CN110021403A - Recommend method and apparatus, household electrical appliance and the machine readable storage medium of food materials - Google Patents
Recommend method and apparatus, household electrical appliance and the machine readable storage medium of food materials Download PDFInfo
- Publication number
- CN110021403A CN110021403A CN201711037394.8A CN201711037394A CN110021403A CN 110021403 A CN110021403 A CN 110021403A CN 201711037394 A CN201711037394 A CN 201711037394A CN 110021403 A CN110021403 A CN 110021403A
- Authority
- CN
- China
- Prior art keywords
- food materials
- user
- matrix
- preference
- similarity
- 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
- 235000013305 food Nutrition 0.000 title claims abstract description 756
- 239000000463 material Substances 0.000 title claims abstract description 756
- 238000000034 method Methods 0.000 title claims abstract description 52
- 239000011159 matrix material Substances 0.000 claims abstract description 194
- 230000006399 behavior Effects 0.000 claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 28
- 238000004891 communication Methods 0.000 claims abstract description 4
- 230000000050 nutritive effect Effects 0.000 claims description 27
- 230000003542 behavioural effect Effects 0.000 claims description 19
- 230000000694 effects Effects 0.000 claims description 17
- 239000000654 additive Substances 0.000 claims description 8
- 230000000996 additive effect Effects 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 6
- 235000016709 nutrition Nutrition 0.000 claims 2
- 230000035764 nutrition Effects 0.000 claims 2
- 235000013601 eggs Nutrition 0.000 description 39
- 239000000047 product Substances 0.000 description 38
- 235000009811 Momordica charantia Nutrition 0.000 description 24
- 235000009812 Momordica cochinchinensis Nutrition 0.000 description 24
- 240000001740 Momordica dioica Species 0.000 description 24
- 235000018365 Momordica dioica Nutrition 0.000 description 24
- 235000018290 Musa x paradisiaca Nutrition 0.000 description 15
- 235000013311 vegetables Nutrition 0.000 description 14
- 241000272525 Anas platyrhynchos Species 0.000 description 13
- 240000005561 Musa balbisiana Species 0.000 description 11
- 235000005254 Allium ampeloprasum Nutrition 0.000 description 10
- 240000006108 Allium ampeloprasum Species 0.000 description 10
- 230000036772 blood pressure Effects 0.000 description 10
- 201000006549 dyspepsia Diseases 0.000 description 10
- 230000036039 immunity Effects 0.000 description 10
- 239000006210 lotion Substances 0.000 description 10
- 241000238557 Decapoda Species 0.000 description 9
- 206010028980 Neoplasm Diseases 0.000 description 9
- 201000011510 cancer Diseases 0.000 description 9
- 239000008280 blood Substances 0.000 description 8
- 210000004369 blood Anatomy 0.000 description 8
- 210000004556 brain Anatomy 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 230000002218 hypoglycaemic effect Effects 0.000 description 5
- 241000234295 Musa Species 0.000 description 4
- 241000657513 Senna surattensis Species 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 4
- 235000013372 meat Nutrition 0.000 description 4
- 210000002784 stomach Anatomy 0.000 description 4
- 241000251468 Actinopterygii Species 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 235000013399 edible fruits Nutrition 0.000 description 3
- 150000002632 lipids Chemical class 0.000 description 3
- 235000018645 Allium odorum Nutrition 0.000 description 2
- 240000008654 Allium ramosum Species 0.000 description 2
- 241000530454 Litopenaeus schmitti Species 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 2
- UQMRAFJOBWOFNS-UHFFFAOYSA-N butyl 2-(2,4-dichlorophenoxy)acetate Chemical compound CCCCOC(=O)COC1=CC=C(Cl)C=C1Cl UQMRAFJOBWOFNS-UHFFFAOYSA-N 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004438 eyesight Effects 0.000 description 2
- 235000014571 nuts Nutrition 0.000 description 2
- 239000002689 soil Substances 0.000 description 2
- 235000001674 Agaricus brunnescens Nutrition 0.000 description 1
- 244000226021 Anacardium occidentale Species 0.000 description 1
- 241000272814 Anser sp. Species 0.000 description 1
- 241000512259 Ascophyllum nodosum Species 0.000 description 1
- 235000017166 Bambusa arundinacea Nutrition 0.000 description 1
- 235000017491 Bambusa tulda Nutrition 0.000 description 1
- 241001474374 Blennius Species 0.000 description 1
- 240000007124 Brassica oleracea Species 0.000 description 1
- 235000003899 Brassica oleracea var acephala Nutrition 0.000 description 1
- 235000011301 Brassica oleracea var capitata Nutrition 0.000 description 1
- 235000001169 Brassica oleracea var oleracea Nutrition 0.000 description 1
- 235000010149 Brassica rapa subsp chinensis Nutrition 0.000 description 1
- 235000000536 Brassica rapa subsp pekinensis Nutrition 0.000 description 1
- 241000499436 Brassica rapa subsp. pekinensis Species 0.000 description 1
- 241000219112 Cucumis Species 0.000 description 1
- 235000015510 Cucumis melo subsp melo Nutrition 0.000 description 1
- 240000008067 Cucumis sativus Species 0.000 description 1
- 235000010799 Cucumis sativus var sativus Nutrition 0.000 description 1
- 241000287828 Gallus gallus Species 0.000 description 1
- 244000017020 Ipomoea batatas Species 0.000 description 1
- 235000002678 Ipomoea batatas Nutrition 0.000 description 1
- 235000006439 Lemna minor Nutrition 0.000 description 1
- 244000242291 Lemna paucicostata Species 0.000 description 1
- 235000007688 Lycopersicon esculentum Nutrition 0.000 description 1
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 241000286209 Phasianidae Species 0.000 description 1
- 244000082204 Phyllostachys viridis Species 0.000 description 1
- 235000015334 Phyllostachys viridis Nutrition 0.000 description 1
- 235000008331 Pinus X rigitaeda Nutrition 0.000 description 1
- 235000011613 Pinus brutia Nutrition 0.000 description 1
- 241000018646 Pinus brutia Species 0.000 description 1
- 244000088415 Raphanus sativus Species 0.000 description 1
- 235000006140 Raphanus sativus var sativus Nutrition 0.000 description 1
- 240000003768 Solanum lycopersicum Species 0.000 description 1
- 244000061458 Solanum melongena Species 0.000 description 1
- 235000002597 Solanum melongena Nutrition 0.000 description 1
- 240000008042 Zea mays Species 0.000 description 1
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 1
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 1
- FJJCIZWZNKZHII-UHFFFAOYSA-N [4,6-bis(cyanoamino)-1,3,5-triazin-2-yl]cyanamide Chemical compound N#CNC1=NC(NC#N)=NC(NC#N)=N1 FJJCIZWZNKZHII-UHFFFAOYSA-N 0.000 description 1
- 230000001093 anti-cancer Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 239000011425 bamboo Substances 0.000 description 1
- 235000013527 bean curd Nutrition 0.000 description 1
- 235000015278 beef Nutrition 0.000 description 1
- 235000020226 cashew nut Nutrition 0.000 description 1
- 239000013065 commercial product Substances 0.000 description 1
- 235000005822 corn Nutrition 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000013364 duck meat Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 235000021374 legumes Nutrition 0.000 description 1
- 210000004185 liver Anatomy 0.000 description 1
- 235000013336 milk Nutrition 0.000 description 1
- 239000008267 milk Substances 0.000 description 1
- 210000004080 milk Anatomy 0.000 description 1
- 231100000614 poison Toxicity 0.000 description 1
- 235000015277 pork Nutrition 0.000 description 1
- 210000000582 semen Anatomy 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 210000003813 thumb Anatomy 0.000 description 1
- 239000003440 toxic substance Substances 0.000 description 1
- 235000013618 yogurt Nutrition 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
The embodiment of the present invention provides a kind of method and apparatus, household electrical appliance and machine readable storage medium for recommending food materials, belongs to household appliance technical field.This method comprises: receiving the commodity title of food materials to be recommended；Processing commodity title is to generate food materials portrait；It is drawn a portrait according to food materials and generates food materials eigenmatrix；Receive the user behaviors log about recipe of user；User's food materials preference portrait is generated according to user behaviors log and recipe；It is drawn a portrait according to user's food materials preference and generates user's food materials preference profiles matrix；The similarity of a food materials and user's food materials preference is calculated according to food materials eigenmatrix and user's food materials preference profiles matrix；And food materials information is exported according to similarity.The device includes；Communication module, processing module and output module.The household electrical appliance include above-mentioned apparatus.Instruction is stored on the machine readable storage medium, which is used for so that the machine above method.Whereby, the matching degree of recommended food materials and user preferences is improved.
Description
Technical field
The present invention relates to household appliance technical field, more particularly to a kind of method and apparatus for recommending food materials, household electrical appliance and
Machine readable storage medium.
Background technique
In the prior art, recommender system is based on directly on the similarity between user or between product mostly as user's progress
Products Show, this way of recommendation can make recommendation results generate very large deviation with user's actual preferences.
Summary of the invention
A kind of recommend the method and apparatus of food materials, household electrical appliance the purpose of the embodiment of the present invention is that providing and machine readable deposit
The matching degree for recommending food materials and user preferences can be improved in storage media.
To achieve the goals above, the one aspect of the embodiment of the present invention provides a kind of for recommending the side of food materials to user
Method, this method comprises: receiving the commodity title of food materials to be recommended；The commodity title is handled to generate food materials portrait；According to institute
It states food materials portrait and generates food materials eigenmatrix；Receive the user behaviors log about recipe of the user；According to the user behaviors log
User's food materials preference portrait is generated with the recipe；It is drawn a portrait according to user's food materials preference and generates user's food materials preference profiles square
Battle array, wherein the user tag one in food materials label and user's food materials preference profiles matrix in the food materials eigenmatrix
One is corresponding；A food materials and user's food materials preference are calculated according to the food materials eigenmatrix and user's food materials preference profiles matrix
Similarity；And according to the similarity export food materials information, wherein the food materials information include food materials title and/or with this
The corresponding commodity title of food materials title.
Optionally, this method further include: the output food materials label corresponding with the food materials title.
Optionally, the processing commodity title includes: to segment to the commodity title to generate food materials portrait；
The stop words for including in the word that filtering participle obtains；Food materials title is marked to filtered word；And the institute according to mark
It states food materials title and the food materials label corresponding with the food materials title generates food materials portrait.
Optionally, described drawn a portrait according to the user behaviors log and recipe generation user's food materials preference includes: to utilize layer
Fractional analysis determines the weight of the behavioral parameters in the user behaviors log；Count the number of the behavioral parameters；According to the row
The weight of the recipe is determined for the weight of parameter and the number；Corresponding to the recipe is determined according to the weight of the recipe
Food materials weight；Determine the food materials label corresponding to food materials corresponding to the recipe；According to corresponding to the recipe
The weights of food materials determine the weight of the food materials label；And the user tag is corresponding with the food materials label, according to
The food materials label, the weight of the food materials label and the user tag generate user's food materials preference portrait.
It optionally, include preset value food materials eigenmatrix for the food materials eigenmatrix of same food materials, it is described pre-
If the food materials label for including in the food materials eigenmatrix in value food materials eigenmatrix belongs to same class food materials label, institute
Stating user's food materials preference profiles matrix includes preset value user's food materials preference profiles matrix, and the preset value user food materials are inclined
The user tag for including in user's food materials preference profiles matrix in good eigenmatrix belongs to same class user tag, institute
The classification for stating food materials label is corresponding with the classification of the user tag；It is described according to the food materials eigenmatrix and the user
It includes: to calculate separately the described of a food materials to preset that food materials preference profiles matrix, which calculates a food materials and the similarity of user's food materials preference,
It is worth each food materials eigenmatrix in food materials eigenmatrix and user's food materials corresponding with each food materials eigenmatrix
The characteristic similarity of preference profiles matrix, wherein the correspondence of the food materials eigenmatrix and user's food materials preference profiles matrix
Relationship is determined according to the corresponding relationship of the classification of the food materials label and the classification of the user tag；And according to the feature
Similarity, feature weight corresponding with the characteristic similarity and the preset value calculate the similarity.
Optionally, the characteristic similarity is calculated using cosine similarity formula.
Optionally, the preset value food materials eigenmatrix includes food materials classification matrix, food materials heat matrix and food materials battalion
Effect matrix is supported, the preset value user food materials preference profiles matrix includes user's food materials classification preference matrix, user's food materials
Heat preference matrix and user health attribute matrix；The characteristic similarity includes that the food materials classification matrix and the user eat
The of the fisrt feature similarity of material classification preference matrix, the food materials heat matrix and user's food materials heat preference matrix
The third feature similarity of two characteristic similarities and the food materials nutritive effect matrix and the user health attribute matrix；It is described
The similarity packet is calculated according to the characteristic similarity, feature weight corresponding with the characteristic similarity and the preset value
It includes: by the fisrt feature similarity, the second feature similarity and the third feature similarity respectively multiplied by corresponding
The feature weight；The value linear, additive that will be obtained after multiplication；And the value after will add up obtains the similarity divided by 3.
The another aspect of the embodiment of the present invention provides a kind of device for user's recommendation food materials, which includes；It is logical
Believe module, is used for: receiving the commodity title of food materials to be recommended；And receive the user behaviors log about recipe of the user；With
And processing module, it is used for: handling the commodity title to generate food materials portrait；It is drawn a portrait according to the food materials and generates food materials feature square
Battle array；User's food materials preference portrait is generated according to the user behaviors log and the recipe；It is drawn a portrait and is given birth to according to user's food materials preference
At user's food materials preference profiles matrix, wherein the food materials label and user's food materials preference in the food materials eigenmatrix are special
The user tag levied in matrix corresponds；And according to the food materials eigenmatrix and user's food materials preference profiles matrix
Calculate the similarity of a food materials and user's food materials preference；And output module, for exporting food materials information according to the similarity,
Wherein the food materials information includes food materials title and/or the commodity title corresponding with the food materials title.
Optionally, the output module is also used to export the food materials label corresponding with the food materials title.
Optionally, it includes: to the commodity that the processing module, which handles the commodity title to generate the food materials portrait,
Title is segmented；The stop words for including in the word that filtering participle obtains；Food materials title is marked to filtered word；According to
The food materials title of mark and the food materials label corresponding with the food materials title generate food materials portrait.
Optionally, the processing module generates user's food materials preference portrait packet according to the user behaviors log and the recipe
It includes: determining the weight of the behavioral parameters in the user behaviors log using analytic hierarchy process (AHP)；Count the number of the behavioral parameters；Root
The weight of the recipe is determined according to the weight and the number of the behavioral parameters；The food is determined according to the weight of the recipe
Compose the weight of corresponding food materials；Determine the food materials label corresponding to food materials corresponding to the recipe；According to the recipe institute
The weight of corresponding food materials determines the weight of the food materials label；And the user tag is corresponding with the food materials label,
User's food materials preference portrait is generated according to the food materials label, the weight of the food materials label and the user tag.
It optionally, include preset value food materials eigenmatrix for the food materials eigenmatrix of same food materials, it is described pre-
If the food materials label for including in the food materials eigenmatrix in value food materials eigenmatrix belongs to same class food materials label, institute
Stating user's food materials preference profiles matrix includes preset value user's food materials preference profiles matrix, and the preset value user food materials are inclined
The user tag for including in user's food materials preference profiles matrix in good eigenmatrix belongs to same class user tag, institute
The classification for stating food materials label is corresponding with the classification of the user tag；The processing module according to the food materials eigenmatrix and
It includes: to calculate separately a food materials that user's food materials preference profiles matrix, which calculates a food materials and the similarity of user's food materials preference,
Each food materials eigenmatrix in the preset value food materials eigenmatrix and corresponding with each food materials eigenmatrix described
The characteristic similarity of user's food materials preference profiles matrix, wherein the food materials eigenmatrix and user's food materials preference profiles square
The corresponding relationship of battle array is determined according to the corresponding relationship of the classification of the food materials label and the classification of the user tag；And according to
The characteristic similarity, feature weight corresponding with the characteristic similarity and the preset value calculate the similarity.
Optionally, the processing module calculates the characteristic similarity using cosine similarity formula.
Optionally, the preset value food materials eigenmatrix includes food materials classification matrix, food materials heat matrix and food materials battalion
Effect matrix is supported, the preset value user food materials preference profiles matrix includes user's food materials classification preference matrix, user's food materials
Heat preference matrix and user health attribute matrix；The characteristic similarity includes that the food materials classification matrix and the user eat
The of the fisrt feature similarity of material classification preference matrix, the food materials heat matrix and user's food materials heat preference matrix
The third feature similarity of two characteristic similarities and the food materials nutritive effect matrix and the user health attribute matrix；It is described
Processing module is according to the calculating of the characteristic similarity, feature weight corresponding with the characteristic similarity and the preset value
Similarity includes: to multiply the fisrt feature similarity, the second feature similarity and the third feature similarity respectively
With the corresponding feature weight；The value linear, additive that will be obtained after multiplication；And the value after will add up obtained divided by 3 it is described
Similarity.
The another aspect of the embodiment of the present invention provides a kind of household electrical appliance, which includes above-mentioned device.
Optionally, the household electrical appliance are refrigerator.
The another aspect of the embodiment of the present invention provides a kind of machine readable storage medium, deposits on the machine readable storage medium
Instruction is contained, which is used for so that machine executes above-mentioned method.
Through the above technical solutions, handling the commodity title of food materials to be recommended to generate food materials eigenmatrix, user is handled
It is special according to food materials eigenmatrix and user's food materials preference about the user behaviors log of recipe to generate user's food materials preference profiles matrix
The similarity that matrix calculates a food materials and user's food materials preference is levied, food materials are recommended to user according to the similarity.In this way, analysis is used
Family obtains the food materials preference of user about the user behaviors log of recipe, inclined according to food materials to be recommended and the user's food materials analyzed
Good similarity recommends food materials to user, improves the matching degree of recommended food materials and user preferences.
The other feature and advantage of the embodiment of the present invention will the following detailed description will be given in the detailed implementation section.
Detailed description of the invention
Attached drawing is to further understand for providing to the embodiment of the present invention, and constitute part of specification, under
The specific embodiment in face is used to explain the present invention embodiment together, but does not constitute the limitation to the embodiment of the present invention.Attached
In figure:
Fig. 1 is the flow chart for the method for recommending food materials to user that one embodiment of the invention provides；
Fig. 2 be another embodiment of the present invention provides for user recommend food materials method flow chart；
Fig. 3 be another embodiment of the present invention provides generate food materials portrait flow chart；
Fig. 4 be another embodiment of the present invention provides generate user's food materials preference portrait flow chart；And
Fig. 5 be another embodiment of the present invention provides for user recommend food materials device structural block diagram.
Description of symbols
1 communication module, 2 processing module
3 output modules
Specific embodiment
It is described in detail below in conjunction with specific embodiment of the attached drawing to the embodiment of the present invention.It should be understood that this
Locate described specific embodiment and be merely to illustrate and explain the present invention embodiment, is not intended to restrict the invention embodiment.
The one aspect of the embodiment of the present invention provides a kind of for recommending the method for food materials to user.Fig. 1 is the present invention one
The flow chart for the method for recommending food materials to user that embodiment provides.As shown in Figure 1, this approach includes the following steps.
In step slo, the commodity title of food materials to be recommended is received.In step s 11, processing commodity title is to generate food
Material portrait.It include commodity title, food materials title and food materials label in food materials portrait.In step s 12, it is drawn a portrait and is given birth to according to food materials
At food materials eigenmatrix.In step s 13, the user behaviors log about recipe of user is received.In step S14, according to behavior
Log and recipe generate user's food materials preference portrait.In step S15, being drawn a portrait according to user's food materials preference, it is inclined to generate user's food materials
Good eigenmatrix, wherein the user tag one in food materials label and user's food materials preference profiles matrix in food materials eigenmatrix
One is corresponding.The one-to-one correspondence that the place is said includes that title correspondence is corresponding with position in a matrix.Such as food materials label is heat-clearing
Removing toxic substances, then corresponding user tag is that concern is clearing heat and detoxicating.The clearing heat and detoxicating position in food materials eigenmatrix is the first row the
One column, then paying close attention to the clearing heat and detoxicating position in user's food materials preference profiles matrix is also the first row first row.In step S16
In, the similarity of a food materials and user's food materials preference is calculated according to food materials eigenmatrix and user's food materials preference profiles matrix.To
Any food materials in food materials are recommended to have oneself corresponding food materials eigenmatrix, according to the corresponding food materials eigenmatrix meter of a certain food materials
Calculate the similarity of the food materials and user's food materials preference profiles matrix.In this step, all food materials to be recommended and user's food are calculated
The similarity of material preference.In step S17, according to similarity export food materials information, wherein food materials information include food materials title and/
Or commodity title corresponding with the food materials title.For example, the similarity of all food materials to be recommended is subjected to ranking from high to low, it will
The food materials of N recommend user before similarity comes, and export the food materials information of this N number of food materials.
The commodity title of food materials to be recommended is handled to generate food materials eigenmatrix, handles user behaviors log of the user about recipe
To generate user's food materials preference profiles matrix, according to food materials eigenmatrix and user's food materials preference profiles matrix calculate a food materials and
The similarity of user's food materials preference recommends food materials to user according to the similarity.In this way, behavior day of the analysis user about recipe
Will obtains the food materials preference of user, is pushed away according to the similarity of food materials to be recommended and the user's food materials preference analyzed to user
Food materials are recommended, the matching degree of recommended food materials and user preferences is improved.
Fig. 2 be another embodiment of the present invention provides for user recommend food materials method flow chart.It is shown in Fig. 2
Method and method shown in FIG. 1 the difference is that, this method further includes the following contents.In step S28, output and food materials
The corresponding food materials label of title, to inform why user recommends recommended food materials, is improved as the explanation to recommendation results
The degree of belief of user.For example, classification, heat and the nutritive effect of output food materials, to inform why user is pushed away to it
The food materials recommended.
Optionally, in embodiments of the present invention, processing commodity title may include the following contents to generate food materials portrait, such as scheme
Shown in 2.In step s 30, commodity title is segmented.In step S31, include in the word that filtering participle obtains stops
Word.In step s 32, food materials title is marked to filtered word.In step S33, according to the food materials title of mark and
Food materials label corresponding with the food materials title generates food materials portrait.For example, according to food materials title with the food materials that are stored in database
Label is matched, and food materials portrait is generated.
Optionally, in embodiments of the present invention, generating user's food materials preference portrait according to user behaviors log and recipe can wrap
Include the following contents.As shown in figure 4, in step s 40, the power of the behavioral parameters in user behaviors log is determined using analytic hierarchy process (AHP)
Weight.In step S41, the number of behavioral parameters is counted.For example, behavioral parameters include browse, search for, thumbing up, commenting on, collecting,
It uploads, counts the number that lower user browses a certain recipe, searches for, thumbing up, commenting on, collecting, uploading respectively.In step S42
In, the weight of recipe is determined according to the weight of behavioral parameters and number.In step S43, recipe is determined according to the weight of recipe
The weight of corresponding food materials.It should be noted that the described recipe at this can be only a recipe, it also may include more
A recipe.By taking multiple recipes as an example, the weight for determining food materials is illustrated how.Such as recipe includes that fiddlehead fries balsam pear, duck silk mixes fragrant-flowered garlic
Dish, balsam pear scrambled eggs, it is fiddlehead and balsam pear that fiddlehead, which fries the corresponding food materials of balsam pear, and it is duck and fragrant-flowered garlic that duck silk, which mixes the corresponding food materials of leek,
Dish, the corresponding food materials of balsam pear scrambled eggs are balsam pear and egg, after the weight of each recipe has been determined, determine the corresponding food of recipe
The weight of material, that is, determine fiddlehead, balsam pear, duck, leek, egg weight, it should be noted that fiddlehead fries balsam pear and balsam pear and fries
There is balsam pear in egg, the weight of balsam pear need to be determined according to the weight of this two recipes.In step S44, recipe pair is determined
Food materials label corresponding to the food materials answered.For example, determining that the corresponding food materials of recipe are fiddlehead, hardship by taking above three recipe as an example
Melon, duck, leek, egg determine fiddlehead, balsam pear, duck, leek, the corresponding food materials label of egg respectively in this step.?
In step S45, the weight of the food materials according to corresponding to recipe determines the weight of food materials label.In step S46, by user tag
It is corresponding with food materials label, for example, food materials label be stem leafy vegetable and product, then corresponding user tag be stem leafy vegetable and
Product fan, food materials label be it is low in calories, then corresponding user tag is to pay close attention to food materials low in calories.According to food materials label, food
The weight and user tag of material label generate user's food materials preference portrait.
It optionally, in embodiments of the present invention, include that preset value food materials are special for the food materials eigenmatrix of same food materials
Matrix is levied, the food materials label for including in the food materials eigenmatrix in preset value food materials eigenmatrix belongs to same class food materials mark
Label.User's food materials preference profiles matrix includes preset value user's food materials preference profiles matrix, preset value user's food materials preference
The user tag for including in user's food materials preference profiles matrix in eigenmatrix belongs to same class user tag, food materials label
Classification it is corresponding with the classification of user tag.For example, food materials label is divided into two major classes, belong to food materials classification is one kind,
Belong to food materials heat is a kind of.Then for the food materials eigenmatrix of same food materials, including two food materials eigenmatrixes, one
A food materials eigenmatrix for belonging to food materials classification for included food materials label, can be named as food materials classification matrix, another is
Included food materials label belongs to the food materials eigenmatrix of food materials heat, can be named as food materials heat matrix.By user tag
When classification, user tag is also divided into two classes, corresponding with food materials classification is one kind, and corresponding with food materials heat is a kind of.Then
User's food materials preference profiles matrix includes two matrixes, and included user tag is corresponding with food materials classification in a matrix
, it can be named as user's food materials classification preference matrix, included user tag is corresponding with food materials heat in another matrix,
It is named as user's food materials heat preference matrix.The similarity for calculating a food materials and user's food materials preference includes: to calculate separately a food
Each food materials eigenmatrix in the preset value food materials eigenmatrix of material and user corresponding with each food materials eigenmatrix
The characteristic similarity of food materials preference profiles matrix, the wherein corresponding relationship of food materials eigenmatrix and user's food materials preference profiles matrix
It is determined according to the corresponding relationship of the classification of food materials label and the classification of user tag；And according to characteristic similarity, with feature phase
Similarity is calculated like the corresponding feature weight of degree and preset value.Wherein, feature weight is related to practical application scene.Such as general
It is corresponding with the similarity of food materials eigenmatrix about food materials heat when the method for the recommendation food materials is applied in scene of reducing weight
Feature weight is more relatively high.When applying the method for the recommendation food materials in health scene, and about nutritive effect
The corresponding feature weight of the similarity of food materials eigenmatrix is more relatively high.Optionally, it is calculated using cosine similarity formula
Characteristic similarity.
Optionally, in the embodiment of the present invention, preset value food materials eigenmatrix includes food materials classification matrix, food materials heat square
Battle array and food materials nutritive effect matrix, preset value user's food materials preference profiles matrix include user's food materials classification preference matrix, use
Family food materials heat preference matrix and user health attribute matrix；Characteristic similarity includes food materials classification matrix and user's food materials classification
The fisrt feature similarity of preference matrix, the second feature similarity of food materials heat matrix and user's food materials heat preference matrix and
The third feature similarity of food materials nutritive effect matrix and user health attribute matrix；According to characteristic similarity, similar to feature
Spending corresponding feature weight and preset value calculating similarity includes: by fisrt feature similarity, second feature similarity and third
Characteristic similarity is respectively multiplied by corresponding feature weight；The value linear, additive that will be obtained after multiplication；And the value after will add up is removed
With 3 to obtain similarity.
Below in another embodiment of the invention, food materials label is divided into food materials classification, food materials heat and food materials to seek
Support effect three classes for, illustrate how in a certain food materials store to user recommend food materials.In this embodiment, Jin Jinlie
Having lifted includes several food materials in food materials store.It should be noted that the illustration at this is used for the purpose of better illustrating this
Invention is not used in the limitation present invention.
Process one: food materials data → food materials portrait → food materials eigenmatrix
1, food materials classification is as shown in Table 1:
| Food materials classification | Citing |
| Neat of animal and product | Pork, beef, mutton etc. |
| Meat of poultris and product | Chicken, duck, goose etc. |
| Eggs and product | Egg, duck's egg, quail egg etc. |
| Nut, kind subclass | Pine nut, cashew nut etc. |
| Newborn class and product | Milk, Yoghourt etc. |
| Cauline leaf class greengrocery and product | Chinese cabbage, cabbage, rape etc. |
| Root piece kind vegetable class and product | Sweet potato, lotus rhizome, radish, bamboo shoots etc. |
| Melon-fruit-like vegetable class and product | Corn, eggplant, cucumber, tomato etc. |
| Legume vegetable and product | Pea, semen viciae fabae, bean curd etc. |
| Homonemeae | Agaric, mushroom etc. |
| Aquatic product | Fish, seaweed, kelp etc. |
| Fruits and product | Apple etc. |
One food materials classification standard of table
2, the specific food materials label that the classification of food materials label and every a kind of food materials label include is as shown in Table 2.
Two food materials label of table
3, food materials mall information, which arranges, generates food materials portrait.Food materials commodity title is segmented；Stop words is filtered；It is right
Food materials title is labeled；It is carried out according to food materials title with food materials classification, heat and the nutritive effect field stored in database
Matching, exports content as shown in Table 3, that is, generates food materials portrait.
| Food materials commodity title | Food materials title | Food materials classification | Food materials heat | Food materials nutritive effect |
| Dabie Mountain egg with green shell 40 | Egg | Eggs and product | It is low in calories | Brain healthy, strengthen immunity, cancer-resisting, beautifying face and moistering lotion |
| 40 pieces of the organic race mountain soil egg in Jinzhai County | Egg | Eggs and product | It is low in calories | Brain healthy, strengthen immunity, cancer-resisting, beautifying face and moistering lotion |
| Hainan banana 9000-10000g/ case | Banana | Fruits and product | It is low in calories | Weight reducing oil extraction, improving eyesight, appetite-stimulating and indigestion-relieving, blood pressure lowering |
| 8, Ecuador's banana (GD) | Banana | Fruits and product | It is low in calories | Weight reducing oil extraction, improving eyesight, appetite-stimulating and indigestion-relieving, blood pressure lowering |
| Lan Xue Ecuador white shrimp 2kg/ box | Shrimp | Aquatic product | It is low in calories | Protect liver |
| Sharp 600g/ bags of the slice of fish of Vietnam dragon | Long Liyu | Aquatic product | It is low in calories | Norcholesterol, appetite-stimulating and indigestion-relieving, blood pressure lowering |
| Denmark's imperial crown day Rangu raises trotter 700g | Trotter | Neat of animal and product | High heat | Beautifying face and moistering lotion |
Three food materials of table portrait
4, food materials eigenmatrix is generated.As shown in Table 3, it is determined that the corresponding food materials of included food materials in food materials store
Title and the corresponding food materials label of food materials title.In this embodiment, by food materials label be divided into food materials classification, food materials heat and
Food materials nutritive effect three classes, then food materials eigenmatrix include three matrixes, respectively food materials classification matrix, food materials heat matrix and
Food materials nutritive effect matrix.The food materials label that food materials classification, food materials heat and food materials nutritive effect respectively include is as shown in Table 2.
For a food materials, in a certain food materials eigenmatrix, if food materials meet food materials label, 1 is marked, is marked if not meeting
0.For example, food materials label is low in calories, medium heat and high heat in food materials heat matrix.For egg, egg category
In food materials low in calories, in the food materials heat matrix of egg, place low in calories marks 1, marks 0 at medium heat and high heat.It is right
For trotter, trotter belongs to high heat food materials, then in the food materials heat matrix of trotter, mark 1 at high heat is low in calories in
0 is marked at equal heats.Other food materials eigenmatrixes are obtained with same method.The food materials classification matrix of each food materials, food materials heat
Moment matrix and food materials nutritive effect matrix are respectively as shown in chart four to table six.
Four food materials classification matrix of table
Five food materials heat matrix of table
Six food materials nutritive effect matrix of table
Process two: user behavior data → user's food materials preference portrait → user's food materials preference profiles matrix
1, the weight of parameters is determined using analytic hierarchy process (AHP) (AHP), that is, determine the parameters for being directed to a certain recipe
Weight, and count the number of parameters.By taking fiddlehead fries balsam pear as an example, weight and number are as shown in Table 7.
| Behavioral parameters | Browsing | Search | It thumbs up | Comment | Collection | It uploads |
| Weight (AHP method) | 2 | 1 | 3 | 4 | 4 | 5 |
| Number (statistics) | 10 | 5 | 1 | 2 | 1 | 0 |
Seven user of table browses recipe behavioral parameters table
2, recipe weight is determined according to the weight of parameters and number, determines that recipe corresponds to food materials.Balsam pear is fried with fiddlehead
For, recipe weight is 2 × 10+1 × 5+3 × 1+4 × 2+4 × 1+5 × 0=40, and corresponding food materials are fiddlehead, balsam pear.At this
In embodiment, by recipe include these three recipes for determining for fiddlehead fries balsam pear, duck silk mixes leek, balsam pear scrambled eggs
Weight and corresponding food materials it is as shown in Table 8.
| Recipe name | Recipe weight | Corresponding food materials |
| Fiddlehead fries balsam pear | 40 | Fiddlehead, balsam pear |
| Duck silk mixes leek | 55 | Duck, leek |
| Balsam pear scrambled eggs | 73 | Balsam pear, egg |
Eight recipe weight of table and corresponding food materials list
3, according to the weight of recipe determine recipe corresponding to food materials weight, calculating process as shown in Table 9 determines food
The weight of material.
| Food materials title | Food materials weight | Calculating process |
| Fiddlehead | 40 | 40 × 1=40 |
| Balsam pear | 113 | 40 × 1+73 × 1=113 |
| Duck | 55 | 55 × 1=55 |
| Leek | 55 | 55 × 1=56 |
| Egg | 73 | 73 × 1=73 |
Nine food materials weight calculation table of table
4, food materials label corresponding to food materials corresponding to recipe is determined.In this embodiment, the food materials being related to include
Fiddlehead, balsam pear, duck, leek, egg, corresponding food materials label are as shown in Table 10.
| Food materials title | Food materials classification | Food materials heat | Food materials nutritive effect |
| Fiddlehead | Stem leafy vegetable and product | It is low in calories | Clearing heat and detoxicating, strengthen immunity, lowering blood pressure and blood fat, norcholesterol |
| Balsam pear | Melon-fruit-like vegetable and product | It is low in calories | Hypoglycemic, cancer-resisting |
| Duck | Meat of poultris and product | High heat | Enriching the blood and tonifying qi, nourishing the stomach |
| Leek | Stem leafy vegetable and product | It is low in calories | Blood-stasis-dissipating blood-circulation-promoting, appetite-stimulating and indigestion-relieving |
| Egg | Eggs and product | It is low in calories | Brain healthy, strengthen immunity, cancer-resisting, beautifying face and moistering lotion |
The corresponding food materials label of the corresponding food materials of ten recipe of table
5, food materials class weight, food materials nutritive effect weight and food materials heat weight table are obtained according to " 3 " and " 4 " step,
Respectively as shown in table 11 to 13.
| Food materials classification | Weight | Weight computations |
| Stem leafy vegetable and product | 96 | 40 × 1+56 × 1=96 |
| Melon-fruit-like vegetable and product | 113 | 113 × 1=113 |
| Meat of poultris and product | 55 | 55 × 1=55 |
| Eggs and product | 73 | 73 × 1=73 |
11 food materials class weight table of table
| Food materials nutritive effect | Weight | Weight computations |
| It is clearing heat and detoxicating | 40 | 40 × 1=40 |
| Strengthen immunity | 113 | 40 × 1+73 × 1=113 |
| Blood pressure lowering | 40 | 40 × 1=40 |
| Reducing blood lipid | 40 | 40 × 1=40 |
| Norcholesterol | 40 | 40 × 1=40 |
| It is hypoglycemic | 113 | 113 × 1=113 |
| Cancer-resisting | 186 | 113 × 1+73 × 1=186 |
| Enriching the blood and tonifying qi | 55 | 55 × 1=55 |
| Nourishing the stomach | 55 | 55 × 1=55 |
| Blood-stasis-dissipating blood-circulation-promoting | 55 | 55 × 1=55 |
| Appetite-stimulating and indigestion-relieving | 55 | 55 × 1=55 |
| Brain healthy | 73 | 73 × 1=73 |
| Beautifying face and moistering lotion | 73 | 73 × 1=73 |
12 food materials nutritive effect weight table of table
| Food materials heat | Weight | Weight computations |
| It is low in calories | 271 | 40 × 1+113 × 1+55 × 1+73 × 1=271 |
| Medium heat | 0 | 0 |
| High heat | 55 | 55 × 1=55 |
13 food materials heat weight table of table
6, food materials label is corresponding with user tag, the power of a kind of food materials label is gone out according to the weight calculation of food materials label
It is overlapped meter, and then calculates the weight proportion of a certain food materials label, obtains user's food materials preference portrait.User's food materials classification preference
Portrait, user's food materials heat preference portrait, user health attribute portrait are respectively as shown in table 14 to 16.
Weight adds up to=96+113+55+73=337
| Food materials classification | Weight | Corresponding user's food materials preference label | Weight proportion |
| Stem leafy vegetable and product | 96 | Stem leafy vegetable and product fan | 96 ÷ 337=38.49% |
| Melon-fruit-like vegetable and product | 113 | Melon-fruit-like vegetable and product fan | 113 ÷ 337=33.53% |
| Meat of poultris and product | 55 | Meat of poultris and product fan | 55 ÷ 337=16.32% |
| Eggs and product | 73 | Eggs and product fan | 73 ÷ 337=21.66% |
14 user's food materials classification preference of table portrait
Weight adds up to=40+113+40+40+40+113+186+55+55+55+55+73+73=938
| Food materials nutritive effect | Weight | Corresponding user's food materials preference label | Weight proportion |
| It is clearing heat and detoxicating | 40 | It pays close attention to clearing heat and detoxicating | 40 ÷ 938=4.26% |
| Strengthen immunity | 113 | Pay close attention to strengthen immunity | 113-938=12.05% |
| Blood pressure lowering | 40 | Pay close attention to blood pressure lowering | 40 ÷ 938=4.26% |
| Reducing blood lipid | 40 | Pay close attention to reducing blood lipid | 40 ÷ 938=4.26% |
| Norcholesterol | 40 | Pay close attention to norcholesterol | 40 ÷ 938=4.26% |
| It is hypoglycemic | 113 | It pays close attention to hypoglycemic | 113 ÷ 938=12.05% |
| Cancer-resisting | 186 | Pay close attention to cancer-resisting | 186 ÷ 938=19.86% |
| Enriching the blood and tonifying qi | 55 | Pay close attention to enriching the blood and tonifying qi | 55 ÷ 938=5.86% |
| Nourishing the stomach | 55 | Pay close attention to nourishing the stomach | 55 ÷ 938=5.86% |
| Blood-stasis-dissipating blood-circulation-promoting | 55 | Pay close attention to blood-stasis-dissipating blood-circulation-promoting | 55 ÷ 938=5.86% |
| Appetite-stimulating and indigestion-relieving | 55 | Pay close attention to appetite-stimulating and indigestion-relieving | 55 ÷ 938=5.86% |
| Brain healthy | 73 | Pay close attention to brain healthy | 73 ÷ 938=7.78% |
| Beautifying face and moistering lotion | 73 | Pay close attention to beautifying face and moistering lotion | 73 ÷ 938=7.78% |
15 user health attribute of table portrait
| Food materials heat | Weight | Corresponding user's food materials preference label | Weight proportion |
| It is low in calories | 271 | Pay close attention to food materials low in calories | 271 ÷ (271+55)=83.12% |
| Medium heat | 0 | Pay close attention to medium heat food materials | 0 |
| High heat | 55 | Pay close attention to high heat food materials | 16.88% |
16 user's food materials heat preference of table portrait
Food available by table 14 to 16, which has 19.86% a possibility that go purchase anticancer effect
Material has 12.05% a possibility that buy hypoglycemic and strengthen immunity food materials, has 83.12% a possibility that buy low in calories
Food materials.
7, it is drawn a portrait according to user's food materials preference and generates user's food materials preference profiles matrix.As shown in table 14 to 16,
User tag is divided into three classes, and the classification method is corresponding with the classification of food materials label, then user's food materials preference profiles matrix includes
Three, respectively user's food materials classification preference matrix, user health attribute matrix and user's food materials heat preference matrix, user's food
Included user tag is corresponded with food materials label included in food materials classification matrix in material classification preference matrix, user
Included user tag is corresponded with food materials label included in food materials nutritive effect matrix in healthy attribute matrix, is used
Included user tag is corresponded with food materials label included in food materials heat matrix in the food materials heat preference matrix of family.
In user's food materials preference profiles matrix, if user tag in table 14 to 16 corresponding weight proportion be 0, with
The corresponding position mark 0 of family food materials preference profiles matrix, if weight proportion is not 0, marks its weight in corresponding position
Ratio.User's food materials preference special medical treatment matrix generated is as shown in table 17 to 19.
Table 17
Table 18
Table 19
Process three: food materials eigenmatrix and user's food materials preference profiles matrix similarity calculation
Using cosine similarity formula respectively by the food materials eigenmatrix of user's food materials preference profiles matrix and different food materials
Carry out similarity calculation, that is, for any food materials, calculate separately the phase of food materials classification matrix and user's food materials classification preference matrix
Like degree, the similarity of food materials nutritive effect matrix and user health attribute matrix, food materials heat matrix and user's food materials heat are inclined
The similarity of good matrix, then will by each similarity of a food materials multiplied by feature weight after, carry out linear, additive, obtain the food materials
With the final similarity of user's food materials preference.
Cosine similarity formula:
Wherein, number or corresponding user food materials preference profiles square of the n for the food materials label in a food materials eigenmatrix
The number of user tag in battle array, xiFor the corresponding value of a food materials label, y in a food materials eigenmatrixiFor user's food materials preference
The corresponding value of a user tag in eigenmatrix.
1, it substitutes into formula and calculates food materials classification matrix and user's food materials preference matrix similarity: by food materials each in table four
Food materials classification matrix brings above-mentioned cosine similarity formula into respectively, calculates the class with user's food materials classification preference matrix (table 17)
Other similarity, calculated egg, banana, shrimps, the classification similarity difference of Long Liyu and trotter are as follows:
(1) egg classification similarity=21.66%
(2) banana classification similarity=0
(3) other similarity=0 of shrimps
(4) other similarity=0 of the sharp fish of dragon
(5) trotter classification similarity=0
2, it substitutes into formula and calculates food materials nutritive effect matrix and user health attribute matrix similarity: by food each in table six
The food materials nutritive effect matrix of material brings above-mentioned cosine similarity formula into respectively, calculates and user health attribute matrix (table 18)
The effect of similarity, similarity difference is as follows the effect of calculated egg, banana, shrimps, Long Liyu and trotter:
(6) egg effect similarity=47.47%
(7) banana effect similarity=4.26%
(8) shrimp effect similarity=0
(9) Long Liyu effect similarity=14.38%
(10) trotter effect similarity=7.78%
3, it substitutes into formula and calculates food materials heat matrix and user's food materials heat preference matrix similarity: by food each in table five
The food materials heat matrix of material brings above-mentioned cosine similarity formula into respectively, calculates and user's food materials heat preference matrix (table 19)
Heat similarity, calculated egg, banana, shrimps, the heat similarity difference of Long Liyu and trotter be as follows:
(11) egg heat similarity=83.12%
(12) banana heat similarity=83.12%
(13) shrimp heat similarity=83.12%
(14) Long Liyu heat similarity=83.12%
(15) trotter heat similarity=16.88%
4, the final similarity calculation of food materials
The same each characteristic similarity of food materials multiplied by feature weight (feature weight according to different scenes carry out reasonable set),
Then food materials and the final similarity of user preference can be obtained in linear, additive.Each feature weight of this example setting is equal, is 1.Then,
(1) egg and user's food materials preference similarity
=(classification similarity+effect similarity+heat similarity) ÷ 3
=(21.66%+47.47%+83.12%) ÷ 3
=50.75%
(2) banana and user's food materials preference similarity
=(classification similarity+effect similarity+heat similarity) ÷ 3
=(0+4.26%+83.12%) ÷ 3
=29.13%
(3) shrimp and user's food materials preference similarity
=(classification similarity+effect similarity+heat similarity) ÷ 3
=(0+0+83.12%) ÷ 3
=27.71%
(4) Long Liyu and user's food materials preference similarity
=(classification similarity+effect similarity+heat similarity) ÷ 3
=(0+14.38%+83.12%) ÷ 3
=32.50%
(5) trotter and user's food materials preference similarity
=(classification similarity+effect similarity+heat similarity) ÷ 3
=(0+7.78%+16.88%) ÷ 3
=8.22%
Process four: according to sequencing of similarity, as shown in table 20.
Calculated result (similarity percentage) is sorted from high to low.
Table 20
Process five: recommendation food materials commodity of the output Top-N as primary election
Output comes food materials Recommendations of the preceding N food materials as primary election, pushes away as shown in table 21, and in food materials
The classification for indicating the food materials in reason, heat and nutritive effect are recommended, as the explanation for recommending the food materials, is pushed away with improving user to this
Recommend the degree of belief of system.
| Corresponding goods title | Food materials title | Rationale for the recommendation |
| Dabie Mountain egg with green shell 40 | Egg | It is eggs, strengthen immunity, beautifying face and moistering lotion, cancer-resisting, brain healthy, low in calories |
| 40 pieces of the organic mountain soil egg in Jinzhai County | Egg | It is eggs, strengthen immunity, beautifying face and moistering lotion, cancer-resisting, brain healthy, low in calories |
| Sharp 600g/ bags of the slice of fish of Vietnam dragon | Long Liyu | It is appetite-stimulating and indigestion-relieving, blood pressure lowering, norcholesterol, low in calories |
| 8, Ecuador's banana (GD) | Banana | It is appetite-stimulating and indigestion-relieving, blood pressure lowering, low in calories |
| Hainan banana 9000-10000g/ case | Banana | It is appetite-stimulating and indigestion-relieving, blood pressure lowering, low in calories |
| Lan Xue Ecuador white shrimp 2kg/ box | Shrimp | It is low in calories |
| Denmark's imperial crown day Rangu raises trotter 700g | Trotter | Beautifying face and moistering lotion |
Table 21
Correspondingly, the another aspect of the embodiment of the present invention provides a kind of for recommending the device of food materials to user.Fig. 5 is this
Invent the structural block diagram for the device for recommending food materials to user that another embodiment provides.As shown in figure 5, the device includes logical
Believe module 1, processing module 2 and output module 3.Communication module 1 is used for: being received the commodity title of food materials to be recommended, and is received
The user behaviors log about recipe of user.Processing module 2 is used for: processing commodity title is to generate food materials portrait；It is drawn according to food materials
As generating food materials eigenmatrix；User's food materials preference portrait is generated according to user behaviors log and recipe；It is drawn according to user's food materials preference
As generating user's food materials preference profiles matrix, wherein food materials label and user's food materials preference profiles square in food materials eigenmatrix
User tag in battle array corresponds；And according to food materials eigenmatrix and user's food materials preference profiles matrix calculate a food materials and
The similarity of user's food materials preference.Output module 3 is used to export food materials information according to similarity, and wherein food materials information includes food materials
Title and/or the commodity title corresponding with the food materials title.
The commodity title of food materials to be recommended is handled to generate food materials eigenmatrix, handles user behaviors log of the user about recipe
To generate user's food materials preference profiles matrix, according to food materials eigenmatrix and user's food materials preference profiles matrix calculate a food materials and
The similarity of user's food materials preference recommends food materials to user according to the similarity.In this way, behavior day of the analysis user about recipe
Will obtains the food materials preference of user, is pushed away according to the similarity of food materials to be recommended and the user's food materials preference analyzed to user
Food materials are recommended, the matching degree of recommended food materials and user preferences is improved.
Optionally, in embodiments of the present invention, output module is also used to export food materials label corresponding with food materials title, makees
The degree of belief of user is improved for the explanation to recommendation results to inform why user recommends recommended food materials.
Optionally, in embodiments of the present invention, processing module processing commodity title includes: to commodity to generate food materials portrait
Title is segmented；The stop words for including in the word that filtering participle obtains；Food materials title is marked to filtered word；According to
The food materials title of mark and food materials label corresponding with the food materials title generate food materials portrait.
Optionally, in embodiments of the present invention, processing module generates user's food materials preference according to user behaviors log and recipe and draws
As including: the weight for determining the behavioral parameters in user behaviors log using analytic hierarchy process (AHP)；Count the number of behavioral parameters；According to row
The weight of recipe is determined for the weight and number of parameter；Determine that recipe corresponds to the weight of food materials according to the weight of recipe；Determine food
The corresponding food materials label of spectrum；The weight of the food materials according to corresponding to recipe determines the weight of food materials label；And user is marked
Label are corresponding with food materials label, generate user's food materials preference portrait according to food materials label, the weight of food materials label and user tag.
It optionally, in embodiments of the present invention, include that preset value food materials are special for the food materials eigenmatrix of same food materials
Matrix is levied, the food materials label for including in the food materials eigenmatrix in preset value food materials eigenmatrix belongs to same class food materials mark
Label, user's food materials preference profiles matrix includes preset value user's food materials preference profiles matrix, preset value user's food materials preference
The user tag for including in user's food materials preference profiles matrix in eigenmatrix belongs to same class user tag, food materials label
Classification it is corresponding with the classification of user tag.Processing module is according to food materials eigenmatrix and user's food materials preference profiles matrix meter
It includes: every in the preset value food materials eigenmatrix for calculate separately a food materials for calculating the similarity of a food materials and user's food materials preference
The characteristic similarity of a food materials eigenmatrix and user's food materials preference profiles matrix corresponding with each food materials eigenmatrix,
The corresponding relationship of middle food materials eigenmatrix and user's food materials preference profiles matrix is according to the classification of food materials label and user tag
The corresponding relationship of classification determines；And according to characteristic similarity, feature weight corresponding with the characteristic similarity and described pre-
If value calculates the similarity.
Optionally, in embodiments of the present invention, processing module calculates characteristic similarity using cosine similarity formula.
Optionally, in embodiments of the present invention, preset value food materials eigenmatrix includes food materials classification matrix, food materials heat
Matrix and food materials nutritive effect matrix, preset value user's food materials preference profiles matrix include user's food materials classification preference matrix,
User's food materials heat preference matrix and user health attribute matrix；Characteristic similarity includes food materials classification matrix and user's food materials class
Fisrt feature similarity, the second feature similarity of food materials heat matrix and user's food materials heat preference matrix of other preference matrix
With the third feature similarity of food materials nutritive effect matrix and user health attribute matrix；Processing module according to characteristic similarity,
It includes: by fisrt feature similarity, the second spy that feature weight corresponding with characteristic similarity and preset value, which calculate the similarity,
Similarity and third feature similarity are levied respectively multiplied by corresponding feature weight；The value linear, additive that will be obtained after multiplication；And
Value after will add up is divided by 3 to obtain similarity.
It is provided in an embodiment of the present invention for user recommend food materials device concrete operating principle and benefit with it is above-mentioned
The concrete operating principle and benefit of method provided in an embodiment of the present invention for recommending from food materials to user are similar, here will no longer
It repeats.
In addition, the another aspect of the embodiment of the present invention provides a kind of household electrical appliance, which includes above-described embodiment
Described in device.Optionally, which can be refrigerator.
In addition, the another aspect of the embodiment of the present invention also provides a kind of machine readable storage medium, the machine readable storage
Instruction is stored on medium, which is used for so that machine executes method described in above-described embodiment.
User's food materials preference portrait is improved in conclusion calculating by User action log, by analyzing food to be recommended
The merchandise news of material obtains food materials portrait, and in the case where fully understanding user's food materials preference and food materials information, the two is carried out
Matching, by the high food materials commercial product recommending of matching degree to user, improves the matching degree of recommended food materials and user preferences.In addition,
Rationale for the recommendation is provided, as the explanation to recommendation results, improves user to the degree of belief of recommendation.
The optional embodiment of the embodiment of the present invention is described in detail in conjunction with attached drawing above, still, the embodiment of the present invention is simultaneously
The detail being not limited in above embodiment can be to of the invention real in the range of the technology design of the embodiment of the present invention
The technical solution for applying example carries out a variety of simple variants, these simple variants belong to the protection scope of the embodiment of the present invention.
It is further to note that specific technical features described in the above specific embodiments, in not lance
In the case where shield, it can be combined in any appropriate way.In order to avoid unnecessary repetition, the embodiment of the present invention pair
No further explanation will be given for various combinations of possible ways.
It will be appreciated by those skilled in the art that implementing the method for the above embodiments is that can pass through
Program is completed to instruct relevant hardware, which is stored in a storage medium, including some instructions are used so that single
Piece machine, chip or processor (processor) execute all or part of the steps of each embodiment the method for the application.And it is preceding
The storage medium stated includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory
The various media that can store program code such as (RAM, Random Access Memory), magnetic or disk.
In addition, any combination can also be carried out between a variety of different embodiments of the embodiment of the present invention, as long as it is not
The thought of the embodiment of the present invention is violated, equally should be considered as disclosure of that of the embodiment of the present invention.
Claims (17)
1. a kind of for recommending the method for food materials to user, which is characterized in that this method comprises:
Receive the commodity title of food materials to be recommended；
The commodity title is handled to generate food materials portrait；
It is drawn a portrait according to the food materials and generates food materials eigenmatrix；
Receive the user behaviors log about recipe of the user；
User's food materials preference portrait is generated according to the user behaviors log and the recipe；
It is drawn a portrait according to user's food materials preference and generates user's food materials preference profiles matrix, wherein in the food materials eigenmatrix
Food materials label and user's food materials preference profiles matrix in user tag correspond；
A food materials and user's food materials preference are calculated according to the food materials eigenmatrix and user's food materials preference profiles matrix
Similarity；And
According to the similarity export food materials information, wherein the food materials information include food materials title and/or with the food materials title
The corresponding commodity title.
2. the method according to claim 1, wherein this method further include:
Export the food materials label corresponding with the food materials title.
3. according to claim 1 or method described in 2, which is characterized in that the processing commodity title is to generate food materials
Portrait includes:
The commodity title is segmented；
The stop words for including in the word that filtering participle obtains；
Food materials title is marked to filtered word；And
Food materials portrait is generated according to the food materials title of mark and the food materials label corresponding with the food materials title.
4. according to claim 1 or method described in 2, which is characterized in that described according to the user behaviors log and the recipe
Generating user's food materials preference portrait includes:
The weight of the behavioral parameters in the user behaviors log is determined using analytic hierarchy process (AHP)；
Count the number of the behavioral parameters；
The weight of the recipe is determined according to the weight of the behavioral parameters and the number；
According to the weight of the recipe determine the recipe corresponding to food materials weight；
Determine the food materials label corresponding to food materials corresponding to the recipe；
The weight of the food materials label is determined according to the weight of food materials corresponding to the recipe；And
The user tag is corresponding with the food materials label, according to the food materials label, the weight of the food materials label and institute
It states user tag and generates user's food materials preference portrait.
5. according to claim 1 or method described in 2, which is characterized in that for the food materials eigenmatrix of same food materials
Including preset value food materials eigenmatrix, the institute that includes in the food materials eigenmatrix in the preset value food materials eigenmatrix
It states food materials label and belongs to same class food materials label, user's food materials preference profiles matrix includes preset value user's food materials preference
Eigenmatrix, the institute for including in user's food materials preference profiles matrix in the preset value user food materials preference profiles matrix
It states user tag and belongs to same class user tag, the classification of the food materials label is corresponding with the classification of the user tag；
It is described inclined according to the food materials eigenmatrix and user's food materials preference profiles matrix one food materials of calculating and user's food materials
Good similarity includes:
Calculate separately each food materials eigenmatrix in the preset value food materials eigenmatrix of a food materials and with each food
The characteristic similarity of the corresponding user's food materials preference profiles matrix of material eigenmatrix, wherein the food materials eigenmatrix and institute
The corresponding relationship of user's food materials preference profiles matrix is stated according to the classification and the classification of the user tag of the food materials label
Corresponding relationship determines；And
It is calculated according to the characteristic similarity, feature weight corresponding with the characteristic similarity and the preset value described similar
Degree.
6. according to the method described in claim 5, it is characterized in that, similar using the cosine similarity formula calculating feature
Degree.
7. according to the method described in claim 5, it is characterized in that, the preset value food materials eigenmatrix includes food materials classification
Matrix, food materials heat matrix and food materials nutritive effect matrix, the preset value user food materials preference profiles matrix includes user
Food materials classification preference matrix, user's food materials heat preference matrix and user health attribute matrix；
The characteristic similarity includes the fisrt feature phase of the food materials classification matrix with user's food materials classification preference matrix
Like the second feature similarity and the food materials nutrition of degree, the food materials heat matrix and user's food materials heat preference matrix
The third feature similarity of effect matrix and the user health attribute matrix；
It is described according to the characteristic similarity, feature weight corresponding with the characteristic similarity and the preset value calculate described in
Similarity includes:
By the fisrt feature similarity, the second feature similarity and the third feature similarity respectively multiplied by corresponding
The feature weight；
The value linear, additive that will be obtained after multiplication；And
Value after will add up obtains the similarity divided by 3.
8. a kind of for recommending the device of food materials to user, which is characterized in that the device includes；
Communication module is used for:
Receive the commodity title of food materials to be recommended；And
Receive the user behaviors log about recipe of the user；And
Processing module is used for:
The commodity title is handled to generate food materials portrait；
It is drawn a portrait according to the food materials and generates food materials eigenmatrix；
User's food materials preference portrait is generated according to the user behaviors log and the recipe；
It is drawn a portrait according to user's food materials preference and generates user's food materials preference profiles matrix, wherein in the food materials eigenmatrix
Food materials label and user's food materials preference profiles matrix in user tag correspond；And
A food materials and user's food materials preference are calculated according to the food materials eigenmatrix and user's food materials preference profiles matrix
Similarity；And
Output module, for according to the similarity export food materials information, wherein the food materials information include food materials title and/or
The commodity title corresponding with the food materials title.
9. device according to claim 8, which is characterized in that the output module is also used to export and the food materials title
The corresponding food materials label.
10. device according to claim 8 or claim 9, the processing module handles the commodity title to generate the food materials
Portrait includes:
The commodity title is segmented；
The stop words for including in the word that filtering participle obtains；
Food materials title is marked to filtered word；
Food materials portrait is generated according to the food materials title of mark and the food materials label corresponding with the food materials title.
11. device according to claim 8 or claim 9, the processing module is generated according to the user behaviors log and the recipe
User's food materials preference is drawn a portrait
The weight of the behavioral parameters in the user behaviors log is determined using analytic hierarchy process (AHP)；
Count the number of the behavioral parameters；
The weight of the recipe is determined according to the weight of the behavioral parameters and the number；
Determine that the recipe corresponds to the weight of food materials according to the weight of the recipe；
Determine the food materials label corresponding to food materials corresponding to the recipe；
The weight of the food materials label is determined according to the weight of food materials corresponding to the recipe；And
The user tag is corresponding with the food materials label, according to the food materials label, the weight of the food materials label and institute
It states user tag and generates user's food materials preference portrait.
12. device according to claim 8 or claim 9, which is characterized in that for the food materials eigenmatrix packet of same food materials
Preset value food materials eigenmatrix is included, include in the food materials eigenmatrix in the preset value food materials eigenmatrix is described
Food materials label belongs to same class food materials label, and user's food materials preference profiles matrix includes that preset value user's food materials preference is special
Matrix is levied, include in user's food materials preference profiles matrix in the preset value user food materials preference profiles matrix is described
User tag belongs to same class user tag, and the classification of the food materials label is corresponding with the classification of the user tag；
The processing module calculates a food materials and use according to the food materials eigenmatrix and user's food materials preference profiles matrix
The similarity of family food materials preference includes:
Calculate separately each food materials eigenmatrix in the preset value food materials eigenmatrix of a food materials and with each food
The characteristic similarity of the corresponding user's food materials preference profiles matrix of material eigenmatrix, wherein the food materials eigenmatrix and institute
The corresponding relationship of user's food materials preference profiles matrix is stated according to the classification and the classification of the user tag of the food materials label
Corresponding relationship determines；And
It is calculated according to the characteristic similarity, feature weight corresponding with the characteristic similarity and the preset value described similar
Degree.
13. device according to claim 12, which is characterized in that the processing module is calculated using cosine similarity formula
The characteristic similarity.
14. device according to claim 12, which is characterized in that the preset value food materials eigenmatrix includes food materials class
Other matrix, food materials heat matrix and food materials nutritive effect matrix, the preset value user food materials preference profiles matrix include using
Family food materials classification preference matrix, user's food materials heat preference matrix and user health attribute matrix；
The characteristic similarity includes the fisrt feature phase of the food materials classification matrix with user's food materials classification preference matrix
Like the second feature similarity and the food materials nutrition of degree, the food materials heat matrix and user's food materials heat preference matrix
The third feature similarity of effect matrix and the user health attribute matrix；
The processing module is according to the characteristic similarity, feature weight corresponding with the characteristic similarity and the preset value
Calculating the similarity includes:
By the fisrt feature similarity, the second feature similarity and the third feature similarity respectively multiplied by corresponding
The feature weight；
The value linear, additive that will be obtained after multiplication；And
Value after will add up obtains the similarity divided by 3.
15. a kind of household electrical appliance, which is characterized in that the household electrical appliance include dress described in any one of claim 8-14
It sets.
16. household electrical appliance according to claim 15, which is characterized in that the household electrical appliance are refrigerator.
17. a kind of machine readable storage medium, it is stored with instruction on the machine readable storage medium, which is used for so that machine
Perform claim requires method described in any one of 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711037394.8A CN110021403A (en) | 2017-10-30 | 2017-10-30 | Recommend method and apparatus, household electrical appliance and the machine readable storage medium of food materials |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711037394.8A CN110021403A (en) | 2017-10-30 | 2017-10-30 | Recommend method and apparatus, household electrical appliance and the machine readable storage medium of food materials |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110021403A true CN110021403A (en) | 2019-07-16 |
Family
ID=67186708
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711037394.8A Pending CN110021403A (en) | 2017-10-30 | 2017-10-30 | Recommend method and apparatus, household electrical appliance and the machine readable storage medium of food materials |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110021403A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110838356A (en) * | 2019-10-14 | 2020-02-25 | 深圳和而泰家居在线网络科技有限公司 | Data processing method and device and storage medium |
| CN111564199A (en) * | 2020-05-08 | 2020-08-21 | 成都尚医信息科技有限公司 | Intelligent nutrition intervention method and terminal |
| CN112786154A (en) * | 2021-01-18 | 2021-05-11 | 京东方科技集团股份有限公司 | Recipe recommendation method and device, electronic equipment and storage medium |
| CN116545790A (en) * | 2023-06-14 | 2023-08-04 | 深圳市翰视科技有限公司 | Intelligent conference system based on Internet of things |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104731846A (en) * | 2014-11-17 | 2015-06-24 | 陕西师范大学 | Individuation catering recommendation method and system based on multiple targets |
| CN105426552A (en) * | 2015-12-31 | 2016-03-23 | 山东大学 | Method for intelligently recommending menus and food and application |
| CN106651533A (en) * | 2016-12-29 | 2017-05-10 | 合肥华凌股份有限公司 | User behavior-based personalized product recommendation method and apparatus |
-
2017
- 2017-10-30 CN CN201711037394.8A patent/CN110021403A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104731846A (en) * | 2014-11-17 | 2015-06-24 | 陕西师范大学 | Individuation catering recommendation method and system based on multiple targets |
| CN105426552A (en) * | 2015-12-31 | 2016-03-23 | 山东大学 | Method for intelligently recommending menus and food and application |
| CN106651533A (en) * | 2016-12-29 | 2017-05-10 | 合肥华凌股份有限公司 | User behavior-based personalized product recommendation method and apparatus |
Non-Patent Citations (1)
| Title |
|---|
| 王玉山等: "基于皮尔逊相似度的食材推荐算法研究", 《信息与电脑(理论版)》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110838356A (en) * | 2019-10-14 | 2020-02-25 | 深圳和而泰家居在线网络科技有限公司 | Data processing method and device and storage medium |
| CN111564199A (en) * | 2020-05-08 | 2020-08-21 | 成都尚医信息科技有限公司 | Intelligent nutrition intervention method and terminal |
| CN112786154A (en) * | 2021-01-18 | 2021-05-11 | 京东方科技集团股份有限公司 | Recipe recommendation method and device, electronic equipment and storage medium |
| CN116545790A (en) * | 2023-06-14 | 2023-08-04 | 深圳市翰视科技有限公司 | Intelligent conference system based on Internet of things |
| CN116545790B (en) * | 2023-06-14 | 2024-04-02 | 深圳市翰视科技有限公司 | Intelligent conference system based on Internet of things |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110021403A (en) | Recommend method and apparatus, household electrical appliance and the machine readable storage medium of food materials | |
| Ueda et al. | User’s food preference extraction for personalized cooking recipe recommendation | |
| Suwonsichon | The importance of sensory lexicons for research and development of food products | |
| Guthman et al. | Magical disruption? Alternative protein and the promise of de-materialization | |
| Jeremiah | Freezing effects of food quality. | |
| Farrell | The effect of mastication on the digestion of food. | |
| Negara et al. | Application of deep, vacuum, and air frying methods to fry chub mackerel (Scomber japonicus) | |
| Echegaray et al. | Influence of the Inclusion of chestnut (Castanea sativa Miller) in the finishing diet and cooking technique on the physicochemical parameters and volatile profile of biceps femoris muscle | |
| Lamb et al. | Nutritive Value of Canned Foods. XXI. Retention of Nutrients during Commercial Production of various Canned Fruits and Vegetables. | |
| Martins et al. | Feasibility study of a fish by-product recovery plant with a processing capacity of 1,000 kg/day | |
| Jackson et al. | The retention of thiamine, riboflavin and niacin in cooking pork and in processing bacon. | |
| Tinagli et al. | Effect of Cooking with Superheated (SHS) vs. Standard Steam Oven on the Fatty Acids Profile of Different Kinds of Meat and Fish | |
| Mehta et al. | Drying of banana-stepwise effect in drying air temperature on drying kinetics | |
| Edmonds et al. | An economic evaluation of the wet coconut process developed at the Tropical Products Institute. | |
| Monier-Williams | Lead in food. | |
| Kim et al. | Complexity and similarity of recipes based on entropy measurement | |
| Meyer et al. | Retention of some vitamins of the B-complex in beef during cooking. | |
| Atter et al. | Healthy Food Africa Project Report-2021 | |
| Cover et al. | Retention of the B vitamins in beef and lamb after stewing. 1. Experimental design and standardized cooking procedure. J. | |
| Wang DongQing et al. | Dietary patterns are associated with metabolic outcomes among adult Samoans in a cross-sectional study. | |
| Nowacka-Polaczyk et al. | Assessment of frequency of food product consumption by selected, professional athletes, before and after nutritional education. | |
| WO2024033916A1 (en) | Machine learning development of food recipes | |
| Ojagh et al. | Study on the changes of approximate composition and physical parameters of nugget of silver carp (Hypophthalmichthys molitrix) during the initial frying in several vegetable oils at different temperature levels. | |
| Ikejiri et al. | Surprising Recipe Extraction based on Rarity and Generality of Ingredients. | |
| Seong PilNam et al. | The impact of ripening time on technological quality traits, chemical change and sensory characteristics of dry-cured loin. |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 230601 Building No. 198, building No. 198, Mingzhu Avenue, Anhui high tech Zone, Anhui Applicant after: Hefei Midea Intelligent Technology Co.,Ltd. Address before: 230601 No. 176 Jinxiu Avenue, Hefei economic and Technological Development Zone, Anhui, China Applicant before: Hefei Midea Intelligent Technology Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190716 |