CN110491475A - A kind of menu recommendation process method and device - Google Patents
A kind of menu recommendation process method and device Download PDFInfo
- Publication number
- CN110491475A CN110491475A CN201910770773.0A CN201910770773A CN110491475A CN 110491475 A CN110491475 A CN 110491475A CN 201910770773 A CN201910770773 A CN 201910770773A CN 110491475 A CN110491475 A CN 110491475A
- Authority
- CN
- China
- Prior art keywords
- user
- menu
- measurement parameter
- target collection
- sample
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- 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)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Epidemiology (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Nutrition Science (AREA)
- Probability & Statistics with Applications (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The embodiment of the invention discloses a kind of menu recommendation process method and device, method includes: acquisition environmental information, according to the user characteristics of the environmental information and target collection, generates menu corresponding with the user characteristics of the target collection；The menu is recommended to the user terminal of the target collection.The embodiment of the present invention generates menu corresponding with the user characteristics of target collection by combining environmental information, recommends user terminal, can not only customize the specific menu for meeting trophic level, and can carry out personalized menu for the different demands of specific crowd and recommend.
Description
Technical field
The present invention relates to field of computer technology, and in particular to a kind of menu recommendation process method and device.
Background technique
With the improvement of people's living standards, the requirement to nutrient diet is higher and higher, occur many dietary therapies of Chinese medicine with
Nutrition catering system, these systems combine modern diet nutritional theory and the dietary therapy of Chinese medicine is theoretical, and are joined according to the body of user
Number, judgement need the information such as the properties of foods, nutritional ingredient and effect of selection, the food for preventing and treating corresponding disease are provided for user
Spectrum scheme.
Intelligence nutritious recipe collocation system in the prior art is by family information unit, healthy assistant's unit and nutrition
Recording unit connection, and by family's nutritional need data, click data, body index data, trophic analysis data and nutrition
Data of arranging in pairs or groups combine, and constitute daily family's recipe；There are also food materials recipe, health to identify the matching analysis system, and guidance user understands
The health status of oneself, the food materials recipe information for reminding user to be needed to register according to the health status of itself, is scanned by mobile phone
Mode, obtain food materials recipe information, analyze food materials ingredient composition；And self-service dietotherapy service system, Ke Yigen
According to the constitution and conditions of different individuals, corresponding dietotherapy menu, etc. is produced.
But the prior art can not provide nutrient formulation that is specific, meeting user demand, Wu Fashi for different users
The personalized recommendation of menu is now carried out for specific crowd.
Summary of the invention
Since existing method is there are the above problem, the embodiment of the present invention proposes a kind of menu recommendation process method and device.
In a first aspect, the embodiment of the present invention proposes a kind of menu recommendation process method, comprising:
Environmental information is obtained, according to the user characteristics of the environmental information and target collection, is generated and the target collection
The corresponding menu of user characteristics；
The menu is recommended to the user terminal of the target collection.
Optionally, the acquisition environmental information, according to the user characteristics of the environmental information and target collection, generation and institute
Before the corresponding menu of user characteristics for stating target collection, further includes:
The measurement parameter for obtaining reflection user's body state, analyzes user according to the measurement parameter, obtains institute
State target collection.
Optionally, the measurement parameter for obtaining reflection user's body state carries out user according to the measurement parameter
Analysis, obtains the target collection, specifically includes:
The measurement parameter for obtaining reflection user's body state is based on density using representational according to the measurement parameter
Clustering algorithm to user carry out clustering, according to maximal density derived from density reachability relation be connected sample set, obtain
To target collection.
Optionally, the measurement parameter for obtaining reflection user's body state carries out user according to the measurement parameter
Analysis, obtains the target collection, specifically includes:
The measurement parameter of user is obtained as sample, if judgement sample quantity is greater than preset value, according to K space tree or ball
Tree search arest neighbors determines the sample distance in clustering algorithm using Euclidean distance according to the arest neighbors, and according to the sample
This distance and the measurement parameter carry out clustering to user, obtain target collection.
Optionally, the measurement parameter includes user's figure, user's constitution and user health situation.
Second aspect, the embodiment of the present invention also propose a kind of menu recommendation process device, comprising:
Menu generation module, it is raw according to the user characteristics of the environmental information and target collection for obtaining environmental information
At menu corresponding with the user characteristics of the target collection；
Menu recommending module, for the menu to be recommended to the user terminal of the target collection.
Optionally, the menu recommendation process device further include:
Cluster Analysis module, for obtain reflection user's body state measurement parameter, according to the measurement parameter to
Family is analyzed, and target collection is obtained.
Optionally, the Cluster Analysis module is specifically used for obtaining the measurement parameter of reflection user's body state, according to institute
It states measurement parameter and clustering is carried out to user using representational density-based algorithms, according to density reachability relation
The connected sample set of derived maximal density, obtains target collection.
Optionally, the Cluster Analysis module is specifically used for obtaining the measurement parameter of user as sample, if judgement sample
Quantity is greater than preset value, then searches for arest neighbors according to K space tree or ball tree, is determined and is gathered using Euclidean distance according to the arest neighbors
Sample distance in class algorithm, and clustering is carried out to user according to the sample distance and the measurement parameter, obtain mesh
Mark set.
Optionally, the measurement parameter includes user's figure, user's constitution and user health situation.
The third aspect, the embodiment of the present invention also propose a kind of electronic equipment, comprising:
At least one processor；And
At least one processor being connect with the processor communication, in which:
The memory is stored with the program instruction that can be executed by the processor, and the processor calls described program to refer to
Order is able to carry out the above method.
Fourth aspect, the embodiment of the present invention also propose a kind of non-transient computer readable storage medium, the non-transient meter
Calculation machine readable storage medium storing program for executing stores computer program, and the computer program makes the computer execute the above method.
As shown from the above technical solution, the embodiment of the present invention is generated special with the user of target collection by combining environmental information
Corresponding menu is levied, user terminal is recommended, the specific menu for meeting trophic level can not only be customized, and can be for specific
The different demands of crowd carry out personalized menu and recommend.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these figures.
Fig. 1 is a kind of flow diagram for menu recommendation process method that one embodiment of the invention provides；
Fig. 2 is a kind of schematic diagram for sample clustering process that one embodiment of the invention provides；
Fig. 3 be another embodiment of the present invention provides a kind of menu recommendation process method flow diagram；
Fig. 4 is a kind of structural schematic diagram for menu recommendation process device that one embodiment of the invention provides；
Fig. 5 is the logic diagram for the electronic equipment that one embodiment of the invention provides.
Specific embodiment
With reference to the accompanying drawing, further description of the specific embodiments of the present invention.Following embodiment is only used for more
Technical solution of the present invention is clearly demonstrated, and not intended to limit the protection scope of the present invention.
Fig. 1 shows a kind of flow diagram of menu recommendation process method provided in this embodiment, comprising:
S101, environmental information is obtained, according to the user characteristics of the environmental information and target collection, generated and the target
The corresponding menu of the user characteristics of set.
Wherein, the environmental information includes the information such as current weather, time and region.
The target collection is after being analyzed according to measurement parameter user, and what is obtained meets the collection of expected user
It closes, such as the set of people with hyperlipidemia, the set of Hypertensive Population, or, the set of pregnant woman.
The user characteristics of the target collection are the correlated characteristic for the abnormalities of sugar/lipid metabolism that the user in target collection shares.
Specifically, it is carried out according to the measurement parameter of user's figure, user's constitution and user health situation these users itself
Analysis, obtains the set of a people with hyperlipidemia；When customizing menu for people with hyperlipidemia, in conjunction with current weather, time
With the feature of the environmental informations such as region and people with hyperlipidemia abnormalities of sugar/lipid metabolism, the customization menu of people with hyperlipidemia is generated.
S102, the user terminal that the menu is recommended to the target collection.
For example, people with hyperlipidemia, Hypertensive Population or the constitution of pregnant woman multiplicity in Chinese medicine, symptom is different, the cause of disease
It is complicated.Different symptom types, corresponding dietotherapy effect is different, if can not be using the theory of Chinese medicine to sick people's
Menu is recommended in disease analysis, and special pantry service can not be just provided for above-mentioned specific crowd.The health prison that the present embodiment uses
Measurement equipment is that body fat claims and sphygmomanometer, body fat claim to can measure the basic body datas such as user's height, body fat, body fat rate, muscle mass,
Sphygmomanometer can measure user's correlation blood pressure data, and dynamically upload data information to healthy cloud platform.
When recommending to customize menu, need to follow healthy rule.Healthy rule refers to for analyzing user information and sentencing
The rule of disconnected user's figure, user's constitution and user health situation.
Specifically, by taking people with hyperlipidemia as an example, when progress user's figure judges, according to personal considerations' (input of user
Information), user is divided to corresponding crowd's label (hyperlipidemia population), matches corresponding effect (prevention and treatment hyperlipidemia etc.) guidance
Scientific nutrition pantry；According to the user information of collection, information is standardized, comprising: information names fuzzy matching, list
Position conversion, information secondary treatment；And BMI (Body Mass Index, body-mass index) value: calculation formula is calculated by formula
For BMI=weight (kg)/height (m)2, as a result generally retain 2 decimals.
When carrying out user health situation and judge, hyperlipidemia is directly made a definite diagnosis by biochemical reflect otherwise, and by information
Typing cloud platform；And by establishing hyperlipidemia health Prediction Model, judge user whether the risk of patient's hyperlipidemia.
By the study found that hyperlipidemia, vascular hypertension or pregnant woman's reaction and the gender of user, the age, BMI, body fat rate,
The relevant property of muscle rate, basal metabolic rate, fat percentage.By establishing multiple parameters equation, by figure (BMI value), the age,
Gender (male is 1, female 2), metabolic disease (quantity) information, indirectly judge whether user suffers from hyperlipidemia or hypertension
Disease, or whether be pregnant woman.For data, by taking hyperlipemia as an example, figure (BMI value) (X1), age (X2), gender are found
(X3), metabolic disease quantity (X4), body fat rate (X5), muscle rate, basal metabolic rate (X6) and fat percentage (X7) and hyperlipemia
Disease (Y) is in highly relevant.
The present embodiment generates menu corresponding with the user characteristics of target collection by combining environmental information, recommends user
Terminal can not only customize the specific menu for meeting trophic level, and can carry out individual character for the different demands of specific crowd
Change menu to recommend.
Further, on the basis of above method embodiment, before S101, further includes:
S100, the measurement parameter for obtaining reflection user's body state, analyze user according to the measurement parameter, obtain
To target collection.
Wherein, the measurement parameter includes user's figure, user's constitution and user health situation.
Specifically, by the analysis to measurement parameter, user can be divided into different crowds, every kind of crowd corresponding one
A set facilitates subsequent being customized of the crowd ground menu in each set to recommend.
Further, on the basis of above method embodiment, S100 is specifically included:
The measurement parameter for obtaining reflection user's body state is based on density using representational according to the measurement parameter
Clustering algorithm to user carry out clustering, according to maximal density derived from density reachability relation be connected sample set, obtain
To target collection.
Wherein, representational density-based algorithms (DBSCAN, the Density-Based Spatial
Clustering of Applications with Noise) it is different from division and hierarchy clustering method, cluster is defined as close by it
The maximum set of the connected point of degree can be cluster having region division highdensity enough, and can be in the spatial data of noise
The cluster of arbitrary shape is found in library.DBSCAN describes the tightness degree of sample set based on one group of neighborhood, reaction all ages and classes,
The degree of correlation of the dimensions such as body fat rate and hyperlipidemia morbidity, the sample distribution that parameter (∈, MinPts) is used to describe neighborhood are close
Degree, such as whether have apparent aggregation in the hyperlipidemia population that specified dimension influences (aggregation is higher, it was demonstrated that parameter
Correlation is bigger).Wherein, ∈ describes the neighborhood distance threshold of a certain sample, and the distance that MinPts describes a certain sample is
The threshold value of number of samples in the neighborhood of ∈.
As an example it is assumed that sample set is D=(x1,x2,...,xm) (it is equal to body mass index (BMI), body fat rate, muscle
Rate, basal metabolic rate, fat percentage), then the specific density description of DBSCAN is defined as follows:
∈-neighborhood: for xj∈ D, ∈-neighborhood include sample set D in xjDistance be not more than ∈ subsample collection,
As crowd sample of the BMI between 30 and 50 be represented by D comprising BMI dimension N ∈ (D)=BMI ∈ D | distance
(30.0,50.0)≤∈ }, the number of this subsample collection is denoted as | N ∈ (BMI) |.
Kernel object: for any sample xj(BMI) ∈ D, if its ∈-neighbor assignment N ∈ (BMI) is included at least
MinPts sample, i.e., if | N ∈ (BMI) | >=MinPts, BMI are kernel objects.
Density is through: if the data X of single user is located at x in samplej(BMI) in ∈-neighborhood, and xjIt (BMI) is core
Object then claims X by xj(BMI) density is through.Otherwise it not necessarily sets up, cannot say x at this timej(BMI) it is gone directly by X density, unless and
X is also kernel object.
Density is reachable: for X and xj(BMI), if there is sample sequence p1,p2,...,pT, meet p1=X, pT=xj
And p (BMI),t+1By ptDensity is through, then claims xj(BMI) reachable by X density.That is, density is reachable to meet transitivity.This
Transfer samples p in time series1,p2,...,pT-1It is kernel object, because only that kernel object can just make other sample rates
It is through.Up to symmetry is also unsatisfactory for, this asymmetry that can be gone directly by density obtains density.
Density is connected: for X and xj(BMI), if there is kernel object sample xk(BMI), make X and xj(BMI) by xk
(BMI) density is reachable, then claims X and xj(BMI) density is connected.Density associated relation meets symmetry.
Above-mentioned algorithm is illustrated for intuitive example, as shown in Fig. 2, the point that MinPts=5 arrow connects in figure is all
Kernel object, because of the neighborhood of its ∈-at least 5 samples.Point in addition to the point of arrow connection is non-kernel object.It is all
Kernel object density goes directly sample in the sphere centered on red kernel object, and it is straight to be unable to density if not in sphere
It reaches.The reachable sample sequence of density is constituted with the kernel object that arrow has connected in figure.In the reachable sample sequence of these density
∈-neighborhood in all sample be mutually all that density is connected.
Specifically, the cluster of DBSCAN is very simple: the connected sample set of the maximal density as derived from density reachability relation,
The classification as finally clustered, in other words a cluster.There can be one or more core inside the cluster of this DBSCAN
Object.If only one kernel object, other non-core object samples are all in ∈-neighborhood of this kernel object cluster
In；If there is multiple cores object, then centainly there is an other core in ∈-neighborhood of any one kernel object in cluster
Object, otherwise the two kernel objects can not density it is reachable.The collection of all samples is combined in ∈-neighborhood of these kernel objects
At a DBSCAN clustering cluster.
In cluster process, arbitrarily selects the core health dimension of a not no classification as seed, then find all
This core health dimension can the reachable sample set of density, (crowd of a group specific dimension is poly- for as clustering cluster
Collection)；It then continues to that another is selected not have the kernel object of classification to look for the reachable sample set of density, thus obtains another
One clustering cluster；The core customer area that institute's unsoundness dimension all finds oneself is run to always, then in other words remote dimension
The sample point being free on outside cluster on a small quantity is labeled as noise point.
Concrete application of the DBSCAN clustering algorithm in healthy dimension weight is as follows:
Input: sample set D=(body mass index (BMI), body fat rate, muscle rate, basal metabolic rate, fat percentage ..., moisture
Rate), Neighbourhood parameter (∈, MinPts), sample distance metric mode.
Output: cluster divides C and its density (weight that can be converted into healthy dimension).
The present embodiment introduces DBSCAN algorithm according to existing user health big data, according to different parameters to different close
The influence of degree user's sample size obtains the accurate weight of parameter, proceeds from reality in everything we do.
DBSCAN clustering algorithm in actual application, including step in detailed below:
A1, initialization kernel object setInitialization cluster number of clusters k=0, initializes non-access-sample set Γ
=D, cluster divide
A2, for j=1,2 ... m is found out all kernel objects by following step:
By distance metric mode, ∈-neighborhood subsample collection N ∈ (BMI) of sample xj (BMI) is found；Or, if increment
This collection number of samples meets | N ∈ (BMI) | kernel object sample set: Ω=Ω ∪ is added in sample B MI by >=MinPts
{BMI}；
If A3, kernel object setThen algorithm terminates, and is otherwise transferred to step A4；
A4, in kernel object set omega, randomly choose next kernel object o (muscle rate), initialize current cluster core
Heart object queue Ω cur={ o (muscle rate) } initializes classification sequence number k=k+1, initializes current cluster sample set Ck={ o
(muscle rate) }, update non-access-sample set Γ=Γ-{ o (muscle rate) }；
If A5, current cluster kernel object queueThen current clustering cluster Ck generation finishes, and updates cluster and divides C
=BMI, muscle rate ..., kernel object set omega=Ω-Ck is updated, step A3 is transferred to；
A6, a kernel object o ' is taken out in current cluster kernel object queue Ω cur, looked for by neighborhood distance threshold ∈
All ∈-neighborhood subsample collection N ∈ (o ') out enables Δ=N ∈ (o ') ∩ Γ, updates current cluster sample set Ck=Ck ∪
Δ updates non-access-sample set Γ=Γ-Δ, updates Ω cur=Ω cur ∪ (Δ ∩ Ω)-o ', is transferred to step A5.
Iteration carries out Density Distribution calculating, until exporting final result are as follows: cluster divides C={ body fat rate, age, basic generation
Thank rate, body mass index (BMI), moisture rate ... } in 350,000 samples, the dnesity index of each core cluster respectively only takes preceding 7
[4.4429,0.4270,0.4289,0.3634,0.3267,0.2097,0.0343]
So far, regression equation is substituted into are as follows: y ∧=4.4429X1+0.4270X2+0.4289X3+0.3634X4+0.3267X5
+ 0.2097X6+0.0343X7, residual variance return standard deviation Sy=0.38, the results of analysis of variance, F value=88.65 > F0.01
(5.55), P < 0.01 illustrates that the effect of regression equation is preferable.But it since 7 yuan of regressor parameters are excessive, uses still just
It is pretty troublesome, to solve this deficiency, each influence factor and hyperlipidemia illness rate are compared.Pass through partial correlation coefficient
(Ri) and partial regression coefficient (Pi) factor analysis that carries out show it is high with hyperlipidemia degree of correlation, and to estimation body fat rate shadow
Ringing big mainly has three body fat rate, age, basal metabolic rate elements.
By taking the corresponding hyperlipidemia of people with hyperlipidemia as an example, Hyperlipidemia (Y) and body fat rate (X are rebuild1), age (X2), base
Plinth metabolic rate (X3) matrix model, the specific method is as follows:
B1, building decision factor set U={ age, BMI, body fat rate, basal metabolic rate }；
Set V={ hyperlipidemia, non-hyperlipidemia } is judged in B2, building；
B3, simple element evaluation collection Ri={ R is calculatedi1,Ri2}；
B4, construction jdgement matrix, are unfolded as follows Ri, that is, obtain judging set.
Hyperlipidemia R1=R1_u1v1, R1_u2v2, R1_u3v1, R1_u4v1, R1_u3v2, R1_u2v1, R1_u1v2,
R1_u4v2}；
Non- hyperlipidemia R2={ R2_u1v1, R2_u2v2, R2_u3v1, R2_u4v1, R2_u3v2, R2_u2v1, R2_
u1v2,R1_u4v2}；
B5, comprehensive weight judgement is carried out, to weight E=(e1, e2, e3, e4 ...), calculates B=AE；
B6, result is normalized, maximum value is the ownership of Diseases diagnosis after normalization；
B7, verifying: by 2300 number of cases in Health database based on, 2100 is randomly selected and uses above-mentioned fuzzy number
It learns and statistical method establishes model, using remaining 200 number of cases according to as test sample, the hyperlipidemia for testing each user respectively is sentenced
Fixed accuracy, test result show accuracy rate up to 92.1%.
Specifically, the decision procedure phase of the decision procedure of the corresponding vascular hypertension of Hypertensive Population and above-mentioned hyperlipemia
Seemingly, i.e., decision factor set U={ age, BMI, body fat rate, basal metabolic rate } is constructed first, then set V=is judged in building
{ vascular hypertension, non-hypertensive disease }, subsequent implementation procedure is referring to above-mentioned B3-B7 step.Similarly, the decision procedure of pregnant woman's reaction
Also similar to the decision procedure of above-mentioned hyperlipemia.
By taking hyperlipemia as an example, detailed process such as Fig. 3 institute of the menu recommendation process method of corresponding people with hyperlipidemia
Show (vascular hypertension is similar with pregnant woman's reaction), input terminal is collected and user information data, and generates user's body parameter.Cloud platform
The environmental informations such as weather, time and region when energy intelligent recognition user is using system.In addition, the collection of human body master data can
It, can also be by intelligent appliance automatic feedback user information by user's voluntarily typing (e.g., information such as height, weight, age).Processing end
By hyperlipidemia healthy early warning model, determine whether user suffers from hyperlipidemia risk；To hyperlipidemia patient and hyperlipemia
Disease risk subscribers carry out dietary management；In conjunction with tcm theory, medicine typing is carried out to user, by effect algorithm, determines food
Material type；In conjunction with modern nutriology disease trophic level, solves food materials and quantify problem.It is defeated by the data processing for introducing DBSCAN
Outlet exports customization menu.
Further, on the basis of above method embodiment, S100 is specifically included:
The measurement parameter of user is obtained as sample, if judgement sample quantity is greater than preset value, according to K space tree or ball
Tree search arest neighbors determines the sample distance in clustering algorithm using Euclidean distance according to the arest neighbors, and according to the sample
This distance and the measurement parameter carry out clustering to user, obtain target collection.
Specifically, when solving the problems, such as DBSCAN distance metric, according to the sample of user's big data, using arest neighbors thought
(because crowd health parameters will not difference it is very big, always in certain index), weighed using a certain distance metric
Measure sample distance, such as Euclidean distance.This is identical with the arest neighbors thought of KNN sorting algorithm.
In general, corresponding a small amount of sample, finding arest neighbors can directly go to calculate the distance of all samples；If sample
This amount is larger, then general to rapidly search for arest neighbors using the space K (KD) tree or ball tree.
Fig. 4 shows a kind of structural schematic diagram of menu recommendation process device provided in this embodiment, and described device includes:
Menu generation module 401 and menu recommending module 402, in which:
The menu generation module 401 is for obtaining environmental information, according to the user of the environmental information and target collection
Feature generates menu corresponding with the user characteristics of the target collection；
The menu recommending module 402 is used to recommend the menu user terminal of the target collection.
Specifically, the menu generation module 401 obtains environmental information, according to the use of the environmental information and target collection
Family feature generates menu corresponding with the user characteristics of the target collection；The menu recommending module 402 pushes away the menu
It recommends to the user terminal of the target collection.
The present embodiment generates menu corresponding with the user characteristics of target collection by combining environmental information, recommends user
Terminal can not only customize the specific menu for meeting trophic level, and can carry out individual character for the different demands of specific crowd
Change menu to recommend.
Further, on the basis of above-mentioned apparatus embodiment, the menu recommendation process device further include:
Cluster Analysis module, for obtain reflection user's body state measurement parameter, according to the measurement parameter to
Family is analyzed, and target collection is obtained；
Further, on the basis of above-mentioned apparatus embodiment, the Cluster Analysis module is specifically used for obtaining reflection use
The measurement parameter of family physical condition, according to the measurement parameter using representational density-based algorithms to user into
Row clustering obtains target collection according to the sample set that maximal density derived from density reachability relation is connected.
Further, on the basis of above-mentioned apparatus embodiment, the Cluster Analysis module is specifically used for obtaining user's
Measurement parameter is as sample, if judgement sample quantity is greater than preset value, searches for arest neighbors according to K space tree or ball tree, according to
The arest neighbors determines the sample distance in clustering algorithm using Euclidean distance, and is joined according to the sample distance and the measurement
Several couples of users carry out clustering, obtain target collection.
Further, on the basis of above-mentioned apparatus embodiment, the measurement parameter include user's figure, user's constitution and
User health situation.
Menu recommendation process device described in the present embodiment can be used for executing above method embodiment, principle and technology
Effect is similar, and details are not described herein again.
Referring to Fig. 5, the electronic equipment, comprising: processor (processor) 501, memory (memory) 502 and total
Line 503；
Wherein,
The processor 501 and memory 502 complete mutual communication by the bus 503；
The processor 501 is used to call the program instruction in the memory 502, to execute above-mentioned each method embodiment
Provided method.
The present embodiment discloses a kind of computer program product, and the computer program product includes being stored in non-transient calculating
Computer program on machine readable storage medium storing program for executing, the computer program include program instruction, when described program instruction is calculated
When machine executes, computer is able to carry out method provided by above-mentioned each method embodiment.
The present embodiment provides a kind of non-transient computer readable storage medium, the non-transient computer readable storage medium
Computer instruction is stored, the computer instruction makes the computer execute method provided by above-mentioned each method embodiment.
The apparatus embodiments described above are merely exemplary, wherein described, unit can as illustrated by the separation member
It is physically separated with being or may not be, component shown as a unit may or may not be physics list
Member, it can it is in one place, or may be distributed over multiple network units.It can be selected according to the actual needs
In some or all of the modules achieve the purpose of the solution of this embodiment.Those of ordinary skill in the art are not paying creativeness
Labour in the case where, it can understand and implement.
Through the above description of the embodiments, those skilled in the art can be understood that each embodiment can
It realizes by means of software and necessary general hardware platform, naturally it is also possible to pass through hardware.Based on this understanding, on
Stating technical solution, substantially the part that contributes to existing technology can be embodied in the form of software products in other words, should
Computer software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including several fingers
It enables and using so that a computer equipment (can be personal computer, server or the network equipment etc.) executes each implementation
Method described in certain parts of example or embodiment.
It is noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although reference
Invention is explained in detail for previous embodiment, those skilled in the art should understand that: it still can be right
Technical solution documented by foregoing embodiments is modified or equivalent replacement of some of the technical features；And this
It modifies or replaces, the spirit and model of technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution
It encloses.
Claims (12)
1. a kind of menu recommendation process method characterized by comprising
Environmental information is obtained, according to the user characteristics of the environmental information and target collection, generates the use with the target collection
The corresponding menu of family feature；
The menu is recommended to the user terminal of the target collection.
2. menu recommendation process method according to claim 1, which is characterized in that the acquisition environmental information, according to institute
The user characteristics of environmental information and target collection are stated, before generating menu corresponding with the user characteristics of the target collection, also
Include:
The measurement parameter for obtaining reflection user's body state, analyzes user according to the measurement parameter, obtains the mesh
Mark set.
3. menu recommendation process method according to claim 2, which is characterized in that the acquisition reflects user's body state
Measurement parameter, user is analyzed according to the measurement parameter, the target collection is obtained, specifically includes:
The measurement parameter for obtaining reflection user's body state, according to the measurement parameter using representational poly- based on density
Class algorithm carries out clustering to user, according to the sample set that maximal density derived from density reachability relation is connected, obtains mesh
Mark set.
4. menu recommendation process method according to claim 2, which is characterized in that the acquisition reflects user's body state
Measurement parameter, user is analyzed according to the measurement parameter, the target collection is obtained, specifically includes:
The measurement parameter for obtaining user is searched if judgement sample quantity is greater than preset value according to K space tree or ball tree as sample
Rope arest neighbors determines the sample distance in clustering algorithm using Euclidean distance according to the arest neighbors, and according to the sample away from
Clustering is carried out to user from the measurement parameter, obtains target collection.
5. according to the described in any item menu recommendation process methods of claim 2-4, which is characterized in that the measurement parameter includes
User's figure, user's constitution and user health situation.
6. a kind of menu recommendation process device characterized by comprising
Menu generation module, for obtaining environmental information, according to the user characteristics of the environmental information and target collection, generate with
The corresponding menu of the user characteristics of the target collection；
Menu recommending module, for the menu to be recommended to the user terminal of the target collection.
7. menu recommendation process device according to claim 6, which is characterized in that the menu recommendation process device also wraps
It includes:
Cluster Analysis module, for obtain reflection user's body state measurement parameter, according to the measurement parameter to user into
Row analysis, obtains target collection.
8. menu recommendation process device according to claim 7, which is characterized in that the Cluster Analysis module is specifically used for
The measurement parameter for obtaining reflection user's body state is calculated according to the measurement parameter using representational density clustering
Method carries out clustering to user, according to the sample set that maximal density derived from density reachability relation is connected, obtains object set
It closes.
9. menu recommendation process device according to claim 7, which is characterized in that the Cluster Analysis module is specifically used for
The measurement parameter of user is obtained as sample, if judgement sample quantity is greater than preset value, is searched for most according to K space tree or ball tree
Neighbour determines the sample distance in clustering algorithm using Euclidean distance according to the arest neighbors, and according to the sample distance and
The measurement parameter carries out clustering to user, obtains target collection.
10. according to the described in any item menu recommendation process devices of claim 7-9, which is characterized in that the measurement parameter packet
Include user's figure, user's constitution and user health situation.
11. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and on a processor
Machine program, which is characterized in that the processor realizes that menu as described in any one in claim 1-5 pushes away when executing described program
Recommend processing method.
12. a kind of non-transient computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer
Menu recommendation process method as described in any one in claim 1-5 is realized when program is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910770773.0A CN110491475A (en) | 2019-08-20 | 2019-08-20 | A kind of menu recommendation process method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910770773.0A CN110491475A (en) | 2019-08-20 | 2019-08-20 | A kind of menu recommendation process method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110491475A true CN110491475A (en) | 2019-11-22 |
Family
ID=68552437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910770773.0A Pending CN110491475A (en) | 2019-08-20 | 2019-08-20 | A kind of menu recommendation process method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110491475A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111090812A (en) * | 2019-12-24 | 2020-05-01 | 珠海格力电器股份有限公司 | Menu recommendation method and device, electronic equipment and storage medium |
| CN111145027A (en) * | 2019-12-31 | 2020-05-12 | 众安信息技术服务有限公司 | Suspected money laundering transaction identification method and device |
| CN111276215A (en) * | 2020-01-21 | 2020-06-12 | 星络智能科技有限公司 | Menu recommendation method, intelligent home controller and storage medium |
| CN111524277A (en) * | 2020-04-16 | 2020-08-11 | 迪国生物高科技有限公司 | Intelligent health product selling cabinet and automatic product selling method and device thereof |
| CN113048529A (en) * | 2019-12-26 | 2021-06-29 | 佛山市云米电器科技有限公司 | Cooking information prompting method, range hood and computer readable storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609860A (en) * | 2012-01-20 | 2012-07-25 | 彭立发 | Method and system suitable for categorizing and recommending e-commerce commodities and information |
| CN103455699A (en) * | 2012-05-31 | 2013-12-18 | 杨玉峰 | Method and system for generating personalized lifestyle intervention scheme automatically |
| CN108806770A (en) * | 2018-05-16 | 2018-11-13 | 北京豆果信息技术有限公司 | A kind of diet recommendation method based on user |
-
2019
- 2019-08-20 CN CN201910770773.0A patent/CN110491475A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609860A (en) * | 2012-01-20 | 2012-07-25 | 彭立发 | Method and system suitable for categorizing and recommending e-commerce commodities and information |
| CN103455699A (en) * | 2012-05-31 | 2013-12-18 | 杨玉峰 | Method and system for generating personalized lifestyle intervention scheme automatically |
| CN108806770A (en) * | 2018-05-16 | 2018-11-13 | 北京豆果信息技术有限公司 | A kind of diet recommendation method based on user |
Non-Patent Citations (1)
| Title |
|---|
| 曹叔彦: "CLIQUE网格聚类算法在医学空间数据中的应用", 《中国优秀硕士学位论文全文数据库 医药卫生科技辑》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111090812A (en) * | 2019-12-24 | 2020-05-01 | 珠海格力电器股份有限公司 | Menu recommendation method and device, electronic equipment and storage medium |
| CN113048529A (en) * | 2019-12-26 | 2021-06-29 | 佛山市云米电器科技有限公司 | Cooking information prompting method, range hood and computer readable storage medium |
| CN111145027A (en) * | 2019-12-31 | 2020-05-12 | 众安信息技术服务有限公司 | Suspected money laundering transaction identification method and device |
| CN111276215A (en) * | 2020-01-21 | 2020-06-12 | 星络智能科技有限公司 | Menu recommendation method, intelligent home controller and storage medium |
| CN111524277A (en) * | 2020-04-16 | 2020-08-11 | 迪国生物高科技有限公司 | Intelligent health product selling cabinet and automatic product selling method and device thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110491475A (en) | A kind of menu recommendation process method and device | |
| Schwab et al. | Clinical predictive models for COVID-19: systematic study | |
| Yang et al. | Yum-me: a personalized nutrient-based meal recommender system | |
| Chen et al. | Eating healthier: Exploring nutrition information for healthier recipe recommendation | |
| Harper | A review and comparison of classification algorithms for medical decision making | |
| Nagaraj et al. | Artificial flora algorithm-based feature selection with gradient boosted tree model for diabetes classification | |
| Aguilar et al. | Regularized uncertainty-based multi-task learning model for food analysis | |
| WO2022133985A1 (en) | Food product recommendation method and apparatus, and storage medium and electronic device | |
| US20220230730A1 (en) | Recipe recommendation method and device, computing device and storage medium | |
| WO2021139116A9 (en) | Method, apparatus and device for intelligently grouping similar patients, and storage medium | |
| CN109817302B (en) | Expert system for fitness | |
| US20160117457A1 (en) | Method and apparatus for analyzing patient's constitutional peculiarity | |
| CN107146160A (en) | Customer insured's analysis on the health status method and server | |
| Esmaeily et al. | Comparing three data mining algorithms for identifying the associated risk factors of type 2 diabetes | |
| US20210057077A1 (en) | Systems and methods for arranging transport of adapted nutrimental artifacts with user-defined restriction requirements using artificial intelligence | |
| Moscato et al. | Community detection over feature-rich information networks: An eHealth case study | |
| US10832172B1 (en) | Systems and methods for arranging transport of adapted nutrimental artifacts with user-defined restriction requirements using artificial intelligence | |
| US11232259B1 (en) | Methods and systems for personal recipe generation | |
| US11687813B2 (en) | Systems and methods for ranking alimentary combinations using machine-learning | |
| US10910112B2 (en) | Apparatus for patient record identification | |
| CN113569140A (en) | Information recommendation method and device, electronic equipment and computer-readable storage medium | |
| Zhang | Innovative food recommendation systems: a machine learning approach | |
| Rahman et al. | Review on food recommender system for diabetes mellitus | |
| JP2017010438A (en) | Multidimensional correlation data extraction device and multidimensional correlation data extraction method | |
| Song et al. | Federated Logistic Regression for Reliable Prediction Models in Privacy-Preserving Healthcare Networks |
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 |