CN118069931B - Intelligent diet recommendation method for diabetics based on artificial intelligence - Google Patents
Intelligent diet recommendation method for diabetics based on artificial intelligence Download PDFInfo
- Publication number
- CN118069931B CN118069931B CN202410465044.5A CN202410465044A CN118069931B CN 118069931 B CN118069931 B CN 118069931B CN 202410465044 A CN202410465044 A CN 202410465044A CN 118069931 B CN118069931 B CN 118069931B
- Authority
- CN
- China
- Prior art keywords
- food materials
- food
- similarity
- diabetics
- food material
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 235000005911 diet Nutrition 0.000 title claims abstract description 45
- 230000037213 diet Effects 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 16
- 239000000463 material Substances 0.000 claims abstract description 328
- 235000013305 food Nutrition 0.000 claims abstract description 322
- 206010012601 diabetes mellitus Diseases 0.000 claims abstract description 141
- 230000036541 health Effects 0.000 claims abstract description 72
- 235000016709 nutrition Nutrition 0.000 claims abstract description 35
- 239000008280 blood Substances 0.000 claims abstract description 26
- 210000004369 blood Anatomy 0.000 claims abstract description 26
- 239000004615 ingredient Substances 0.000 claims abstract description 17
- 235000012041 food component Nutrition 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000010606 normalization Methods 0.000 claims description 11
- WQZGKKKJIJFFOK-GASJEMHNSA-N Glucose Natural products OC[C@H]1OC(O)[C@H](O)[C@@H](O)[C@@H]1O WQZGKKKJIJFFOK-GASJEMHNSA-N 0.000 claims description 8
- 239000008103 glucose Substances 0.000 claims description 8
- 239000000203 mixture Substances 0.000 claims description 8
- 235000015097 nutrients Nutrition 0.000 claims description 6
- 230000001105 regulatory effect Effects 0.000 claims description 4
- 230000035764 nutrition Effects 0.000 abstract description 15
- 235000019640 taste Nutrition 0.000 abstract description 10
- 238000011156 evaluation Methods 0.000 description 14
- NOESYZHRGYRDHS-UHFFFAOYSA-N insulin Chemical compound N1C(=O)C(NC(=O)C(CCC(N)=O)NC(=O)C(CCC(O)=O)NC(=O)C(C(C)C)NC(=O)C(NC(=O)CN)C(C)CC)CSSCC(C(NC(CO)C(=O)NC(CC(C)C)C(=O)NC(CC=2C=CC(O)=CC=2)C(=O)NC(CCC(N)=O)C(=O)NC(CC(C)C)C(=O)NC(CCC(O)=O)C(=O)NC(CC(N)=O)C(=O)NC(CC=2C=CC(O)=CC=2)C(=O)NC(CSSCC(NC(=O)C(C(C)C)NC(=O)C(CC(C)C)NC(=O)C(CC=2C=CC(O)=CC=2)NC(=O)C(CC(C)C)NC(=O)C(C)NC(=O)C(CCC(O)=O)NC(=O)C(C(C)C)NC(=O)C(CC(C)C)NC(=O)C(CC=2NC=NC=2)NC(=O)C(CO)NC(=O)CNC2=O)C(=O)NCC(=O)NC(CCC(O)=O)C(=O)NC(CCCNC(N)=N)C(=O)NCC(=O)NC(CC=3C=CC=CC=3)C(=O)NC(CC=3C=CC=CC=3)C(=O)NC(CC=3C=CC(O)=CC=3)C(=O)NC(C(C)O)C(=O)N3C(CCC3)C(=O)NC(CCCCN)C(=O)NC(C)C(O)=O)C(=O)NC(CC(N)=O)C(O)=O)=O)NC(=O)C(C(C)CC)NC(=O)C(CO)NC(=O)C(C(C)O)NC(=O)C1CSSCC2NC(=O)C(CC(C)C)NC(=O)C(NC(=O)C(CCC(N)=O)NC(=O)C(CC(N)=O)NC(=O)C(NC(=O)C(N)CC=1C=CC=CC=1)C(C)C)CC1=CN=CN1 NOESYZHRGYRDHS-UHFFFAOYSA-N 0.000 description 10
- 238000001914 filtration Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 102000004877 Insulin Human genes 0.000 description 5
- 108090001061 Insulin Proteins 0.000 description 5
- 229940125396 insulin Drugs 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 230000002641 glycemic effect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000037396 body weight Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000000378 dietary effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 235000021049 nutrient content Nutrition 0.000 description 2
- 235000003170 nutritional factors Nutrition 0.000 description 2
- 208000024891 symptom Diseases 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 206010006895 Cachexia Diseases 0.000 description 1
- 102000015781 Dietary Proteins Human genes 0.000 description 1
- 108010010256 Dietary Proteins Proteins 0.000 description 1
- 208000017701 Endocrine disease Diseases 0.000 description 1
- 206010020710 Hyperphagia Diseases 0.000 description 1
- 208000004880 Polyuria Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 150000001720 carbohydrates Chemical class 0.000 description 1
- 235000014633 carbohydrates Nutrition 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 235000013325 dietary fiber Nutrition 0.000 description 1
- 235000021196 dietary intervention Nutrition 0.000 description 1
- 208000016097 disease of metabolism Diseases 0.000 description 1
- 230000035619 diuresis Effects 0.000 description 1
- 208000026500 emaciation Diseases 0.000 description 1
- 230000002124 endocrine Effects 0.000 description 1
- 208000030172 endocrine system disease Diseases 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 206010016256 fatigue Diseases 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000002347 injection Methods 0.000 description 1
- 239000007924 injection Substances 0.000 description 1
- 210000004153 islets of langerhan Anatomy 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 208000030159 metabolic disease Diseases 0.000 description 1
- 239000003538 oral antidiabetic agent Substances 0.000 description 1
- 229940127209 oral hypoglycaemic agent Drugs 0.000 description 1
- 206010036067 polydipsia Diseases 0.000 description 1
- 208000022530 polyphagia Diseases 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
- G16H20/60—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to nutrition control, e.g. diets
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Epidemiology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Nutrition Science (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The invention relates to the technical field of intelligent diet recommendation, in particular to an artificial intelligence-based intelligent diet recommendation method for diabetics. Acquiring physical data of a diabetic patient, nutritional ingredient data of food materials and scoring the food materials; determining a health degree coefficient according to the body data; the related indexes of the scoring blood sugar content of the food materials are adjusted through the health degree coefficient, and the scoring confidence of the diabetics is determined; combining the similarity of the nutritional component data among the food materials, and determining the similarity among the food materials through scoring of the food materials weighted by the scoring confidence; and determining the intelligent score of the food materials according to the similarity among the food materials, the score of the diabetes patients on the food materials and the time length of last eating of the food materials, and carrying out diet recommendation on the diabetes patients. According to the invention, the preference of diabetics to the taste of the food materials is considered, the nutrition components of the food materials are ensured, and the accuracy of recommendation of the diabetics by the intelligent recommendation algorithm is improved.
Description
Technical Field
The invention relates to the technical field of intelligent diet recommendation, in particular to an artificial intelligence-based intelligent diet recommendation method for diabetics.
Background
Diabetes is a common endocrine and metabolic disease with a certain genetic tendency, and is manifested by polydipsia, polyphagia, diuresis, fatigue, emaciation and other symptoms. In the comprehensive prevention and treatment of diabetes mellitus, nutrition education and dietary intervention are one of the most basic and important treatment means, so that no matter whether diabetes mellitus is serious or serious, no matter insulin or oral hypoglycemic agents are used, dietary treatment is required to relieve the burden of islet cells, reduce blood sugar and improve symptoms. At present, common technologies such as machine learning, deep learning and the like can be combined with information such as personal health data, diet preference, nutritional requirements and the like of diabetics, an intelligent diet recommendation system can be developed, and the diabetics can be helped to better manage diet by providing personalized diet recommendation, menu recommendation, food material purchase guidelines and the like, so that the purposes of balancing nutrition supply and controlling blood sugar are achieved while the personal preference is met.
At present, a method for intelligently recommending diabetics is generally to consider a collaborative filtering mode to realize personalized diet advice for different users; however, in the traditional collaborative filtering method, the predictive scores of the food materials are quantified only by using the evaluation scores of different historical users on each food material, but when the food materials are recommended for diabetics, the scores of different users on the food materials are required to be considered, whether the nutritional components corresponding to different food materials are balanced or not, and whether the personal health condition can accept the food materials with different glycemic indexes or not is required to be considered.
Disclosure of Invention
In order to solve the technical problems that whether the nutrition components of diet are balanced or not is not considered in the traditional method for recommending the diet of the diabetics through collaborative filtering, the predictive score of the food is quantized only by using the evaluation scores of different historical users on each food, and the invention aims to provide the artificial intelligence-based method for recommending the diet of the diabetics, which adopts the following technical scheme:
Acquiring body data of a diabetic patient, nutrient component data of food materials and scores of the diabetic patient on the food materials, wherein the nutrient component data comprises blood sugar content;
determining a health degree coefficient of the diabetic patient according to the body data;
according to the health degree coefficient, the grading of the diabetics to the food materials and the related indexes of the blood sugar content of the food materials are regulated, and the grading confidence of the diabetics is determined;
Combining the similarity of the nutritional component data between the food materials and the scoring of the food materials by the diabetics weighted by the scoring confidence, determining the similarity between the food materials;
Determining intelligent grading of the food materials according to the similarity among the food materials, grading of the food materials by diabetics and the time length of last eating of the food materials; and combining the intelligent grading of the food materials to carry out diet recommendation on the diabetics.
Preferably, the step of adjusting the related indexes of the score of the diabetes patient on the food material and the blood sugar content of the food material according to the health degree coefficient to determine the score confidence of the diabetes patient comprises the following steps:
Taking the ith diabetes patient as the current diabetes patient, the calculation formula of the grading confidence of the current diabetes patient is as follows:
；
Wherein, Scoring confidence for the ith diabetic patient; is a normalization function; A health degree coefficient for the ith diabetic patient; Is the first A favorite scoring sequence of a plurality of food materials evaluated by individual diabetics; Is the first A sequence of blood glucose levels of a plurality of food materials evaluated by individual diabetics; Is the first Pearson correlation coefficients of a favorite scoring sequence and a blood glucose content sequence of a plurality of food materials evaluated by individual diabetics.
Preferably, the determining the similarity between the food materials by combining the similarity of the nutritional composition data between the food materials and the scoring of the food materials by the diabetics weighted by the scoring confidence comprises:
determining a first food material similarity according to the similarity of the nutritional component data between the food materials;
determining the similarity between food materials by scoring the food materials by the diabetics weighted by the scoring confidence, and determining the second food material similarity;
And carrying out weighted summation on the first food material similarity and the second food material similarity based on the health degree coefficient of the diabetic patient to obtain the similarity between the food materials.
Preferably, the determining the first food material similarity according to the similarity of the nutritional component data between the food materials includes:
Taking the u-th food material and the v-th food material as examples, the calculation formula of the first food material similarity of the u-th food material and the v-th food material is as follows:
；
Wherein, A first food material similarity that is a nth food material and a v-th food material; is a normalization function; to be commonly evaluated by Seed material and the firstThe number of diabetics seeded with food material; to be commonly evaluated by Seed material and the firstScoring confidence of the ith diabetic patient of both food materials; to be commonly evaluated by Seed material and the firstIth diabetic patient pair of seed foodScoring the seed food materials; to be commonly evaluated by Seed material and the firstIth diabetic patient pair of seed foodScoring the seed food materials; For each diabetic patient The grading average of the seed food materials; For each diabetic patient The score average value of the food materials.
Preferably, the determining the similarity between food materials by scoring the food materials by the diabetics weighted by the scoring confidence degree, and determining the second food material similarity includes:
taking the u-th food material and the v-th food material as examples, the calculation formula of the second food material similarity of the u-th food material and the v-th food material is as follows:
；
Wherein, A second food material similarity that is a nth food material and a v-th food material; the dimension number of the nutritional ingredient data of the food material; is an exponential function based on natural constants; Is the first First of seed foodNutritional composition data in individual dimensions; Is the first First of seed foodNutritional composition data in each dimension.
Preferably, the weighting and summing the similarity of the first food material and the similarity of the second food material based on the health degree coefficient of the diabetic patient to obtain the similarity between food materials includes:
taking the normalized value of the health degree coefficient as the weight of the first food material similarity, and taking the normalized value of the difference between a preset threshold and the health degree coefficient as the weight of the second food material similarity;
And carrying out weighted summation on the first food material similarity and the second food material similarity to obtain the similarity between the food materials.
Preferably, the determining the intelligent score of the food materials according to the similarity between the food materials, the score of the diabetic to the food materials and the time length of last eating the food materials includes:
taking the ith diabetes patient as the current diabetes patient, wherein the calculation formula of the intelligent score is as follows:
；
Wherein the ith diabetic patient scores the z-th food material intelligently; The number of food materials evaluated for the ith diabetic; is a normalization function; The ith diabetic patient is last eaten Time difference between seed and food materials and the present day; Is the first Seed material and the firstSimilarity between the seed materials; For the ith diabetic patient Scoring the seed food material.
Preferably, the determining the health degree coefficient of the diabetic patient according to the body data includes:
The physical data of the diabetic patient is composed of physical characteristic data under a plurality of characteristic dimensions;
Selecting any characteristic dimension as a reference characteristic dimension, taking the difference value between the physical characteristic data of the current diabetic patient in the reference characteristic dimension and the diabetes data comparison value in the reference characteristic dimension as a numerator, taking the difference value between the health data comparison value in the reference characteristic dimension and the diabetes data comparison value as a denominator, and taking the ratio formed by the numerator and the denominator as a single health degree in the reference characteristic dimension;
The mean value of the single health degree of all characteristic dimensions is used as the health degree coefficient of the diabetic patient.
Preferably, the method for obtaining the health data control value comprises the following steps: and calculating the average value of the physical characteristic data of the healthy users in each characteristic dimension, wherein each characteristic dimension is provided with a corresponding healthy data comparison value as a healthy data comparison value.
Preferably, the method for obtaining the diabetes data control value comprises the following steps: and calculating the mean value of the physical characteristic data of the diabetics in each characteristic dimension, wherein each characteristic dimension is provided with a corresponding diabetes data comparison value as a diabetes data comparison value.
The embodiment of the invention has at least the following beneficial effects:
Aiming at the traditional situation that the differences of favorites and nutritional value characteristics of diabetics cannot be well reflected in the scores of different food materials when the diabetics are subjected to diet recommendation through a collaborative filtering method, the physical health data of different users are utilized to measure the physical health coefficients of the users, and the score confidence degrees corresponding to the diabetics are calculated by combining the scores of the food materials with different blood sugar coefficients of the users; the method has the advantages that the similarity measurement between raw food materials is corrected by combining the nutritional ingredient content of different food materials, so that the recommendation of users with different health conditions is more accurate, the score of each food material is further corrected by combining historical food data, namely the last time the user eats the food material, the nutrition ingredient diversity of the food materials is ensured while the favorite degree of the diabetes patient on the taste of the food material is considered, and the recommendation accuracy of a recommendation algorithm on the diet of the diabetes patient is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for providing an artificial intelligence based method for intelligent recommendation of diets for diabetics in accordance with one embodiment of the present invention;
Fig. 2 is a flowchart of a method for obtaining similarity between food materials according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of the artificial intelligence based intelligent recommendation method for diabetics based on the artificial intelligence according to the invention, and the specific implementation, structure, characteristics and effects thereof, with reference to the accompanying drawings and the preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The embodiment of the invention provides a specific implementation method of an artificial intelligence based diet intelligent recommendation method for diabetics, which is suitable for diet recommendation scenes of diabetics. In order to solve the technical problems that when the traditional collaborative filtering method is used for recommending the diets of diabetics, the predictive scores of the food materials are quantized only by using the evaluation scores of different historical users on the food materials, and whether the nutritional ingredients of the diets are balanced or not is not considered. According to the invention, on the basis of the traditional grading, the similarity measurement between different food materials is corrected by utilizing the physical health data of different users and the nutritional ingredient data of the food materials, so that the accuracy of recommendation is improved.
The following specifically describes a specific scheme of the artificial intelligence-based diabetes patient diet intelligent recommendation method provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of steps of an artificial intelligence based method for intelligent recommendation of diets for diabetics according to an embodiment of the invention is shown, the method comprising the steps of:
Step S100, acquiring physical data of a diabetic patient, nutrient content data of food materials and scores of the diabetic patient on the food materials, wherein the nutrient content data comprises blood sugar content.
Traditional collaborative filtering simply uses the preferences of a community of interest to recommend information of interest to a user, either by scoring or by having a common experience. The current common method for intelligently recommending the diet of the diabetics is to utilize the traditional method for recommending the diet of the diabetics. However, in the traditional collaborative filtering method, the predictive scores of the food materials are quantized by only using the evaluation scores of different historical users on each food material, but when the food materials are recommended for diabetics, the scores of different users on the food materials are considered, and factors such as whether the nutritional ingredients corresponding to different food materials are balanced or not are considered. Therefore, the invention corrects the similarity measurement between different food materials based on the traditional collaborative filtering method and according to the grading size by utilizing the physical health data of different diabetics and the nutritional ingredient data of the food materials, so as to improve the accuracy of diet recommendation of the diabetics.
Firstly, respectively collecting physical health data of multiple diabetics including weight, blood sugar level, insulin usage, etc., and respectively recording standardized data asTo the point ofAnd nutritional ingredient data and glycemic index (GI index) of different food materials, wherein the nutritional ingredient data of the food materials comprise energy, blood sugar content, protein, dietary fiber, carbohydrate, etc., and the standardized data is recorded asTo the point ofAnd further, according to the data conditions, on the basis of a traditional collaborative filtering algorithm, the predictive scores of different diabetics on food materials are improved. The insulin usage is daily injection.
The favorite scores of different users for food materials in the diet are obtained by analyzing the acquired behavior data of the diabetes patients, such as purchase history, browsing records and search records and combining the modes of user surveys or questionnaires, so that the favorite scores of the diabetes patients for each food material are obtained and stored.
In the embodiment of the invention, the favorite scores of the diabetics on the food materials can be the favorite scores of the diabetics on the food materials according to personal tastes or favorites, the favorite scores can be in the range of [0,10], and can also be in the range of [0,100], the favorite scores of each diabetics on the food materials are obtained through a user survey or questionnaire survey mode, and the favorite score sequences of the food materials evaluated by the diabetics are formed by the scores of various food materials. It should be noted that each diabetic patient corresponds to a favorite scoring sequence of food materials. The purpose of obtaining the scores of the diabetics on each food material is to avoid that the diabetics are only in the nutrition component when the diabetics are subjected to intelligent recommendation of diet, and the personal hobbies and tastes of the diabetics are completely ignored, and the subjective favorite scores of the diabetics on the food materials are used for adjusting the intelligent scores of the various food materials by combining the nutrition components of the food materials, so that the diabetics are subjected to the intelligent scores of the food materials obtained by comprehensive consideration.
In another embodiment of the invention, the existing nutrition scores of the food materials can be combined, simple weighting and averaging can be carried out on the personal favorite scores of the diabetics, the weighted and averaged result value is used as the score of the diabetics on the food materials, the weighted and averaged result value is used as the updated favorite score of the diabetics, namely the weighted and averaged result value is updated to the favorite score of the diabetics, and the favorite score sequence is obtained. The food nutrition score and the personal preference score of the diabetic patient are consistent in size.
Step S200, determining the health degree coefficient of the diabetic patient according to the body data.
Considering that the traditional user score is only a single numerical value, the traditional user score cannot better show whether the score is based on taste or nutritional ingredients, so that the reliability of historical score data is required to be considered before recommendation is carried out; therefore, on the basis of scoring, the similarity among the nutritional ingredients is considered, quantification can be carried out according to the content of substances contained in the food materials, so that the consideration of partial nutritional ingredients is removed on the basis of scoring, and the similarity acquired under different angles is fused according to the physical health data of different users as weight coefficients, so that the accuracy of a follow-up recommendation algorithm is improved.
Considering the difference of the physical health condition of each patient, as the physical health data corresponding to different users are different and a plurality of healthy users exist, the influence of the physical health data of different users on the grading is required to be considered when the follow-up recommendation is carried out, for example, the grading of the diabetes patient on the food material is more biased to the nutrition component of the user, and the relatively healthy user may consider the nutrition component of the food material, and also consider the factors such as the taste of the food material more, so that the comprehensive evaluation of the food material is carried out, and the health degree coefficient of different users is required to be quantified according to the physical state data of different users. A health factor of the diabetic patient can be determined from the body data.
The physical data of the diabetic patient is composed of physical feature data in a plurality of feature dimensions. That is, each feature dimension corresponds to one physical feature data, for example, when there are three physical feature data of body weight, blood glucose level, insulin usage, then there are three feature dimensions. Wherein the body data comprises body weight, blood glucose level, insulin usage.
And calculating the average value of the physical characteristic data of the healthy users in each characteristic dimension, wherein each characteristic dimension is provided with a corresponding healthy data comparison value as a healthy data comparison value.
And calculating the mean value of the physical characteristic data of the diabetics in each characteristic dimension, wherein each characteristic dimension is provided with a corresponding diabetes data comparison value as a diabetes data comparison value.
Selecting any characteristic dimension as a reference characteristic dimension, taking the difference value between the physical characteristic data of the current diabetic patient in the reference characteristic dimension and the diabetes data comparison value in the reference characteristic dimension as a numerator, taking the difference value between the health data comparison value in the reference characteristic dimension and the diabetes data comparison value as a denominator, and taking the ratio of the numerator and the denominator as a single health degree in the reference characteristic dimension.
The mean value of the single health degree of all characteristic dimensions is used as the health degree coefficient of the diabetic patient.
Taking the ith diabetes patient as a current diabetes patient, taking the r characteristic dimension as a reference characteristic dimension, and calculating the health degree coefficient of the current diabetes patient according to the formula:
；
Wherein, A health degree coefficient for the ith diabetic patient; a number of feature dimensions for the acquired body data; An r-th feature dimension personal health data for an i-th diabetic patient; the average value of physical characteristic data of a plurality of diabetics in the r characteristic dimension is the diabetes data comparison value; The average value of the physical characteristic data of the healthy users in the r characteristic dimension is the comparison value of the healthy data.
According to the obtainedData in each characteristic dimension are respectively calculated, and data differences of a single user and a diabetic patient in corresponding dimensions in each characteristic dimension are respectively calculatedAs an indicator size of the state of health of the body in a single characteristic dimension; and then utilizing the difference of the data mean value of the healthy user and the diabetic patient in the characteristic dimensionCompared with the prior art, the method eliminates the influence of dimensions in different dimensions, and further obtains the health degree coefficient of each diabetic patientThe smaller the value of (C) is, the more serious the diabetes is.
And step S300, regulating the scores of the diabetics on the food materials and the related indexes of the blood sugar content of the food materials according to the health degree coefficient, and determining the score confidence of the diabetics.
After the health degree coefficient of each diabetic patient is obtained, the evaluation condition of the user on the food materials under different health degrees is considered, and for the patient with smaller health degree coefficient, namely the patient with serious diabetes condition, the glycemic index of the selected food materials is usually in inverse relation with the evaluation index, the evaluation of factors such as taste of the food materials is ignored, and the evaluation of the food materials by partial diabetic patients is caused to be slightly deviated, so that the similarity degree among the subsequent food materials is greatly influenced, the confidence degree of each diabetic patient during the evaluation needs to be calculated, and the accuracy of the subsequent similarity calculation is further improved. The confidence level of a diabetic patient in an evaluation can also be understood as a referent level for the evaluation of the diabetic patient.
Taking the ith diabetes patient as the current diabetes patient, the calculation formula of the grading confidence of the current diabetes patient is as follows:
；
Wherein, Scoring confidence for the ith diabetic patient; is a normalization function; A health degree coefficient for the ith diabetic patient; Is the first A favorite scoring sequence of a plurality of food materials evaluated by individual diabetics; Is the first A sequence of blood glucose levels of a plurality of food materials evaluated by individual diabetics; Is the first Pearson correlation coefficients of a favorite scoring sequence and a blood glucose content sequence of a plurality of food materials evaluated by individual diabetics.
Wherein,The higher the index is, the more serious the current diabetic is, whereas the lower the index is, the more slight the current diabetic is. The worse the physical health of the diabetic patient, the worse the dietary control ability of the diabetic patient is reflected compared to healthy users. Meanwhile, the pearson correlation coefficients of the multiple food materials evaluated by the diabetics reflect the correlation degree of the two sequences, namely the approximation degree of the two sequences, so that when the correlation coefficient of the pearson correlation coefficients of the two sequences is higher, the food materials with higher blood sugar content are higher in the diabetic patients, the food materials with higher blood sugar content are favored by the diabetics, and the corresponding scoring confidence of the diabetics is lower.
Step S400, combining the similarity of the nutritional component data among the food materials and the scoring of the food materials by the diabetics weighted by the scoring confidence, and determining the similarity among the food materials.
After the scoring confidence degrees corresponding to the diabetics are obtained, the similarity among the foods scored by the original diabetics is weighted and corrected according to the obtained scoring confidence degrees, so that the influence of partial abnormal scores on the calculation of the similarity is reduced, but the similarity score eliminates the consideration of nutrition factors, so that the health degree coefficient is used as a weight factor between the rest factors such as food favorites and the health factors, and the similarity among the scoring similarity after correction and the similarity among food nutrition components is weighted, so that more accurate similarity measurement is obtained.
Referring to fig. 2, fig. 2 is a flowchart of a method for obtaining similarity between food materials. Combining the similarity of nutritional composition data between the food materials, and scoring the food materials by the diabetics weighted by the scoring confidence, the step of determining the similarity between the food materials is specific:
step S401, determining a first food material similarity according to the similarity of the nutritional component data between the food materials.
Taking the u-th food material and the v-th food material as examples, the calculation formula of the first food material similarity of the u-th food material and the v-th food material is as follows:
；
Wherein, A first food material similarity that is a nth food material and a v-th food material; is a normalization function; to be commonly evaluated by Seed material and the firstThe number of diabetics seeded with food material; to be commonly evaluated by Seed material and the firstScoring confidence of the ith diabetic patient of both food materials; to be commonly evaluated by Seed material and the firstIth diabetic patient pair of seed foodScoring the seed food materials; to be commonly evaluated by Seed material and the firstIth diabetic patient pair of seed foodScoring the seed food materials; For each diabetic patient The grading average of the seed food materials; For each diabetic patient The score average value of the food materials.
The similarity of the food materials is corrected through the similarity of the nutritional component data among the food materials, so that the influence of great scoring noise on the taste of the food materials caused by factors of diabetics can be avoided.
Step S402, determining the similarity between food materials and determining the second food material similarity through scoring the food materials by the diabetics weighted by the scoring confidence.
Taking the u-th food material and the v-th food material as examples, the calculation formula of the second food material similarity of the u-th food material and the v-th food material is as follows:
；
Wherein, A second food material similarity that is a nth food material and a v-th food material; the dimension number of the nutritional ingredient data of the food material; is an exponential function based on natural constants; Is the first First of seed foodNutritional composition data in individual dimensions; Is the first First of seed foodNutritional composition data in each dimension.
The similarity of the food materials is corrected by scoring the food materials by the diabetics weighted by the scoring confidence, the similarity is measured only according to the content of the nutritional ingredients of the food materials, and the obtained similarity measure is more prone to consideration of nutritional factors for patients with serious diabetes.
Step S403, performing weighted summation on the first food material similarity and the second food material similarity based on the health degree coefficient of the diabetic patient, to obtain the similarity between the food materials.
Taking the normalized value of the health degree coefficient as the weight of the first food material similarity, and taking the normalized value of the difference between a preset threshold and the health degree coefficient as the weight of the second food material similarity; in the embodiment of the present invention, the preset threshold value is 1, and in other embodiments, the operator adjusts the value according to the actual situation.
And carrying out weighted summation on the first food material similarity and the second food material similarity to obtain the similarity between the food materials.
Taking a nth food material and a nth food material as examples, a calculation formula of similarity between the nth food material and the nth food material is as follows:
；
Wherein, Similarity between the nth food material and the v-th food material; is a normalization function; A health degree coefficient for the ith diabetic patient; a first food material similarity that is a nth food material and a v-th food material; a second food material similarity that is a nth food material and a v-th food material; 1 is a preset threshold.
By means of similarity of nutritional component data among food materials and by means of grading of the food materials by diabetics weighted based on grading confidence, similarity of the food materials is corrected, so that the obtained similarity measurement of diabetics with low health degree coefficients can be more prone to consideration of nutritional factors, otherwise, diabetics with high health degree coefficients can be more prone to grading of the diabetics, and taste of users can be more fitted.
Step S500, determining intelligent scores of the food materials according to the similarity among the food materials, the scores of diabetics on the food materials and the time length of last eating of the food materials; and combining the intelligent grading of the food materials to carry out diet recommendation on the diabetics.
In order to make the food ingested by diabetics diversified, the predictive score of each food needs to be further improved by combining the historical edible food data of each diabetics, so as to reduce the predictive score of the food used recently, and be beneficial to the health of the diabetics.
Taking the ith diabetes patient as the current diabetes patient, the calculation formula of the intelligent score of the ith diabetes patient to the z-th food material is as follows:
；
Wherein the ith diabetic patient scores the z-th food material intelligently; The number of food materials evaluated for the ith diabetic; is a normalization function; The ith diabetic patient is last eaten Time difference between seed and food materials and the present day; Is the first Seed material and the firstSimilarity between the seed materials; For the ith diabetic patient Scoring the seed food material.
On the basis of the traditional predictive scoring, the time data of the eaten food materials are subjected to weighted correction after normalization, so that the accuracy of a recommendation algorithm is improved. When the number of times that a diabetic eats a certain food material is more, the problem of unbalanced body data is more likely to exist, so the food material is intelligently scored by further combining the time difference between the last use of the food material and the present day and the similarity between the scoring of the diabetic and the different types of the food material. The intelligent score combines analysis of multiple factors, is obtained through single data such as nutrition components or user evaluation, analyzes the nutrition components or the user evaluation, further analyzes the current physical health degree of the user on the basis, and finally determines a comprehensive intelligent score.
Selecting a plurality of diet packages corresponding to the food materials according to the scoring sizes of the plurality of food materials corresponding to each diabetic patient, namely respectively calculating the conditions of different food materials contained in different packages or finished products, weighting the scoring according to the content sizes of the food materials to obtain the scoring index of the package diet finally, and then sorting according to the scoring index sizes, and performing intelligent diet recommendation to the diabetic patient from the package diet with the score from high to low.
In summary, the invention relates to the technical field of intelligent diet recommendation. Acquiring body data of a diabetic patient, nutrient component data of food materials and scores of the diabetic patient on the food materials, wherein the nutrient component data comprises blood sugar content; determining a health degree coefficient of the diabetic patient according to the body data; according to the health degree coefficient, the grading of the diabetics to the food materials and the related indexes of the blood sugar content of the food materials are regulated, and the grading confidence of the diabetics is determined; combining the similarity of the nutritional component data between the food materials and the scoring of the food materials by the diabetics weighted by the scoring confidence, determining the similarity between the food materials; determining intelligent grading of the food materials according to the similarity among the food materials, grading of the food materials by diabetics and the time length of last eating of the food materials; and combining the intelligent grading of the food materials to carry out diet recommendation on the diabetics. According to the invention, the preference of diabetics to the taste of food materials is considered, the diversity of the nutritional ingredients of the food materials is ensured, and the accuracy of the intelligent recommendation algorithm for recommending the diabetics is improved.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (5)
1. An artificial intelligence based intelligent recommendation method for diets of diabetics is characterized by comprising the following steps:
Acquiring body data of a diabetic patient, nutrient component data of food materials and scores of the diabetic patient on the food materials, wherein the nutrient component data comprises blood sugar content;
determining a health degree coefficient of the diabetic patient according to the body data;
according to the health degree coefficient, the grading of the diabetics to the food materials and the related indexes of the blood sugar content of the food materials are regulated, and the grading confidence of the diabetics is determined;
Combining the similarity of the nutritional component data between the food materials and the scoring of the food materials by the diabetics weighted by the scoring confidence, determining the similarity between the food materials;
Determining intelligent grading of the food materials according to the similarity among the food materials, grading of the food materials by diabetics and the time length of last eating of the food materials; combining the intelligent grading of the food materials, and carrying out diet recommendation on the diabetics;
Wherein the determining the health degree coefficient of the diabetic patient according to the body data comprises: the physical data of the diabetic patient is composed of physical characteristic data under a plurality of characteristic dimensions;
Selecting any characteristic dimension as a reference characteristic dimension, taking the difference value between the physical characteristic data of the current diabetic patient in the reference characteristic dimension and the diabetes data comparison value in the reference characteristic dimension as a numerator, taking the difference value between the health data comparison value in the reference characteristic dimension and the diabetes data comparison value as a denominator, and taking the ratio formed by the numerator and the denominator as a single health degree in the reference characteristic dimension;
Taking the average value of the single health degree of all characteristic dimensions as the health degree coefficient of the diabetic patient;
The method for acquiring the similarity between the food materials comprises the following steps:
determining a first food material similarity according to the similarity of the nutritional component data between the food materials;
determining the similarity between food materials by scoring the food materials by the diabetics weighted by the scoring confidence, and determining the second food material similarity;
Weighting and summing the first food material similarity and the second food material similarity based on the health degree coefficient of the diabetic patient to obtain the similarity between food materials;
taking the u-th food material and the v-th food material as examples, the calculation formula of the first food material similarity of the u-th food material and the v-th food material is as follows:
；
Wherein, A first food material similarity that is a nth food material and a v-th food material; is a normalization function; to be commonly evaluated by Seed material and the firstThe number of diabetics seeded with food material; to be commonly evaluated by Seed material and the firstScoring confidence of the ith diabetic patient of both food materials; to be commonly evaluated by Seed material and the firstIth diabetic patient pair of seed foodScoring the seed food materials; to be commonly evaluated by Seed material and the firstIth diabetic patient pair of seed foodScoring the seed food materials; For each diabetic patient The grading average of the seed food materials; For each diabetic patient The grading average of the seed food materials;
taking the u-th food material and the v-th food material as examples, the calculation formula of the second food material similarity of the u-th food material and the v-th food material is as follows:
；
Wherein, A second food material similarity that is a nth food material and a v-th food material; the dimension number of the nutritional ingredient data of the food material; is an exponential function based on natural constants; Is the first First of seed foodNutritional composition data in individual dimensions; Is the first First of seed foodNutritional composition data in individual dimensions;
Wherein, based on the health degree coefficient of the diabetic patient, the first food material similarity and the second food material similarity are weighted and summed to obtain the similarity between the food materials, comprising:
taking the normalized value of the health degree coefficient as the weight of the first food material similarity, and taking the normalized value of the difference between a preset threshold and the health degree coefficient as the weight of the second food material similarity;
And carrying out weighted summation on the first food material similarity and the second food material similarity to obtain the similarity between the food materials.
2. The artificial intelligence based diabetes mellitus patient's diet intelligent recommendation method according to claim 1, wherein the adjusting the related index of the diabetes mellitus patient's score to the food material and the blood sugar content of the food material by the health degree coefficient, determining the confidence of the score of the diabetes mellitus patient, comprises:
Taking the ith diabetes patient as the current diabetes patient, the calculation formula of the grading confidence of the current diabetes patient is as follows:
；
Wherein, Scoring confidence for the ith diabetic patient; is a normalization function; A health degree coefficient for the ith diabetic patient; Is the first A favorite scoring sequence of a plurality of food materials evaluated by individual diabetics; Is the first A sequence of blood glucose levels of a plurality of food materials evaluated by individual diabetics; Is the first Pearson correlation coefficients of a favorite scoring sequence and a blood glucose content sequence of a plurality of food materials evaluated by individual diabetics.
3. The artificial intelligence based diabetes patient's diet intelligent recommendation method according to claim 1, wherein the determining the intelligent score of the food material based on the similarity between food materials, the score of the diabetes patient to the food material, and the length of time the food material was last consumed, comprises:
taking the ith diabetes patient as the current diabetes patient, wherein the calculation formula of the intelligent score is as follows:
；
Wherein, Intelligent scoring of the z-th food for the i-th diabetic patient; The number of food materials evaluated for the ith diabetic; is a normalization function; The ith diabetic patient is last eaten Time difference between seed and food materials and the present day; Is the first Seed material and the firstSimilarity between the seed materials; For the ith diabetic patient Scoring the seed food material.
4. The artificial intelligence based intelligent recommendation method for diets of diabetics according to claim 1, wherein the method for obtaining the health data control value is as follows: and calculating the average value of the physical characteristic data of the healthy users in each characteristic dimension, wherein each characteristic dimension is provided with a corresponding healthy data comparison value as a healthy data comparison value.
5. The artificial intelligence based intelligent recommendation method for diets of diabetics according to claim 1, wherein the method for obtaining the diabetes data control value is as follows: and calculating the mean value of the physical characteristic data of the diabetics in each characteristic dimension, wherein each characteristic dimension is provided with a corresponding diabetes data comparison value as a diabetes data comparison value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410465044.5A CN118069931B (en) | 2024-04-18 | 2024-04-18 | Intelligent diet recommendation method for diabetics based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410465044.5A CN118069931B (en) | 2024-04-18 | 2024-04-18 | Intelligent diet recommendation method for diabetics based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN118069931A CN118069931A (en) | 2024-05-24 |
| CN118069931B true CN118069931B (en) | 2024-06-28 |
Family
ID=91111175
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410465044.5A Active CN118069931B (en) | 2024-04-18 | 2024-04-18 | Intelligent diet recommendation method for diabetics based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118069931B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108565004A (en) * | 2018-04-24 | 2018-09-21 | 吉林大学 | A kind of introducing Adaboost probability matrixs decomposition diabetes personalization dietary recommendations continued method |
| CN113051391A (en) * | 2021-03-29 | 2021-06-29 | 深圳软通动力信息技术有限公司 | Personalized diet recommendation method based on food nutrition and health knowledge base |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200058404A1 (en) * | 2016-09-21 | 2020-02-20 | Rejuvenan Global Health, Inc. | Interactive computing system to generate customized preventive health information based on an individual's biomarkers |
| CN110504019A (en) * | 2019-08-30 | 2019-11-26 | 北京妙医佳健康科技集团有限公司 | User individual dietary recommendations continued method, apparatus, electronic equipment and storage medium |
| CN113936765A (en) * | 2021-12-17 | 2022-01-14 | 北京因数健康科技有限公司 | Method and device for generating periodic behavior report, storage medium and electronic equipment |
| CN114974512A (en) * | 2022-05-30 | 2022-08-30 | 中国银行股份有限公司 | Recommendation method and device for meal information |
| CN117573963A (en) * | 2022-08-05 | 2024-02-20 | 合肥美的电冰箱有限公司 | Food material recommending method, food material recommending device, refrigerator and storage medium |
| CN116403678A (en) * | 2023-03-24 | 2023-07-07 | 平安科技（深圳）有限公司 | Healthy diet recommendation method and device based on artificial intelligence and related equipment |
| CN117292792A (en) * | 2023-06-15 | 2023-12-26 | 赵国婧 | Recommendation method of personalized healthy diet based on intelligent algorithm |
-
2024
- 2024-04-18 CN CN202410465044.5A patent/CN118069931B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108565004A (en) * | 2018-04-24 | 2018-09-21 | 吉林大学 | A kind of introducing Adaboost probability matrixs decomposition diabetes personalization dietary recommendations continued method |
| CN113051391A (en) * | 2021-03-29 | 2021-06-29 | 深圳软通动力信息技术有限公司 | Personalized diet recommendation method based on food nutrition and health knowledge base |
Also Published As
| Publication number | Publication date |
|---|---|
| CN118069931A (en) | 2024-05-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110097946B (en) | Diet recommendation method based on nutrient analysis | |
| US20140052722A1 (en) | Optimization-based regimen method and system for personalized diabetes and diet management | |
| RU2472427C2 (en) | Mobile device, method and system for processing factors, influencing sugar level in blood | |
| WO2018148684A1 (en) | Nutrition scoring system | |
| CN108565004B (en) | Adaboost probability matrix-introduced diabetes personalized diet recommendation method | |
| US20040162678A1 (en) | Method of screening for disorders of glucose metabolism | |
| WO2021247922A1 (en) | Predictive guidance systems for personalized health and self-care, and associated methods | |
| US20040197846A1 (en) | Determination of glucose sensitivity and a method to manipulate blood glucose concentration | |
| JP5835580B2 (en) | Lifestyle improvement support system and analysis method of biological index | |
| CN112951430A (en) | Multiple chronic disease joint management apparatus and computer-readable storage medium | |
| CN115602289A (en) | Dynamic diet recommendation method based on crowd attribute-diet knowledge graph | |
| KR101777747B1 (en) | Method for individualized recommanding food for glycosuria patient | |
| JP2023020789A (en) | Program, method, and system | |
| JP5853348B2 (en) | Lifestyle analysis system and lifestyle analysis method | |
| CN115662617B (en) | Result illness state prediction system based on CGM and prediction method thereof | |
| KR101807853B1 (en) | Method for individualized recommanding food for glycosuria patient | |
| CN115985470A (en) | Intelligent nutrition management method and intelligent management system | |
| CN118069931B (en) | Intelligent diet recommendation method for diabetics based on artificial intelligence | |
| US20230162859A1 (en) | Improved method for determining blood glucose responses | |
| Huo et al. | Predicting the meal macronutrient composition from continuous glucose monitors | |
| WO2022050992A1 (en) | Virtually monitoring glucose levels in a patient using machine learning and digital twin technology | |
| US20230255556A1 (en) | Biosensors and food logger systems for personalized health and self-care | |
| CN111489806A (en) | Intelligent diabetes heat management method and system | |
| KR102546256B1 (en) | A deep learning based health care service platform by vitro diagnosis and the recommendation method for health care using the same | |
| US20230274812A1 (en) | Methods and systems for calculating an edible score in a display interface |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |