Computer-Aided Multi-Criteria Food Selection and Optimization System Based on Nutrition, Allergen, and Environmental Data.

Abstract

This invention relates to a computer-aided decision support system and method that provides personalized meal selection and optimization for users. The system processes user nutritional goals, allergen restrictions, health data, user preferences, and contextual data together with nutritional values, allergen content, and environmental impact data of meals to generate optimized meal recommendations. Within the scope of this invention, heterogeneous datasets obtained from different data sources are normalized, meals unsuitable for the user are subjected to a restriction-based pre-filtering process, and multi-criteria suitability scores are calculated for the remaining candidate meals. Based on these calculated scores, meals are ranked and optimized meal recommendations are presented to the user. By performing the filtering process before the score calculation, the system reduces computational load and improves real-time decision-making performance.It also incorporates an adaptive learning structure that can dynamically update parameter weights based on user behavior, feedback, and recurring selection patterns. Thanks to this invention, user safety is enhanced, personalized and sustainable food selection is provided, and an integrated optimization infrastructure is achieved that operates in real-time on menus (in restaurants, cafes, catering companies, mass food production facilities, and digital menu platforms).

Description

1 TARIFF Computer-Aided Multi-Scale Analysis Based on Nutrition, Allergen, and Environmental Data Food Selection and Optimization System Based on Criteria Technical Area This invention; computer-aided decision support systems, artificial intelligence-powered data processing 5 systems, digital nutrition management and restaurant technology related to the technical field and specifically the user's nutritional goals, allergen restrictions, health data, user preferences and environmental sustainability parameters together by evaluating restaurants, cafes, catering companies, mass food production facilities, and digital menus. There are 10 platforms that create user-specific, optimized meal recommendations. It relates to a criteria-based food selection and optimization system and method. The invention allows for the visualization of food data, rather than simply displaying it; it provides nutritional information about meals. values, allergen content, environmental impact data, user preferences, and contextual factors. a computer-based 15 that enables data to be processed within the same decision engine It relates to data processing and optimization infrastructure. In this context, the system, user In this respect, dishes that are unsuitable are subjected to a restriction-based pre-filtering process. Multiparameter suitability scores for candidate dishes remaining after filtering. calculating and optimizing meals based on those scores. It forms the recommendations. Thus, the static data presentation in existing systems is 20 instead; real-time operating, learning, customizable and transactional efficiency An enhanced dynamic decision support infrastructure is provided. State of the art Today, digital restaurant menus, nutrition apps, diet management apps 25 systems and user-specific food recommendation platforms are becoming increasingly common, various software designed to make the food selection process easier for users Solutions are being developed, particularly mobile applications and online dining. In systems that operate through platforms; calorie tracking, nutritional value calculation, Creating dietary recommendations, providing allergen information, and offering sustainable nutrition advice 30 Different technical approaches are used, such as those mentioned. However, existing systems... 2 The vast majority of them consider these parameters independently of each other. It collects and provides the user with only limited data. In most applications... only calorie, macronutrient, or basic ingredient information is provided to the user. parameters such as health constraints, allergen safety and environmental impact are shown. 5 by evaluating them together to create an optimized decision-making mechanism It cannot be converted. MyFitnessPal is one of the applications included in the known state of the art. The system tracks users' daily calorie intake and macro and micronutrient values. It is a nutritional application that enables them to do so. The system in question, It can perform specific analyses and provide partial recommendations based on users' dietary goals. 10 It can offer. However, the system is primarily focused on nutrition only. It focuses on data including allergen restrictions, environmental impact parameters and much more. It does not offer a structure that evaluates criteria-based decision-making mechanisms together. We can also filter the dishes and assign multi-parameter suitability scores to them. It also does not include an optimization engine that sorts in an optimized way. 15 Another application within the known state of the art is the Noom system. Analyzing user behavior to promote weight loss and healthy eating habits. It offers suggestions for improvement. In this system, user data is taken into account. Personalized recommendations can be generated based on this information, as well as the restaurant menu. Real-time food selection optimization is not performed on a per-page basis. Also, 20 The system includes allergen filtering, environmental impact analysis, and multi-criteria score calculation. There is no comprehensive, integrated decision support mechanism in place. The Lifesum app provides diet plans and daily meal plans based on the user's profile. It is known as a nutritional monitoring system that offers recommendations. However, the aforementioned The application operates based on predefined plans and provides real-time restaurant information. 25 It does not perform dynamic optimization based on menu data. Furthermore... a systematic scoring model of allergen data and environmental impact parameters It is not included. The AllergyEats system helps users choose restaurants in terms of allergen safety. It is a platform that allows for evaluation. The 30 in question... The system takes allergen information into account and implements specific safety measures for the user. 3 It offers filtering mechanisms. However, the system only filters allergens. It exhibits a structure focused on; nutritional goals, user preferences and a multi-criteria optimization that evaluates environmental impact parameters together It does not include infrastructure. Additionally, there is a combined suitability score for the dishes. There is no technical decision-making mechanism that creates and ranks these decisions. 5 Spokin is an application that allows individuals with allergies to restaurants and meals. It offers community-based suggestions on the subject. However, the system user It generates recommendations based on its experiences and is a data-driven, multi-parameter approach. It does not include an optimization algorithm. Specifically, filtering and scoring. architecture, environmental impact assessment and mathematical decision mechanism 10 It is not available. In the known state of the art, there are also various patent documents. One of them is "Diet mapping processes and systems" with the number US11328810B2. The patent, titled [Patent Title], aims to influence users' dietary choices in terms of health and environmental factors. It describes a system that analyzes according to sustainability criteria. These 15 The patent includes both nutritional data and environmental impact parameters. This is being evaluated. However, the system uses allergens as a mandatory filtering criterion. not using and real-time filtering + scoring based on restaurant menu + It does not offer a sorting architecture. In the patent applications titled US20190295440A1 / WO2019183404A1, there are 20 Computer-assisted systems that provide personalized nutrition and health management recommendations to users. The systems are described. These systems analyze user data and food data. It produces personalized recommendations by doing so. However, allergens in these structures... multiparameter suitability, where it is used as a mandatory filtering mechanism. Real-time optimization is performed based on restaurant menu scores or 25 an integrated structure in which environmental impact is made a central decision parameter It is not available. “Method and system for evaluating, scoring and presenting” numbered US7974881B2 The patent titled "nutritional value information" refers to the nutritional values ​​of food products. It describes a system for scoring and presenting results to the user. 30 However, the patent in question is a one-dimensional one based solely on nutritional values. 4 It offers an assessment approach that includes allergen safety, environmental impact, and user experience. a multi-criteria decision that considers preferences and contextual data together It does not include a mechanism. Furthermore, unsuitable options are not presented beforehand. There is also no system in place for filtering it. “Systems and methods for diet quality assessment” number US12073935B2 5 The patent, titled [Patent Title], analyzes users' past consumption data to determine their diet. It describes systems for evaluating quality. However, this system is more... It focuses on the analysis of very old dietary habits and restaurant menus. An active decision support system that optimizes instant meal choices. It is not of a certain nature. Furthermore, allergen-based mandatory filtering and multi-criteria compliance are not applicable. There is no score or optimized ranking approach. With application number US20180154350A1 for “Portable allergen detection system” Titled "Detection of allergens using Raman scattering" (US10656145B2) The patent, on the other hand, is for hardware-based systems for the physical detection of allergens. These systems describe sensors, optical analysis, or Raman 15. Allergen detection in food samples is performed using techniques such as scattering. However, these systems provide allergen data for user-specific, multi-parameter decision-making. It does not offer an optimized food selection by integrating it into its mechanism. It also includes filtering, multi-criteria scoring, and dynamic ranking infrastructures. It does not include. 20 Upon reviewing the existing systems and patent documents described above; • Nutritional data, allergen information and environmental impact parameters are mostly used. considered independently of each other, • Options that are not suitable for the user are not systematically filtered, • Optimized ranking with multi-parameter fitness scores after filtering 25 not done, • Lack of real-time decision support infrastructure based on restaurant menus, • A dynamic system that updates parameter weights based on user feedback. no structure was presented, • It appears that allergen data is not at the center of the decision-making process. 30 Developed to overcome the aforementioned disadvantages in the current technology. This invention allows users to take into account their nutritional goals, alleviate restrictions on allergies, and manage environmental factors. Sustainability criteria, user preferences, and contextual data all contribute to the same decision. considered together within the engine; risky for the user. The dishes are pre-filtered; the remaining dishes are rated with multi-criteria suitability scores of 5. The menus are sorted and optimized meal recommendations are presented to the user. Thus, unlike existing systems, instead of passive structures that only provide data; filtering, score calculation, dynamic optimization and personalization capabilities An active decision support and optimization system is obtained. The purpose and technical advantages of the invention. The purpose of the invention is to enable users to create food for restaurants, cafes, catering companies, and mass food production facilities. and the food selection processes that they carry out through digital food platforms; nutritional goals, allergen restrictions, health data, user preferences, and environmental factors. a computer-aided decision that optimizes in line with sustainability criteria 15 The goal is to develop a support and optimization system, especially for existing systems. Unlike passive structures that rely solely on information presentation, food data is seen. Active filtering that processes and produces user-specific optimized decision outputs, The aim is to provide an evaluation and ranking mechanism. Another aim of the invention is to place the user's allergen sensitivities at the center of the system. evaluating food items that may pose a risk to the user by placing them in a specific location. The goal is to ensure its removal from the system at an early stage of the process. In this context... The system does not display allergen data solely for informational purposes; on the contrary... It uses these as mandatory technical filtering criteria. Thus, user 25 Safety is being improved, the risks of poor food choices are being reduced, and serious allergic reactions are being minimized. The aim is to prevent situations that could lead to reactions. The invention also enables different data types to be used together within the same decision-making mechanism. Multiparameter data processing and optimization that enables evaluation 30 It provides the infrastructure. Accordingly, the system provides nutritional values ​​and allergens of foods. 6 normalize content, environmental impact data, user preferences, and contextual data. by making them comparable and weighting the parameters in question It evaluates them together within the mathematical compatibility model. Thus Instead of focusing solely on one-dimensional assessments such as calories or nutritional value; health, 5 that take safety, sustainability and user habits into account together An integrated decision-making mechanism is achieved. One of the key technical advantages of the invention is the scoring of the constraint-based pre-filtering process. This is done before the calculation process. Thanks to this technical structure... Dishes unsuitable for the user are eliminated early in the system, 10 Scoring is only performed on suitable candidate dishes. Therefore... Unnecessary calculations are prevented, processing load is reduced, and processor resources are conserved. It is being used more efficiently, especially in high-volume restaurant menus. Timely decision-making performance is improved. Another technical advantage is the system's multi-criteria suitability scores. It is the ability to calculate. Within the system, for each meal; nutritional suitability, allergen safety, environmental impact, user preference fit, and contextual appropriateness. a combined fitness score by evaluating the parameters together is being created. Thanks to this technical approach, only the appropriate 20 are provided to the user. not only the food, but also the highest level of adaptation to their own needs. Optimized meal options can be offered. Another technical advantage of the invention is its ability to determine environmental sustainability parameters. It is about integrating it into the mechanism. The system tracks the carbon footprint of food and water. consumption, energy consumption and similar environmental impact data are included in the assessment. by encouraging users to choose foods with a lower environmental impact This encourages not only individual health goals, but also sustainable development. Consumption habits are also supported. 7 The invention also dynamically adjusts system parameters by analyzing user behavior. It includes an adaptive structure that can be updated in this way. User history based on choices, interactions, order patterns, and feedback records Parameter weights are being re-optimized and the system becomes more efficient over time. It can produce highly accurate personalized results. This technical structure 5 Thanks to this, the system can learn and adapt, moving away from static recommendation infrastructures. It is transforming into a dynamic decision support system. Another technical advantage of the invention is the real-time display of restaurant menus. It is able to function. When the user accesses the system, the current menu data is displayed instantly. Analysis, filtering, and scoring processes are being performed simultaneously. This is done so that users don't have to navigate through long and complex menus. The need for manual evaluation is reduced and the decision-making process is made technical. is being accelerated. Furthermore, thanks to the system's modular software architecture, existing POS systems can be converted into digital systems. menu infrastructures are integrated with mobile applications and online ordering platforms. This ensures that the system can work with restaurants of different sizes. This allows for its implementation in food service infrastructures. In conclusion, the invention combines user data and food data in the same decision-making mechanism. It operates within a system that performs allergen-based pre-filtration and meets multi-criteria requirements. calculates scores, evaluates environmental sustainability criteria, user can dynamically update parameter weights depending on their behavior and an integrated computer that provides real-time optimized meal recommendations 25 by providing a supported decision support and optimization system for existing technical solutions It addresses the shortcomings that exist. Detailed description of the invention The invention provides 30 computer-aided meal selections and optimizations tailored to the user. It relates to a decision support system and methodology; obtained from different data sources. 8 heterogeneous food data collected, user's health, nutrition, allergen and by working together with environmental sustainability parameters in real time Modular data processing and optimization that generates optimized meal recommendations. It includes architecture. The system is particularly suitable for low-volume menu data. 5 with a multi-layered technical infrastructure that enables analysis with processing load. It is structured. Within the scope of the invention, the system generally; •user interface, • data management layer, • Processing and optimization engine, 10 •decision support layer, •It includes adaptive learning and parametric update layers. User Interface and Data Collection Layer User interface, mobile application, web-based platform, digital restaurant menu, 15 It can operate via kiosk systems or online ordering infrastructures. It is structured. The user interface, the transfer of user data to the system, visualization of system outputs and collection of behavioral interaction data. It works for the purpose of... Within the user interface; 20 • User profile data module, • User restrictions management module, • Contextual data module, • Behavioral preference analysis module is included. The user profile data module includes information about the user's age, weight, metabolic goals, and daily intake. energy requirements, macronutrient targets, diet type, and health parameters are entered into the system. It is conveying. User restrictions management module, allergen sensitivities, prohibited content, religious system data on dietary preferences, intolerance information, and health-based restrictions. It transforms into its structure. 30 The contextual data module provides information about the user's current status; 9 • Time of day, • Meal time, • Daily remaining energy requirement, • Physical activity level, • Location data, 5 • Restaurant density, • It processes contextual variables such as stock availability. The behavioral preference analysis module analyzes the user's past choices and orders. by analyzing behaviors, click history, and recurring selection patterns It creates a parametric preference profile. 10 Thanks to this structure, user data is standardized into a data format. heterogeneous data from different data sources are being transformed and processed in the same way. This ensures that they can be processed together in the optimization engine. Data Management and Data Processing Layer 15 Data management layer, restaurant systems, digital menu infrastructures, third-party APIs. data obtained from systems or AI-powered data production infrastructures It is structured to operate. Within the data management layer; • Food data repository, 20 • Allergen data repository, • Environmental impact data repository, • Data validation module, • It includes an AI-powered data prediction module. The food data repository contains information on calories, protein, carbohydrates, fat, fiber, and other nutrients for meals. It stores the list, portion information, and micronutrient values. The allergen data repository allows for the machine to process allergen content in foods. It stores data in a specific format. The environmental impact data repository includes carbon footprint and water data related to food production processes. It includes consumption, energy consumption, and sustainability coefficients. 30 The data validation module performs a consistency analysis of the food data uploaded to the system. This ensures data integrity. The AI-powered data prediction module predicts missing data records. It uses machine learning algorithms for this purpose. In this context, the system, Based on similar food data, missing nutrients, allergens, or environmental factors 5 It improves data integrity by estimating its parameters. This technical structure allows for the analysis of heterogeneous datasets from different sources. normalization is being performed and high-volume datasets are being integrated within the same data model. processing is ensured. Processing and Optimization Engine The processing and optimization engine, which forms the basic technical structure of the invention, is for the user. Optimized by analyzing data and food data in real time. It produces decision outcomes. Processing and optimization engine; 15 • Data normalization module, • Restriction-based pre-filtering module, • Multi-parameter score calculation module, • Parametric weight optimization module, • Includes a ranking and decision generation module. 20 Data Normalization Module The data normalization module combines different data types into a common computational structure. It transforms. In this context; 25 • Nutritional values, • Allergen risk factors, • Environmental impact parameters, • User preference data, • Contextual variables are standardized and normalized into a standardized data format. 30 It is being transformed. 11 This technical structure allows heterogeneous datasets to be processed using the same computational infrastructure. They can be processed together within it, and multi-parameter optimization operations It can be accomplished. Constraint-Based Pre-Filtering Module 5 The restriction-based pre-filtering module scores dishes that are unsuitable for the user. It removes it from the system before the calculation process. Within this module; • Allergen filtration, • Health restriction control, 10 • Dietary compliance check, • Minimum / maximum nutritional value control, • Prohibited content is checked. In particular, the allergen filtering process is the system's fundamental technical safety mechanism. It consists of foods containing allergens that the user has identified as sensitive. The candidate is automatically excluded from the dataset. By performing the filtering process before the score calculation process: • Unnecessary score calculations are prevented, • Processor load is reduced, • Computational complexity is being reduced, 20 • Data processing delay is reduced, • Real-time decision-making performance is improved. This technical structure improves processing efficiency, especially in high-volume restaurant menus. It increases significantly. Multiparameter Score Calculation Module After pre-filtering, the remaining candidate dishes are subjected to multi-parameter scoring. It is evaluated by the module. Within this module, a combined fitness score is assigned to each dish. is being calculated. 30 12 This technical structure allows different data parameters to be integrated into the same decision model. combined decision outcomes are evaluated and optimized. is being created. In one application, the suitability score for each dish is 5 as follows: is being calculated: UMSSi=w1Ni+w2Ai+w3Ei+w4Pi+w5Ci Here; • Nᵢ: nutritional suitability, 10 • Aᵢ: allergen safety, • Eᵢ: environmental impact parameter, • Pᵢ: user preference matching, • Cᵢ: contextual appropriateness, • w₁–w₅: represents the weighting coefficients for the relevant parameters. 15 The allergen safety parameter (Aᵢ) takes into account the user's allergen sensitivities. It is calculated by taking into account the following, and if it falls below a certain threshold or the user Foods identified as risky in this regard are automatically removed from the candidate dataset. They may be removed or evaluated with a low suitability score. The suitability scores calculated by the scoring engine are used for ranking and decision generation. 20 The candidate meals are transferred to the module; the calculated combined suitability scores are then transferred to the module. The system sorts meals accordingly and provides users with optimized meal recommendations. This technical structure allows different data parameters to be integrated into the same decision model. They can be evaluated together, process efficiency is increased, and the user Personalized and optimized decision outcomes in real time 25 can be created. 13 Parametric Weight Optimization Module The parametric weighting optimization module is used in the score calculation process. It dynamically updates the parameter coefficients. This module; • User selection history, 5 • Behavioral interaction data, • Feedback records, • Parametric optimization through recurring ordering patterns It is carrying out. Thanks to this technical structure, the system achieves higher accuracy and more over time. It can produce optimized decision outputs with a low error rate. Sequencing and Decision Making Module The ranking and decision-making module uses calculated suitability scores to identify candidates. It transforms meals into an optimized decision sequence. 15 This module: • In identifying dishes with a high suitability score, • Creating alternative choice sets, • We perform category-based optimization, • Technical optimization such as low carbon footprint or low allergen risk 20 It produces screenplays. This structure allows the user to not only be presented with data, but also to receive optimized decisions. The outputs are generated in real time. Adaptive Learning and Update Layer 25 The adaptive learning layer adapts the system based on user behavior. It updates its parameters automatically. Within this module; • User feedback, • Order cancellations, 30 • Repeated elections, 14 • Interaction patterns are analyzed and parametric weighting coefficients are recalculated. It is being optimized. Thanks to this technical structure, the system, unlike static proposal infrastructures, saves time. within it a dynamic system that is adaptable and whose accuracy rate is constantly being improved. It is transforming into an optimization system. 5 In conclusion, the invention normalizes heterogeneous datasets using mandatory constraints. Multi-criteria optimization with reduced computational load, performing base-based pre-filtering. using an engine that produces real-time decision outputs and adaptive parametric an integrated computer-assisted food selection and update capability It provides an optimization system. 10 EXAMPLE USAGE SCENARIOS Use Case – Optimization Focused on Sports Nutrition In one example application, the system could target professional athletes or 15-year-olds who exercise regularly. to create optimized meal recommendations for users It is used. Users log in to the system via the mobile application and manage their daily information. protein target, calorie requirement, carbohydrate balance, and training intensity. It is transferring its information to the system. The contextual data module indicates that the user is in the post-workout time frame. and indicates that daily protein requirements have not yet been met. System Analyzing dishes on restaurant menus to find low-protein or high-sugar options. It filters the options that contain these. Leftover food; 25 Protein density, Calorie suitability, Muscle recovery suitability, User preference history, Carbon footprint parameters are scored and optimized for the user. Prepared athlete meal recommendations are provided. Use Case – Allergen-Safety Priority Restaurant Use In another application, the user is an individual with a severe nut and gluten allergy. The user registers with the system via the restaurant's QR menu. It reaches. 5 The restriction-based pre-filtering module filters all dishes on the menu using allergen data. It compares the products with its warehouse and automatically identifies products that are risky for the user. This removes it from the system. Thanks to this process: Only safe foods are shown to the user, The risk of making the wrong choice is reduced. The processing load is being reduced, 15 Unnecessary score calculations are prevented. The system sorts remaining safe meals based on nutritional value, user preference, and environmental factors. Optimized safety features for the user, ranked according to sustainability criteria. It creates a menu. 20 Use Case – Sustainable and Low Carbon Footprint Food The election In one use case, the user selected a 25 that prioritizes environmental sustainability. The profile is selected. The user prefers meals with a low carbon footprint. This is indicated in the system profile. Found in the environmental impact data repository: Carbon footprint, 16 Water consumption, Energy consumption, Sustainability coefficients are evaluated by a scoring engine. The system lowers the score for dishes with a high environmental impact, with scores below 5. It highlights alternatives with a lower carbon footprint. Thus, to the user; Low carbon footprint, 10 In line with nutritional goals, Sustainable food options are suggested in real time. Use Case – Corporate Cafeteria and Catering Management In another application, the system is used in company cafeterias or mass food production. It is used in the facilities. Employees' health data and dietary preferences. It is defined in the user profile. The system analyzes the daily menu and finds: 20 Diabetes patients, Users following a low-sodium diet, Vegan users, Meals unsuitable for employees with allergen sensitivities 25 It filters. The system then generates personalized meal recommendations for each user, and The cafeteria kiosk screens display an optimized menu order. Thanks to this structure, in mass catering systems: 17 Wrong food choices are being reduced, User satisfaction is being increased, Process efficiency is being increased. Use Case – Adaptive Learning and Behavioral Optimization In one use case, the system analyzes the user's past ordering behavior. It automatically updates the parametric weights. The user constantly... It was determined that he preferred high-protein and low-carbohydrate meals. is being done. 10 Adaptive learning layer: Recurring ordering patterns, Order cancellations, Click behaviors, 15 Revising score calculation coefficients by analyzing user feedback. It optimizes. Thus, the system learns the user's habits over time and becomes more It offers highly accurate personalized meal recommendations. 20 FIGURES AND THEIR DESCRIPTIONS Figure 1. Multi-Criteria Analysis Based on the Processing of User Data and Food Data. Workflow Diagram of the Food Selection and Optimization Method 25 Figure 2. Food Choices Based on Nutrition, Allergen and Environmental Data. Modular System Architecture Diagram of the Optimization System Figure 3. Multiparameter Goodness-of-Fit Score Calculation and Decision-Making Process. Scoring Diagram

Claims

18 REQUESTS 1) A computer that provides personalized food selection and optimization for the user. It is a supported method; • User's nutritional goals, allergen restrictions, health data, 5 retrieval of user preferences and contextual data, • Nutritional information, allergen data and environmental impact data for foods obtaining from data sources, • To be used in a common computational structure for the data in question. normalization, 10 • Dishes unsuitable for the user will be excluded from the scoring process. first, it must be subjected to a restriction-based pre-filtering process, • Multi-criteria suitability for candidate dishes remaining after filtering. Calculating the score using a weighted parametric mathematical model, • Meals are sorted and optimized according to their calculated suitability scores. Creating suggested meals, • The filtering process should be performed before the score calculation process. by reducing the workload and improving real-time decision-making performance It is a method characterized by including steps to increase efficiency. 2) This method, according to Claim 1, is characterized by the fact that the pre-filtering process in question; • Foods containing allergens, • Foods that do not comply with health restrictions, • Candidate data 25 for dishes containing user-banned content. This includes automatic removal from the group. 3) This method, according to Claim 1, is characterized by its multi-criteria suitability score; • Nutritional suitability, 30 • Allergen safety, 19 • Environmental impact, • User preference matching, • Weighted combined fit model of contextual fit parameters It is calculated by considering all the elements together. 4) The method according to claim 3 is characterized by its environmental impact parameter; carbon footprint, water consumption, energy consumption and sustainability coefficients It includes. 5) Method according to Claim 1, its characteristics include user behavior and past orders. based on data, feedback records and recurring selection patterns The parameter weights used in the fitness score are dynamically adjusted. It is an update. 15 6) This method, according to Claim 1, is characterized by the use of artificial intelligence to fill in missing food data. This is completed through a supported data forecasting module. 7) It is a system that provides personalized meal selection and optimization for the user; • User interface configured to retrieve user data, • Data management layer that stores food data, • Processing and optimization of user data and food data 25 engine, • A decision support layer that generates optimized decision outcomes, • Dynamically adjusting parameter weights based on user behavior. It is characterized by containing an updated parametric weight optimization module. It is a system; the processing and optimization engine; 30 • Data normalization module, • Restriction-based pre-filtering module, • Multi-criteria score calculation module, • Includes a ranking and decision-making module and a scoring system for filtering. It is configured to be performed before the calculation process. 10 20 30

Images