前述之建議料理清單的格式等等，在此 便不再贅述。 因此Φ於本發明之提供一飲食建議的方法係應用本 發明之飲食建遢系統’所以本發明之提供一飲食建議的方 法可提供使用者一不但符合其飲食偏好，更可達到其營養 需求的建議料理建議清單。 201033926 ❹ 10 15 ❹ 20 如圖14所示，本發明 入單元21、— 了攜式電子裝置包括-資料輪 單元以。其中，運算單元23以及一資料輪出 明之飲食建言I李Γ 攜式電子裝置載有前述之本發 1。此外n、而此飲食建議系統的組成則請參閱圖 令未亍耗：可攜式電子裝置亦可與-顯示裝置(圖 二未：说。，以顯示飲食建議系統所輸出之建議料理清 彳面，本發明之可攜式電子裝置的資料輸入單元 可為一指向裝置或-觸控面板。但在本實施例t，本發 月之可攜式電子裝置的資料輸入單元21係一觸控面板。 前述之飲食建議系統之資料輸入模組係運作於本發明 之可攜式電子裝置的資料輸入單元21。另一方面此飲食 建議系統之資料儲存模組係儲存於本發明之可攜式電子裝 置之記憶單元22,此飲食建議系統之決策分析模組係執行 於本發明之可攜式電子裝置之運算單元23，此飲食建議系 統之資料輸出模組則運作於本發明之可攜式電子裝置之資 料輸出單元24’且資料輸出單元24係一影像輸出埠。 其中’由於前述之本發明之可攜式電子裝置所載有之 飲食建議系統即為本發明之飲食建議系統，其組成與運作 方式便如圖1至圖12所示。因此，此飲食建議系統之各組成 模組（即資料輸入模組、資料儲存模組、決策分析模組及資 料輸出模組）的形式與叙合方式、此飲食建議系統接受輸入 之各項資料的種類、此飲食建議系統根據這些輸入的資料 模糊及儲存於資料儲存模組進行模糊運算的詳細流程、以 17 201033926 及最後輸出前述之建議料理清單的格式等等，在此便不再 贅述。 因此，由於本發明之可攜式電子裝置係載有本發明之 飲食建議系統’所以本發明之可搆式電子裝置可提供使用 5 10 15 者一不但符合其飲食偏好，更可達到其營養需求的建議料 理建議清單。 上述實施例僅係為了方便說明而舉例而已，本發明所 主張之權利範圍自應以申請專利範圍所述為準，而非僅限 於上述實施例。 【圖式簡單說明】 圖1係本發明之飲食建議系統的示意圖。 圓2至圖Π係顯示本發明之飲今读媒么^ ， 飲食建4系統之決策分析模組 運算出一建議料理清單之過程的示意圖。 圖13係本發明之提供一飲食建議的方法的流程圖。 圖14係本發明可攜式電子裝置的示意圖。 【主要元件符號說明】 Η資料輸入模組 13決策分析模組 14資料輸出模組 16營養師管理介面 22記憶單元 24資料輸出單元 12資料儲存模組 13 1模糊多屬性運算單元 1 5網際網路 21資料輸入單元 23運算單元 18Step I (4) · Sort the decision plan. #是决定方案^ is the best choice, then its ^ should be closer to the opposite. If the decision-making scheme is the worst choice, its C, · is closer to zero. In addition, if two or more gas-reduction schemes have the same ς+ at the same time, their order will be randomly arranged. 13 201033926 Therefore, as shown in Fig. 12, the cases are arranged in order according to the calculated value, α4>α3>α2>α >α, 5, that is, in each decision-making scheme (five please serve fresh milk for the most Good choice, the rest of the dishes according to the order for fried sweet potato leaves, roast chicken legs, Changhua meatballs and egg fried rice. The list shown in Figure 12 is calculated by the decision analysis module 13 of the Diet Building 4 system of the present invention. A list of recommended dishes is provided, and each material is arranged in order according to the recommended order. • Thus, the list of recommended dishes output by the above-mentioned data output module 包括 includes the above five recommended dishes (eg fried rice, roasted chicken drumsticks) Sweet potato leaves, 10 fresh milk and Changhua meatballs), and these five dishes are sorted according to a recommended order (the first order is fresh milk, the second order is fried sweet potato leaves, the third order is roasted chicken legs, and the fourth order is Changhua The meat round, the fifth order is the egg fried rice. The data output module 14 is a data output webpage, and the data output module 14 is also coupled to the decision analysis module 13 via the Internet 15, as shown in FIG. Show. 15 The other side As shown in FIG. 1, in another embodiment of the present invention, the diet suggestion system of the present invention may optionally include a dietitian management interface 16' that is incompatible with the decision analysis module 13 and the dietitian management The interface 16 is also coupled to the decision analysis module 13 via the Internet 15 for a nutritionist or related professional to perform at least one of the following actions: 20 1. Adjust the foregoing plurality of stored in the data storage module 12 The semantic variables of the nutrient content of the food (that is, changing the weight of each nutrient), so that the nutritionist can change his judgment on the content of various nutrients in a certain dish at any time. 201033926 2. Increase the camp #师 for a turn The various nutrients contained in the food contain 1 judgment, and the added data is stored in the data storage module 12. 3. Modify the basic information previously input by the user, such as height, body weight, age, physical condition The history of the disease and the nutritional needs recommendations provided by the nutritionist enable the dietitian to reconfirm whether the basic information entered by the user is correct. 4. In the data output module 14 Before the list of recommended dishes mentioned above, let the dietitian confirm that the list of recommended dishes to be exported is reasonable and can be modified. Therefore, the dietary recommendation system of the present invention can effectively prevent nutritionists from nourishing each dish. The ambiguous narrative inevitable in the determination of the content of the ingredients and the inevitable ambiguity of the recommended amount of nutritional requirements for the users of the dietary recommendation system of the present invention have a negative impact on the list of recommended dishes. On the other hand, Since the dietary suggestion system of the present invention can accept the user's dietary record and basic data through the input/module, the decision analysis module of the drinking and drinking system of the present invention can take this when calculating the recommended cooking list. The information is taken into consideration so that users can get a list of recommended dishes that not only meet the = dietary preference but also meet their nutritional needs. As shown in FIG. 13, the method for providing a dietary recommendation of the present invention comprises the following steps: Step A: providing a dietary recommendation system, comprising a data entry module, a data storage module, and a decision analysis module. Group and a data output module, 15 201033926 and the data storage module stores a plurality of nutrient composition data and a nutritional requirement data. The decision analysis module has a fuzzy multi-attribute operation unit: Step B: input a dietary record and A basic data is sent to the data input module; 5 Step C: The decision analysis module receives the dietary record and the basic information from the data input module, and the data storage module accepts the nutrition information of the food and the nutrition Demand data, the fuzzy multi-attribute computing unit calculates a list of recommended dishes based on the dietary record, the basic information, the nutrient composition information of the cuisine, and the nutritional demand data; and 10 step D: the data output module outputs the Suggested cooking list. Among them, the dietary suggestion system provided by the above-mentioned method for providing a dietary recommendation of the present invention is the dietary suggestion system of the present invention, and its composition and operation mode is as shown in Fig. 12. Therefore, the form and the light combination method of each component module of the food recommendation system (ie, the data input module, the data storage module decision analysis module and the data output module), and the information recommended by the dietary recommendation system are accepted. The type of this dietary suggestion system is based on these inputs ^ "^龙糊(四) in the capital (4) detailed process of storing the Yusaki clock, and finally outputting the format of the aforementioned list of recommended dishes, etc., and will not be described here. Therefore, the method for providing a dietary recommendation in the present invention is to apply the dietary building system of the present invention. Therefore, the method for providing a dietary recommendation of the present invention can provide a user with a dietary requirement that meets their nutritional needs. Suggested food recommendation list. 201033926 ❹ 10 15 ❹ 20 As shown in FIG. 14, the present invention enters unit 21, and the portable electronic device includes a data wheel unit. The arithmetic unit 23 and the data-providing food suggestion I Li-Yu portable electronic device carry the aforementioned hairpin 1. In addition, the composition of the dietary recommendation system can be found in the diagram: the portable electronic device can also be used with the - display device (Figure 2: not mentioned.) to display the recommendations of the food recommendation system. The data input unit of the portable electronic device of the present invention can be a pointing device or a touch panel. However, in the present embodiment t, the data input unit 21 of the portable electronic device of the present month is a touch. The data input module of the above-mentioned dietary suggestion system is operated by the data input unit 21 of the portable electronic device of the present invention. On the other hand, the data storage module of the dietary suggestion system is stored in the portable type of the present invention. The memory unit 22 of the electronic device, the decision analysis module of the diet suggestion system is implemented in the computing unit 23 of the portable electronic device of the present invention, and the data output module of the food recommendation system operates in the portable mode of the present invention. The data output unit 24' of the electronic device and the data output unit 24 are an image output port, wherein 'the dietary suggestion system carried by the portable electronic device of the present invention described above The composition and operation mode of the food recommendation system of the present invention is as shown in FIG. 1 to FIG. 12. Therefore, each component module of the dietary recommendation system (ie, data input module, data storage module, decision analysis module) And the format and method of the data output module, the types of data that the dietary recommendation system accepts, and the detailed flow of the food recommendation system based on the data entered and the data storage module for fuzzy calculation, The format of the list of recommended dishes mentioned above and the like, and the like, will not be described here. Therefore, since the portable electronic device of the present invention carries the dietary suggestion system of the present invention, the structure of the present invention is The electronic device can provide a list of recommended dishes that use 5 10 15 not only in accordance with its dietary preferences, but also meet its nutritional needs. The above embodiments are merely examples for convenience of explanation, and the scope of claims claimed by the present invention is self-sufficient. The above description is based on the scope of the patent application, and is not limited to the above embodiments. Schematic diagram of the diet suggestion system. The circle 2 to the figure shows the schematic diagram of the process of calculating the recommended food list by the decision analysis module of the present invention. Fig. 13 is a diagram of the present invention. Figure 14 is a schematic diagram of a method for a portable electronic device. Figure 14 is a schematic diagram of a portable electronic device of the present invention. [Key element symbol description] Η Data input module 13 Decision analysis module 14 Data output module 16 Nutritionist management interface 22 Memory Unit 24 data output unit 12 data storage module 13 1 fuzzy multi-attribute operation unit 1 5 internet 21 data input unit 23 arithmetic unit 18