Special medical food recommendation system based on big data analysis

Abstract

The invention relates to the technical field of food recommendation, in particular to a big data analysis-based special medical food recommendation system, which comprises a data acquisition module for acquiring user data and special medical food data; the glycemic index analysis module is used for constructing a special medical food glycemic index recommendation coefficient according to the collected special medical food glycemic index; the food analysis module is used for constructing a food scoring coefficient; the matching module is used for constructing the nutrition matching degree of the user and the special medical food; the recommendation module is used for constructing the food recommendation coefficient and recommending special medical food to the user according to a construction result; and the blood glucose monitoring module is used for analyzing the blood glucose state of the user according to the blood glucose data of the user collected in the monitoring period and adjusting the construction process of the food recommendation coefficient in the next monitoring period according to an analysis result. According to the invention, the recommendation efficiency of special medical food is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of food recommendation, and in particular to a special medical food recommendation system based on big data analysis. Background Art

[0002] Because of the changes in the body's blood sugar regulation mechanism, diabetic patients have different nutritional needs from healthy people. They need to pay special attention to the amount, quality and time of carbohydrate intake to maintain blood sugar stability. Traditional food recommendation methods or dietary guidelines often fail to match this demand in a refined manner, especially ignoring the impact of food on blood sugar (i.e., differences in glycemic index) and its impact on future blood sugar fluctuations.

[0003] Chinese patent publication number CN118315024A discloses a special medical food recommendation system and method based on artificial intelligence, which relates to the field of intelligent food recommendation. It collects the medical records, health status, food preferences and treatment target data of the patient object, and introduces artificial intelligence-based data processing and semantic understanding algorithms at the back end to perform semantic analysis and association coding of these patient object information, so as to capture the semantics of the patient object's potential dietary needs. At the same time, it also performs semantic analysis on the text descriptions of different special medical foods to capture the semantics of different special medical foods. Then, by calculating the semantic alignment scores between the patient object's potential dietary needs and the semantics of different special medical foods, it is determined whether the special medical food meets the patient object's potential dietary needs, thereby providing personalized food recommendations for patients with specific disease states to recommend suitable special medical foods; it can be seen that this solution only analyzes user information and special medical food information, and there are problems such as low efficiency in special medical food recommendation and poor user experience. Summary of the invention

[0004] The object of the present invention is to provide a special medical food recommendation system based on big data analysis to solve at least one of the problems existing in the prior art.

[0005] To achieve the above object, the present invention adopts the following technical solutions: Data collection module, used to collect user data and special medical food data; The glycemic index analysis module is used to construct the recommended glycemic index coefficient of special medical foods based on the collected glycemic index of special medical foods; Food analysis module, used to construct food scoring coefficients based on the collected special medical food scores and the number of special medical food scores; A matching module is used to match the collected protein demand, fat demand and carbohydrate demand of the user with the protein content, fat content and carbohydrate content of the special medical food, so as to build a nutritional matching degree between the user and the special medical food; A recommendation module, which is used to construct a food recommendation coefficient based on the construction results of the glycemic index recommendation coefficient of special medical food, the construction results of the food scoring coefficient, and the construction results of the nutritional matching degree between the user and the special medical food, and recommend special medical food to the user based on the construction results; The blood sugar monitoring module is used to analyze the user's blood sugar status based on the user's blood sugar data collected during the monitoring period, and adjust the construction process of the food recommendation coefficient for the next monitoring period based on the analysis results.

[0006] Furthermore, the glycemic index analysis module is used to construct a recommended glycemic index coefficient for special medical foods based on the collected glycemic index of the special medical foods. If Ti≤t1, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT1; if t1<Ti<t2, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT2; if Ti≥t2, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT3, where Ti is the glycemic index of the i-th special medical food, t1 is the first preset glycemic index, and t2 is the second preset glycemic index.

[0007] Furthermore, the food analysis module includes a scoring analysis unit and a quantity analysis unit. The scoring analysis unit is used to compare the collected i-th special medical food score with the preset score f0 to analyze the abnormality of the special medical food score. The abnormality of the special medical food score includes normal and abnormal. If the i-th special medical food score is abnormal, the scoring index of the i-th special medical food is set to F1; if the i-th special medical food score is normal, the scoring index of the i-th special medical food is set to F2.

[0008] Furthermore, the quantity analysis unit is used to compare the collected number of special medical food scores of the i-th type with the preset number m1 to analyze the abnormality of the number of special medical food scores. The abnormality of the number of special medical food scores includes normal and abnormal. If the number of scores of the i-th special medical food is abnormal, the preset score is adjusted to f0'.

[0009] Furthermore, the matching module matches the collected protein demand, fat demand and carbohydrate demand of the user with the protein content, fat content and carbohydrate content of the special medical food to construct a nutritional matching degree between the user and the special medical food, and sets the nutritional matching degree between the user and the i-th special medical food as Pi.

[0010] Furthermore, the recommendation module includes a recommendation unit and a history analysis unit, wherein the recommendation unit is used to construct a food recommendation coefficient according to the construction result of the glycemic index recommendation coefficient of the special medical food, the construction result of the food scoring coefficient, and the construction result of the nutritional matching degree between the user and the special medical food, and set the food recommendation coefficient of the i-th special medical food to TJi; The recommendation unit sorts the food recommendation coefficients of each special medical food in descending order, and recommends the special medical foods with the top three food recommendation coefficients as recommended foods to the user.

[0011] Furthermore, the historical analysis unit is used to compare the historical push times di1 of the i-th special medical food with the push threshold d0 to analyze the abnormality of the push times, and the abnormality of the push times includes normal and abnormal. If the historical push times of the i-th special medical food are abnormal and normal, when di2 / di1≤γ, the historical analysis unit adjusts the food recommendation coefficient of the i-th special medical food to TJi', di2 is the historical selection times of the i-th special medical food, and γ is the selection rate threshold.

[0012] Furthermore, the blood glucose monitoring module includes a blood glucose monitoring unit, a state analysis unit and an abnormality analysis unit. The blood glucose monitoring unit is used to compare the user's blood glucose data collected during the monitoring period with the preset blood glucose RT, and analyze the abnormality of the user's blood glucose data based on the comparison result. The abnormality of the user's blood glucose data includes normal and abnormal.

[0013] Furthermore, the status analysis unit is used to analyze the user's blood sugar status according to the analysis results of the abnormality of the user's blood sugar data during the monitoring period, and the user's blood sugar status includes normal and abnormal; if the user's blood sugar status in the current monitoring period is abnormal, the first preset blood sugar index is adjusted to t1', and the second preset blood sugar index is adjusted to t2'.

[0014] Furthermore, the abnormality analysis unit is used to construct a fluctuation coefficient B of the user's blood sugar data based on the user's blood sugar data collected during the monitoring period, and compare B with a preset fluctuation coefficient b to analyze the fluctuation state of the user's blood sugar data. The fluctuation state of the user's blood sugar data includes normal and abnormal. If the fluctuation state of the user's blood sugar data in the current monitoring period is abnormal, the preset abnormal proportion is adjusted to β'.

[0015] The beneficial effects of the present invention are as follows: the system is based on existing blood sugar monitoring equipment and big data technology, and uses machine learning and big data analysis to develop a dynamic personalized recommendation model. The system has automated analysis and intelligent food recommendation functions. Compared with the traditional purely manual experience recommendation method, it can more accurately and continuously provide users with special medical food recommendations that meet their conditions, especially suitable for the dietary management of diabetic patients and patients with chronic diseases. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 This is a schematic diagram of the structure of the special medical food recommendation system based on big data analysis in this embodiment.

[0018] Figure 2 Schematic diagram of the structure of the food analysis module of this embodiment.

[0019] Figure 3 This is a schematic diagram of the structure of the module recommended in this embodiment.

[0020] Figure 4 Schematic diagram of the structure of the blood glucose monitoring module of this embodiment. DETAILED DESCRIPTION

[0021] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and accompanying drawings. Similar components in the accompanying drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the content specifically described below is illustrative rather than restrictive, and should not be used to limit the scope of protection of the present invention.

[0022] It should be noted that, although the terms first, second, third, etc. may be used to describe in the embodiments of the present application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, without departing from the scope of the embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first.

[0023] See also Figure 1 As shown, it is a structural schematic diagram of the special medical food recommendation system based on big data analysis in this embodiment, and the system includes: The data collection module is used to collect user data and special medical food data, wherein the user data includes user nutritional needs and user preferences, the nutritional needs include protein needs, fat needs and carbohydrate needs, and the special medical food data includes glycemic index, protein content, fat content and carbohydrate content in special medical food; the nutritional needs described in this embodiment are the daily nutritional needs of the user; in this embodiment, no specific limitation is made on the collection method of user data and special medical food data, and those skilled in the art can freely set it as long as the collection requirements of user data and special medical food data are met, wherein the user data and special medical food data can be collected through interaction.

[0024] Please continue reading Figure 1 As shown, the system includes a glycemic index analysis module, which is connected to the data acquisition module and is used to construct a recommended glycemic index coefficient of the special medical food according to the collected glycemic index of the special medical food, so as to improve the accuracy of the construction of the recommended glycemic index coefficient of the special medical food, wherein: If Ti≤t1, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT1, and sets XT1=1; if t1<Ti<t2, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT2, and sets XT2=exp[-(Ti-t1) / t1]; if Ti≥t2, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT3, and sets XT3=0, Ti is the glycemic index of the i-th special medical food, t1 is the first preset glycemic index, t2 is the second preset glycemic index, and t1<t2; foods are classified according to the glycemic index of special medical foods to ensure that the recommended foods have better blood sugar regulation capabilities. Through piecewise function design, the system's ability to handle abnormal values ​​is improved, and the recommendation strategy can be adjusted according to different user needs.

[0025] Specifically, in this embodiment, there is no specific limitation on the setting of each preset blood sugar index, and those skilled in the art can set it freely as long as the setting requirements of each preset blood sugar index are met. Among them, the optimal value of t1 is 55, and the optimal value of t2 is 70.

[0026] Please continue reading Figure 1 As shown, the system includes a food analysis module, which is used to construct a food score coefficient based on the collected special medical food scores and the number of special medical food scores.

[0027] Specifically, the method for collecting the special medical food scores and the number of special medical food scores is not specifically limited in this embodiment, and technical personnel in this field can set it freely, as long as the collection requirements of the special medical food scores and the number of special medical food scores are met. Among them, the special medical food scores and the number of special medical food scores can be collected by importing the system background data.

[0028] See also Figure 2 As shown, the food analysis module includes a scoring analysis unit, which is used to compare the collected special medical food score with the preset score f0, and construct a special medical food scoring index according to the comparison result to improve the accuracy of the construction of the special medical food scoring index, wherein: If fi≤f0, the scoring analysis unit determines that the score of the i-th special medical food is abnormal, and sets the scoring index of the i-th special medical food to F1, setting F1=0; if fi>f0, the scoring analysis unit determines that the score of the i-th special medical food is normal, and sets the scoring index of the i-th special medical food to F2, setting F2=(fi-f0) / f0, fi is the score of the i-th special medical food; by identifying foods with abnormal scores, the accuracy of constructing the scoring index is improved, thereby improving the accuracy of recommending special medical foods to users.

[0029] Specifically, the setting of the preset score is not specifically limited in this embodiment, and those skilled in the art can set it freely as long as the setting requirements of the preset score are met. Among them, when the score is a 5-point system, the best value of f0 is 4.

[0030] Please continue reading Figure 2 As shown, the food analysis module includes a quantity analysis unit, which is connected to the score analysis unit and is used to compare the collected special food score quantity with the preset quantity m1 to analyze the abnormality of the special medical food score quantity, and adjust the construction process of the special medical food score index according to the analysis result, wherein: If mi0≤m1, the quantity analysis unit determines that the number of scores for the i-th special medical food is normal and does not make any adjustments; If mi0>m1, the quantity analysis unit determines that the number of scores for the i-th special medical food is abnormal, and adjusts the preset score to f0', setting: f0'=f0×[1-α×(mi0-m1) / (mi0+m0)], α is the preset adjustment coefficient, and mi0 is the number of scores for the i-th special medical food; by analyzing the number of scores, it avoids over-recommendation of foods with very few scores, and makes reasonable adjustments to the scoring criteria to increase the objectivity and credibility of the recommendations.

[0031] Specifically, in this embodiment, there is no specific limitation on the setting of the preset number and the preset adjustment coefficient, and those skilled in the art can set them freely as long as the setting requirements of the preset number and the preset adjustment coefficient are met. Among them, the best value of m1 is 100, and the best value of α is 0.18.

[0032] Please continue reading Figure 1 As shown, the system includes a matching module, which is connected to the data collection module to match the collected user's protein demand, fat demand and carbohydrate demand with the protein content, fat content and carbohydrate content of the special medical food to build a nutritional matching degree between the user and the special medical food, wherein: The matching module sets the nutritional matching degree between the user and the i-th special medical food as Pi, and sets: , Hin is the content of the nth nutrient in the i-th special medical food, Qn is the user's demand for the nth nutrient, Wn is the weight of the nth nutrient, N is the number of nutrients, n=1 is protein, n=2 is fat, and n=3 is carbohydrate.

[0033] Specifically, in this embodiment, there is no specific limitation on the setting of nutrient weights, and those skilled in the art can set them freely as long as the setting requirements of nutrient weights are met, wherein W1+W2+W3=1, the optimal value of W1 is 0.3, the optimal value of W2 is 0.3, and the optimal value of W3 is 0.4; the user's protein, fat and carbohydrate requirements are matched with the corresponding nutrients in the special medical food to ensure that the recommended food can meet the user's nutritional needs, the personalization of the user's nutritional needs can be achieved, and the recommendation effect is improved.

[0034] Please continue reading Figure 1 As shown, the system includes a recommendation module, which is connected to the glycemic index analysis module, the food analysis module and the matching module. The recommendation module is used to construct a food recommendation coefficient according to the construction results of the glycemic index recommendation coefficient of the special medical food, the construction results of the food scoring coefficient and the construction results of the nutritional matching degree between the user and the special medical food, and recommend the special medical food to the user according to the construction results, and also to adjust the recommendation process of the special medical food according to the historical push times and historical selection times of the special medical food.

[0035] Specifically, the historical push times of special medical foods described in this embodiment are the times that special medical foods are pushed to users as recommended foods, and the historical selection times are the times that users select special medical foods after they are pushed to users as recommended foods. In this embodiment, no specific limitation is made on the collection method of the historical push times and the historical selection times of special medical foods, and those skilled in the art may freely set them as long as the collection requirements of the historical push times and the historical selection times of special medical foods are met. Among them, the historical push times and the historical selection times of special medical foods can be obtained by importing system background data.

[0036] See also Figure 3 As shown, the recommendation module includes a recommendation unit, which is used to construct a food recommendation coefficient according to the construction results of the glycemic index recommendation coefficient of special medical food, the construction results of the food scoring coefficient and the construction results of the nutritional matching degree between the user and the special medical food, so as to improve the accuracy of the food recommendation coefficient construction, and recommend special medical food to the user according to the construction results, wherein: The recommendation unit sets the food recommendation coefficient of the i-th special medical food as TJi, and sets: TJi=E1×glycemic index recommendation coefficient of the i-th special medical food+E2×scoring index of the i-th special medical food+E3×Pi; E1 is the glycemic index weight, E2 is the scoring weight, E3 is the nutrition matching weight, and E1+E2+E3=1; The recommendation unit sorts the food recommendation coefficients of each special medical food in descending order, and recommends the special medical foods with the top three food recommendation coefficients to the user as recommended foods; the food recommendation coefficients are calculated by comprehensively considering multiple factors such as glycemic index, user rating index, nutritional matching, etc., to ensure that the recommended foods can meet the user's various needs, thereby improving the recommendation efficiency of special medical foods.

[0037] Specifically, in this embodiment, there is no specific limitation on the setting of the glycemic index weight, the scoring weight and the nutritional matching weight. Those skilled in the art can set them freely as long as the setting requirements of the glycemic index weight, the scoring weight and the nutritional matching weight are met. Among them, the optimal value of E1 is 0.4, the optimal value of E2 is 0.2, and the optimal value of E3 is 0.4.

[0038] Specifically, the sorting method is not specifically limited in this embodiment, and those skilled in the art can freely set it as long as the sorting requirements are met. Among them, the sorting method can be set to a bubble sort method.

[0039] Please continue reading Figure 3 As shown, the recommendation module includes a history analysis unit, which is connected to the recommendation unit and is used to adjust the recommendation process of the special medical food according to the historical push times di1 and the historical selection times di2 of the special medical food, wherein: If di1≤d0, the history analysis unit determines that the number of historical pushes of the i-th special medical food is abnormal and does not make any adjustments; If di0>d0, the historical analysis unit determines that the historical push times of the i-th special medical food are normal. When di2 / di1≤γ, the historical analysis unit determines that the selection rate of the i-th special medical food is abnormal and does not make any adjustments. When di2 / di1>γ, the historical analysis unit determines that the selection rate of the i-th special medical food is normal, and adjusts the food recommendation coefficient of the i-th special medical food to TJi', setting TJi'=TJi+D×(di2 / di1-γ) / γ, di1 is the historical push times of the i-th special medical food, di2 is the historical selection times of the i-th special medical food, d0 is the push threshold, γ is the selection rate threshold, and D is the ratio threshold. By analyzing the historical push times and selection rates of food, the recommendation algorithm is dynamically adjusted to improve the intelligence and responsiveness of the system, thereby improving the recommendation efficiency of special medical food.

[0040] Specifically, in this embodiment, there is no specific limitation on the setting of the push threshold, the selection rate threshold and the ratio threshold. Those skilled in the art can set them freely as long as the setting requirements of the push threshold, the selection rate threshold and the ratio threshold are met. Among them, the best value of d0 is 80, the best value of γ is 0.7, and the best value of D is 0.2.

[0041] Please continue reading Figure 1 As shown, the system includes a blood glucose monitoring module, which is connected to the recommendation module. The blood glucose monitoring module is used to analyze the user's blood glucose status based on the user's blood glucose data collected during the monitoring period, and adjust the construction process of the food recommendation coefficient for the next monitoring period based on the analysis results.

[0042] Specifically, the present embodiment does not impose any specific limitation on the setting of the monitoring period, and those skilled in the art may freely set it as long as the setting requirements of the monitoring period are met, wherein the monitoring period may be set to 60 days, 90 days, etc.

[0043] Specifically, the present embodiment does not specifically limit the method for collecting the user's blood sugar data, and those skilled in the art can freely set it, as long as it meets the requirements for collecting the user's blood sugar data, such as being able to obtain it through a blood sugar meter.

[0044] See also Figure 4 As shown, the blood glucose monitoring module includes a blood glucose monitoring unit, which is used to compare the user's blood glucose data collected during the monitoring period with the preset blood glucose RT to improve the accuracy of the abnormality analysis of the user's blood glucose data, and analyze the abnormality of the user's blood glucose data according to the comparison result, wherein: If js≤RT, the blood glucose monitoring unit determines that the blood glucose data of the sth user in the current monitoring period is normal; if js>RT, the blood glucose monitoring unit determines that the blood glucose data of the sth user in the current monitoring period is abnormal, and js is the blood glucose data of the sth user collected in the monitoring period.

[0045] Specifically, the setting of the preset blood sugar is not specifically limited in this embodiment, and those skilled in the art can set it freely as long as the setting requirements of the preset blood sugar are met. Among them, the optimal value of RT is 7.8mmol / L.

[0046] Please continue reading Figure 4 As shown, the blood glucose monitoring module includes a state analysis unit, which is connected to the blood glucose monitoring unit and is used to analyze the user's blood glucose state according to the analysis result of the abnormality of the user's blood glucose data during the monitoring period, and adjust the construction process of the food recommendation coefficient for the next monitoring period according to the analysis result of the user's blood glucose state, wherein: When ys / S≤β, the state analysis unit determines that the user's blood sugar state in the current monitoring period is normal; when ys / S>β, the state analysis unit determines that the user's blood sugar state in the current monitoring period is abnormal, and adjusts the first preset blood sugar index to t1', setting t1'=t1×{1-ln[3×(ys / S-β)+1] / ln4}, and adjusts the second preset blood sugar index to t2', setting t2'=t2×{1-ln[3×(ys / S-β)+1] / ln4}; Wherein, ys is the number of abnormal blood glucose data of users during the monitoring period, S is the number of blood glucose data of users collected during the monitoring period, β is the preset abnormal proportion, if t1'≤t3, t1' is taken as t3, if t2'≤t4, t2' is taken as t4, t3 is the first blood glucose index threshold, t4 is the second blood glucose index threshold, t3<t4; according to the abnormal blood glucose data of the user during the monitoring period, the blood glucose-related indicators are dynamically adjusted to adapt to the changes in the user's physical condition, so as to ensure that food recommendations can play a positive role in the user's blood glucose management.

[0047] Specifically, in this embodiment, there is no specific limitation on the setting of the preset abnormal proportion and each blood sugar index threshold value. Those skilled in the art can set them freely as long as the setting requirements of the preset abnormal proportion and each blood sugar index threshold are met. Among them, the optimal value of t3 is 45, and the optimal value of t4 is 60.

[0048] Please continue reading Figure 4 As shown, the blood glucose monitoring module includes an abnormality analysis unit, which is connected to the state analysis unit and is used to construct a fluctuation coefficient B of the user's blood glucose data according to the user's blood glucose data collected during the monitoring period, and set: , , jsr is the average value of the user's blood sugar data collected during the monitoring period; The abnormality analysis unit compares B with a preset fluctuation coefficient b to analyze the fluctuation state of the user's blood sugar data, and adjusts the analysis process of the user's blood sugar state according to the analysis result, wherein: If B≤b, the abnormality analysis unit determines that the fluctuation state of the user's blood sugar data in the current monitoring period is normal and no adjustment is made; If B>b, the abnormality analysis unit determines that the fluctuation state of the user's blood sugar data in the current monitoring period is abnormal, and adjusts the preset abnormality ratio to β', setting β'=β×{1+[e 4×（B-b） -1] / [e 4 -1]}, e is the natural logarithm; by calculating the blood sugar fluctuation coefficient, analyzing its fluctuation state, and further refining the system adjustment mechanism, the recommendation efficiency of special medical foods is improved.

[0049] Specifically, in this embodiment, the value of the preset fluctuation coefficient is not specifically limited, and those skilled in the art can freely set it as long as it meets the value requirement of the preset fluctuation coefficient, where the optimal value of b is 0.15.

[0050] Specifically, the system described in this embodiment is specifically used for personalized dietary recommendations for diabetic patients. The system classifies special medical foods according to their glycemic index, and screens and analyzes them according to the user's personal eating habits and special medical food scores. At the same time, it calculates the supply and demand matching of the user's required nutrition, such as protein, fat and carbohydrates, and comprehensively considers multiple factors such as glycemic index, user scoring index, nutritional matching, etc. to calculate the food recommendation coefficient and make food recommendations to the user.

[0051] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not limitations on the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the embodiments here. All obvious changes or modifications derived from the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A special medical food recommendation system based on big data analysis, characterized in that: include: Data collection module, used to collect user data and special medical food data; The glycemic index analysis module is used to construct the recommended glycemic index coefficient of special medical foods based on the collected glycemic index of special medical foods; Food analysis module, used to construct food scoring coefficients based on the collected special medical food scores and the number of special medical food scores; A matching module is used to match the collected protein demand, fat demand and carbohydrate demand of the user with the protein content, fat content and carbohydrate content of the special medical food, so as to build a nutritional matching degree between the user and the special medical food; A recommendation module, which is used to construct a food recommendation coefficient based on the construction results of the glycemic index recommendation coefficient of special medical food, the construction results of the food scoring coefficient, and the construction results of the nutritional matching degree between the user and the special medical food, and recommend special medical food to the user based on the construction results; The blood sugar monitoring module is used to analyze the user's blood sugar status based on the user's blood sugar data collected during the monitoring period, and adjust the construction process of the food recommendation coefficient for the next monitoring period based on the analysis results.

2. The FSMP recommendation system based on big data analysis according to claim 1, characterized in that: The glycemic index analysis module is used to construct the recommended glycemic index coefficient of special medical food according to the collected glycemic index of the special medical food. If Ti≤t1, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT1; if t1<Ti<t2, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT2; if Ti≥t2, the glycemic index analysis module sets the recommended glycemic index coefficient of the i-th special medical food to XT3, where Ti is the glycemic index of the i-th special medical food, t1 is the first preset glycemic index, and t2 is the second preset glycemic index.

3. The FSMP recommendation system based on big data analysis according to claim 2, characterized in that: The food analysis module includes a scoring analysis unit and a quantity analysis unit. The scoring analysis unit is used to compare the collected i-th special medical food score with the preset score f0 to analyze the abnormality of the special medical food score. The abnormality of the special medical food score includes normal and abnormal. If the i-th special medical food score is abnormal, the scoring index of the i-th special medical food is set to F1; if the i-th special medical food score is normal, the scoring index of the i-th special medical food is set to F2.

4. The FSMP recommendation system based on big data analysis according to claim 3, characterized in that: The quantity analysis unit is used to compare the collected number of special medical food scores of the i-th type with the preset number m1 to analyze the abnormality of the number of special medical food scores. The abnormality of the number of special medical food scores includes normal and abnormal. If the number of scores of the i-th special medical food is abnormal, the preset score is adjusted to f0'.

5. The FSMP recommendation system based on big data analysis according to claim 4, characterized in that: The matching module matches the collected user's protein demand, fat demand and carbohydrate demand with the protein content, fat content and carbohydrate content of the special medical food to construct the nutritional matching degree between the user and the special medical food, and sets the nutritional matching degree between the user and the i-th special medical food as Pi.

6. The FSMP recommendation system based on big data analysis according to claim 5, characterized in that: The recommendation module includes a recommendation unit and a history analysis unit, wherein the recommendation unit is used to construct a food recommendation coefficient according to the construction result of the glycemic index recommendation coefficient of the special medical food, the construction result of the food scoring coefficient, and the construction result of the nutritional matching degree between the user and the special medical food, and set the food recommendation coefficient of the i-th special medical food to TJi; The recommendation unit sorts the food recommendation coefficients of each special medical food in descending order, and recommends the special medical foods with the top three food recommendation coefficients as recommended foods to the user.

7. The FSMP recommendation system based on big data analysis according to claim 6, characterized in that: The historical analysis unit is used to compare the historical push times di1 of the i-th special medical food with the push threshold d0 to analyze the abnormality of the push times, and the abnormality of the push times includes normal and abnormal. If the historical push times of the i-th special medical food are abnormal and normal, when di2 / di1≤γ, the historical analysis unit adjusts the food recommendation coefficient of the i-th special medical food to TJi', di2 is the historical selection times of the i-th special medical food, and γ is the selection rate threshold.

8. The FSMP recommendation system based on big data analysis according to claim 7, characterized in that: The blood glucose monitoring module includes a blood glucose monitoring unit, a state analysis unit and an abnormality analysis unit. The blood glucose monitoring unit is used to compare the user's blood glucose data collected during the monitoring period with the preset blood glucose RT, and analyze the abnormality of the user's blood glucose data based on the comparison result. The abnormality of the user's blood glucose data includes normal and abnormal.

9. The FSMP recommendation system based on big data analysis according to claim 8, characterized in that: The state analysis unit is used to analyze the user's blood sugar state according to the analysis results of the abnormality of the user's blood sugar data within the monitoring period. The user's blood sugar state includes normal and abnormal; if the user's blood sugar state is abnormal in the current monitoring period, the first preset blood sugar index is adjusted to t1', and the second preset blood sugar index is adjusted to t2'.

10. The FSMP recommendation system based on big data analysis according to claim 9, characterized in that: The abnormality analysis unit is used to construct a fluctuation coefficient B of the user's blood sugar data based on the user's blood sugar data collected during the monitoring period, and compare B with a preset fluctuation coefficient b to analyze the fluctuation state of the user's blood sugar data. The fluctuation state of the user's blood sugar data includes normal and abnormal. If the fluctuation state of the user's blood sugar data in the current monitoring period is abnormal, the preset abnormal proportion is adjusted to β'.

Images