CN109300059A - Vegetable recommended method and device - Google Patents
Vegetable recommended method and device Download PDFInfo
- Publication number
- CN109300059A CN109300059A CN201811069258.1A CN201811069258A CN109300059A CN 109300059 A CN109300059 A CN 109300059A CN 201811069258 A CN201811069258 A CN 201811069258A CN 109300059 A CN109300059 A CN 109300059A
- Authority
- CN
- China
- Prior art keywords
- vegetable
- vector
- sample
- attribute information
- recommended
- 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.)
- Granted
Links
- 235000013311 vegetables Nutrition 0.000 title claims abstract description 382
- 238000000034 method Methods 0.000 title claims abstract description 54
- 239000013598 vector Substances 0.000 claims abstract description 200
- 238000012549 training Methods 0.000 claims abstract description 95
- 230000006399 behavior Effects 0.000 claims description 58
- 239000011159 matrix material Substances 0.000 claims description 29
- 238000004891 communication Methods 0.000 claims description 17
- 238000013507 mapping Methods 0.000 claims description 13
- 238000012512 characterization method Methods 0.000 description 5
- 235000012054 meals Nutrition 0.000 description 5
- 230000003542 behavioural effect Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000001419 dependent effect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000004615 ingredient Substances 0.000 description 3
- 238000004590 computer program Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/12—Hotels or restaurants
Landscapes
- Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of vegetable recommended method and devices, wherein method includes: by the customer attribute information of target user and scene properties information MAP into corresponding attribute information vector;Obtain the vegetable vector of more portions of vegetables to be recommended;The vegetable vector of each part of attribute information vector sum vegetable to be recommended is input in the obtained related degree model of training, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value;According to association angle value select that vegetable is recommended to recommend to target user from more portions of vegetables to be recommended.The present invention program, the behavior that can order independent of the history of target user carries out vegetable recommendation, even if for the behavior target user not abundant that orders, also can use that related degree model obtains the attribute information vector of corresponding target user and the vegetable vector of vegetable to be recommended is associated with angle value, and association angle value is recommended accordingly, and then promotes the experience of ordering of target user.
Description
Technical field
The present invention relates to field of computer technology, and in particular to a kind of vegetable recommended method and device.
Background technique
With the rise of intelligent restaurant technology, mode of ordering orders from traditional papery and orders hair to Intelligent Terminal
Exhibition, for example, ordering on taking out platform or ordering in the terminal of ordering that dining room provides.At the same time, in order to improve vegetable
Place an order and rate and promote the experience of ordering of user, more and more offers order service platforms can by terminal to user into
Row vegetable is recommended.
In existing suggested design, the preference information of user is usually determined according to user's history data, and preference accordingly
Information is recommended.In the Chinese patent application for being CN1068157745A such as application publication number, believed according to the comment of target user
The item ingredients preference library for determining target user is ceased, and is determined as the dish of target user's recommendation according to the item ingredients preference library
Product.The program dependent on target user historical review information come be determined as user recommendation vegetable, can not for without comment or
It comments on less target user and accurately vegetable recommendation is provided.For another example application publication number is the Chinese patent Shen of CN108230009A
Please in, by preference prediction model trained in advance, based on target user to the interbehavior feature of target object (including point
The behavioural characteristics such as hit, buy, browse and collect) it is predicted, target user is obtained to the preference prediction result of target object.
The program determined dependent on the history interbehavior of target user target user to the preference of target object, can not to mesh
Interbehavior is not generated between mark object or the less target user of interbehavior accurately carries out the prediction of preference, in turn
Can not accurately it be recommended.
Summary of the invention
In view of the above problems, it proposes on the present invention overcomes the above problem or at least be partially solved in order to provide one kind
State the vegetable recommended method and device of problem.
According to an aspect of the invention, there is provided a kind of vegetable recommended method, comprising:
By the customer attribute information of target user and scene properties information MAP at corresponding attribute information vector;
Obtain the vegetable vector of more portions of vegetables to be recommended;
The vegetable vector of described each part of attribute information vector sum vegetable to be recommended is input to the degree of association mould that training obtains
In type, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value;
It is selected that vegetable is recommended to recommend to target user from more portions of vegetables to be recommended according to the association angle value.
Optionally, before the vegetable vector for obtaining more portions of vegetables to be recommended, the method also includes:
The training corpus of vegetable vectorization model is constructed according to vegetable knowledge mapping;
Training sample corpus is obtained from the training corpus, the stroke feature information of the training sample corpus is defeated
Enter into the vegetable vectorization model of initialization and is trained;Vegetable vectorization matrix is obtained according to training result.
Optionally, the vegetable vector for obtaining more portions of vegetables to be recommended further comprises:
The corresponding stroke feature information of the menu name of each part vegetable to be recommended is reflected according to the vegetable vectorization matrix
Penetrate into the vegetable vector of the vegetable to be recommended.
Optionally, the related degree model is obtained by following steps training:
Obtain the vegetable sample vector of more parts of vegetable samples;And the multiple corresponding user's sample of ordering of behavior of ordering of acquisition
The scene sample attribute information of this user's sample attribute information and scene of ordering;By user's sample attribute information and scene
Sample attribute information MAP is at corresponding sample attribute information vector;
According to behavior of repeatedly ordering, the degree of association of the sample attribute information vector and the vegetable sample vector is carried out
Mark, obtains degree of association annotation results;
The vegetable sample vector and the sample attribute information vector are input in degree of association training pattern, closed
The degree of association of connection degree training pattern output exports result;
Using the loss between the degree of association annotation results and degree of association output result, degree of association training pattern is carried out
Training, obtains the related degree model.
Optionally, the vegetable sample vector for obtaining more parts of vegetable samples further comprises:
According to the vegetable vectorization matrix by the corresponding stroke feature information MAP of the menu name of each part vegetable sample
At the vegetable vector of the vegetable sample.
Optionally, the sample attribute information vector and the degree of association of the vegetable sample vector are labeled described
Before, the method also includes:
Lower single vegetable that the behavior of ordering every time includes is carried out with vegetable sample corresponding, obtains lower single vegetable and vegetable sample
Corresponding relationship;
It is described that the sample attribute information vector and the degree of association of the vegetable sample vector are labeled specifically: root
The sample attribute information vector and the degree of association of the vegetable sample vector are labeled according to the corresponding relationship.
Optionally, it is described according to the corresponding relationship to the sample attribute information vector and the vegetable sample vector
The degree of association be labeled and further comprise:
It is according to corresponding relationship, the vegetable sample vector for the vegetable sample for including in the behavior of ordering is corresponding with the behavior of ordering
The degree of association of sample attribute information vector be labeled as first degree of association;
And/or by the vegetable sample vector for the vegetable sample not included in the behavior of ordering sample corresponding with the behavior of ordering
The degree of association of this attribute information vector is labeled as second degree of association.
Optionally, described to be selected that vegetable is recommended to recommend to target to use from more portions of vegetables to be recommended according to the association angle value
Family further comprises:
More portions of vegetables to be recommended are ranked up from high to low according to association angle value, selected and sorted is forward from ranking results
Preset quantity vegetable to be recommended as recommend vegetable recommend to target user.
According to another aspect of the present invention, a kind of vegetable recommendation apparatus is provided, comprising:
Mapping block, suitable for believing the customer attribute information of target user and scene properties information MAP at corresponding attribute
Cease vector;
Module is obtained, suitable for obtaining the vegetable vector of more portions of vegetables to be recommended;
Prediction module, it is trained suitable for the vegetable vector of described each part of attribute information vector sum vegetable to be recommended to be input to
To related degree model in, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value;
Recommending module recommends vegetable to recommend to target suitable for being selected from more portions of vegetables to be recommended according to the association angle value
User.
Optionally, described device further include:
First training module, suitable for constructing the training corpus of vegetable vectorization model according to vegetable knowledge mapping;From institute
Acquisition training sample corpus in training corpus is stated, by the stroke feature information input of the training sample corpus to initialization
It is trained in vegetable vectorization model;Vegetable vectorization matrix is obtained according to training result.
Optionally, the acquisition module is further adapted for:
The corresponding stroke feature information of the menu name of each part vegetable to be recommended is reflected according to the vegetable vectorization matrix
Penetrate into the vegetable vector of the vegetable to be recommended.
Optionally, described device further include:
Second training module, suitable for obtaining the vegetable sample vector of more parts of vegetable samples;And the multiple behavior of ordering of acquisition
The scene sample attribute information of user's sample attribute information of corresponding user's sample of ordering and scene of ordering;By user's sample
This attribute information and scene sample attribute information MAP are at corresponding sample attribute information vector;It is right according to behavior of repeatedly ordering
The sample attribute information vector and the degree of association of the vegetable sample vector are labeled, and obtain degree of association annotation results;It will
The vegetable sample vector and the sample attribute information vector are input in degree of association training pattern, obtain degree of association training mould
The degree of association of type output exports result;Using the loss between the degree of association annotation results and degree of association output result, to pass
Connection degree training pattern is trained, and obtains the related degree model.
Optionally, second training module is further adapted for:
According to the vegetable vectorization matrix by the corresponding stroke feature information MAP of the menu name of each part vegetable sample
At the vegetable vector of the vegetable sample.
Optionally, second training module is further adapted for:
Lower single vegetable that the behavior of ordering every time includes is carried out with vegetable sample corresponding, obtains lower single vegetable and vegetable sample
Corresponding relationship;
It is carried out according to the degree of association of the corresponding relationship to the sample attribute information vector and the vegetable sample vector
Mark.
Optionally, second training module is further adapted for:
It is according to corresponding relationship, the vegetable sample vector for the vegetable sample for including in the behavior of ordering is corresponding with the behavior of ordering
The degree of association of sample attribute information vector be labeled as first degree of association;
And/or by the vegetable sample vector for the vegetable sample not included in the behavior of ordering sample corresponding with the behavior of ordering
The degree of association of this attribute information vector is labeled as second degree of association.
Optionally, the recommending module is further adapted for: by more portions of vegetables to be recommended according to association angle value from high to low into
Row sequence, the vegetable to be recommended of the forward preset quantity of selected and sorted is recommended to target to use as recommendation vegetable from ranking results
Family.
According to another aspect of the invention, provide a kind of calculating equipment, comprising: processor, memory, communication interface and
Communication bus, the processor, the memory and the communication interface complete mutual communication by the communication bus;
For the memory for storing an at least executable instruction, it is above-mentioned that the executable instruction executes the processor
The corresponding operation of vegetable recommended method.
In accordance with a further aspect of the present invention, provide a kind of computer storage medium, be stored in the storage medium to
A few executable instruction, the executable instruction make processor execute such as the corresponding operation of above-mentioned vegetable recommended method.
Vegetable recommended method according to the present invention and device utilize the category of trained related degree model prediction target user
The angle value that is associated with of property information vector and the vegetable vector of each part vegetable to be recommended, obtains target user in current recommendation scene
To the preference of each part vegetable to be recommended, the history for rather than relying on target user orders behavior prediction target user to each part
The preference of vegetable to be recommended, this mode can accurately carry out the pre- of preference to the behavior of ordering target user not abundant
It surveys;Then, vegetable recommendation is carried out according to the association angle value that prediction obtains, uses the recommendation vegetable for recommending target user and target
The matching degree of the scene properties information of the customer attribute information at family and current recommendation scene is high, recommends the same of efficiency improving
When, improve the experience of ordering of user.
The above description is only an overview of the technical scheme of the present invention, in order to better understand the technical means of the present invention,
And it can be implemented in accordance with the contents of the specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, the followings are specific embodiments of the present invention.
Detailed description of the invention
By reading the following detailed description of the preferred embodiment, various other advantages and benefits are common for this field
Technical staff will become clear.The drawings are only for the purpose of illustrating a preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings:
Fig. 1 shows the flow chart of vegetable recommended method according to an embodiment of the invention;
Fig. 2 shows the flow charts of vegetable recommended method in accordance with another embodiment of the present invention;
Fig. 3 shows the flow chart of the training process of related degree model in a specific embodiment of the invention;
Fig. 4 shows the functional block diagram of vegetable recommendation apparatus according to an embodiment of the invention;
Fig. 5 shows a kind of structural schematic diagram for calculating equipment according to an embodiment of the present invention.
Specific embodiment
Exemplary embodiments of the present disclosure are described in more detail below with reference to accompanying drawings.Although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
It is fully disclosed to those skilled in the art.
Fig. 1 shows the flow chart of vegetable recommended method according to an embodiment of the invention.As shown in Figure 1, this method
Include:
Step S101, by the customer attribute information of target user and scene properties information MAP at corresponding attribute information to
Amount.
Wherein, target user refers to the user for currently needing that vegetable recommendation is carried out for it;Customer attribute information includes and user
It draws a portrait relevant information, for example, the information such as gender, age;Scene properties information includes letter relevant to current recommendation scene
Breath, for example, recommending the information such as period, festivals or holidays.
In this step, the customer attribute information and scene properties information of vector characterization target user are utilized.In the present invention,
The concrete mode of customer attribute information and scene properties information vectorization characterization is not limited, those skilled in the art should understand that
It is that arbitrarily can be used for being all contained in the scope of the present invention the method that attribute information carries out vectorization characterization.
Step S102 obtains the vegetable vector of more portions of vegetables to be recommended.
Wherein, more portions of vegetables to be recommended match with current recommendation scene, specifically can be according to the page where target user
Face position is determined;Optionally, if target user clicks to enter specific shop, more portions of vegetables to be recommended are the specific shop
All or part of vegetable;It is corresponding according to the page location where target user if target user does not enter any shop
Popularization demand determines more portions of vegetables to be recommended.
Specifically, it is determined that obtaining this more parts wait push away after the more portions of vegetables to be recommended to match with current recommendation scene
Recommend the vegetable vector of vegetable.The present invention is not specifically limited the mode for obtaining vegetable vector, and optionally, which can be with
Directly got from other channels, or generated using vegetable vectorization method.
The vegetable vector of each part of attribute information vector sum vegetable to be recommended is input to the association that training obtains by step S103
It spends in model, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value.
Specifically, predict to obtain the vegetable vector of attribute information vector and each part vegetable to be recommended using related degree model
Angle value is associated with to get target user is arrived under current recommendation scene to the preference of each part vegetable to be recommended, makes the prediction simultaneously
The process of preference is obtained to order behavioural information also not dependent on the history of target user.
Step S104 according to association angle value selects that vegetable is recommended to recommend to target user from more portions of vegetables to be recommended.
Specifically, vegetable recommendation is carried out according to association angle value, makes the recommendation vegetable for recommending target user and target user
Customer attribute information and current recommendations scene scene properties information matching degree it is high, improving the same of recommendation efficiency
When, improve the experience of ordering of user.
According to vegetable recommended method provided in this embodiment, the category of trained related degree model prediction target user is utilized
The angle value that is associated with of property information vector and the vegetable vector of each part vegetable to be recommended, obtains target user in current recommendation scene
To the preference of each part vegetable to be recommended, the history for rather than relying on target user orders behavior prediction target user to each part
The preference of vegetable to be recommended, this mode can accurately carry out the pre- of preference to the behavior of ordering target user not abundant
It surveys;Then, vegetable recommendation is carried out according to the association angle value that prediction obtains, uses the recommendation vegetable for recommending target user and target
The matching degree of the scene properties information of the customer attribute information at family and current recommendation scene is high, recommends the same of efficiency improving
When, improve the experience of ordering of user.
Fig. 2 shows the flow charts of vegetable recommended method in accordance with another embodiment of the present invention.As shown in Fig. 2, the party
Method includes:
Step S201, training obtain vegetable vectorization matrix.
In the present embodiment, the vegetable vector of more portions of vegetables to be recommended, and the more portions of dishes for training related degree model
The vegetable sample vector of product sample is obtained by vegetable vectorization matrix.Before this, training obtains vegetable vectorization square first
Battle array.
Wherein, vegetable vectorization matrix can instruct the corresponding text feature of menu name using vegetable vectorization model
It gets;Wherein, text feature includes but is not limited to radical, character element, and/or stroke feature.Below only with stroke
Feature is example, illustrates that training obtains the process of vegetable vectorization matrix, but the present invention is not limited with this example:
Specifically, the training corpus of vegetable vectorization model is constructed according to vegetable knowledge mapping;From training corpus
Obtain training sample corpus, by the stroke feature information input of training sample corpus to initialization vegetable vectorization model in into
Row training;Vegetable vectorization matrix is obtained according to training result.Further, by the vegetable knowledge mapping to structuring
Original language material carries out the processing such as segmenting, and obtains non-structured corpus data, and construct training corpus;Choose preset quantity
Corpus is as training sample corpus, and using the stroke feature information of the centre word in each training sample corpus as vegetable vector
Change model training input, and using the term vector of the cliction up and down of centre word as the training of vegetable vectorization model export into
Row training;Vegetable vectorization matrix is obtained according to the parameter of the vegetable vectorization model at the end of training.
Step S202 believes the corresponding stroke feature of the menu name of each part vegetable to be recommended according to vegetable vectorization matrix
Breath is mapped to the vegetable vector of vegetable to be recommended.
By the corresponding stroke feature information of the menu name of each part vegetable to be recommended and vegetable vectorization matrix multiple, mapping
Obtain term vector corresponding with the menu name of the vegetable to be recommended, i.e., the vegetable vector of vegetable to be recommended.
Step S203, by the customer attribute information of target user and scene properties information MAP at corresponding attribute information to
Amount.
Wherein, customer attribute information includes information relevant to user's portrait, optionally, these and user's representation data phase
The information of pass is specially the taste for influencing user, the information of item ingredients preference.For example, the household register of user, it can to a certain degree
The upper taste preference for influencing user.Wherein, scene properties information includes information relevant to current recommendation scene, optionally,
These information relevant to current recommendation scene are to influence the information of the vegetable selection of user.For example, different season or section
Day, the vegetable that user may select is different, alternatively, working day and nonworkdays, the vegetable that user may select is also not
With.
Specifically, account information or account letter in the platform for ordering service are currently being provided according to target user
Corresponding related information or real-name authentication information are ceased, from the user property for obtaining target user in the cooperation channel of platform or platform
Information;And scene properties information is determined according to data such as current date, time and weather;It is obtained with vector characterization is above-mentioned
The customer attribute information and scene properties information of the target user arrived.Further, customer attribute information and scene properties information
It may each comprise the information of multiple dimensions, and when being mapped to corresponding attribute information vector, then the information of each dimension is reflected
Penetrate into corresponding attribute information vector.
For example, the information of three dimensions belongs to as user with choosing gender, age and the household register of target user
Property information, and choose current period, whether be working day and festival information three dimensions information as scene category
The information of above-mentioned six dimensions is mapped to the vector of default dimension then when being characterized with vector by property information respectively.
The vegetable vector of each part of attribute information vector sum vegetable to be recommended is input to the association that training obtains by step S204
It spends in model, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value.
Wherein, related degree model utilizes the customer attribute information of other users (may also comprise the target user) and orders
Behavioural information is trained to obtain, and the history independent of the target user is ordered, behavioural information is trained.Fig. 3 is shown
The flow chart of the training process of related degree model in a specific embodiment of the invention.As shown in figure 3, the training process includes:
Step S301 obtains the vegetable sample vector of more parts of vegetable samples.
Specifically, the corresponding stroke feature information of the menu name of each part vegetable sample is reflected according to vegetable vectorization matrix
Penetrate into the vegetable sample vector of vegetable sample.Further, the corresponding stroke feature of the menu name of each part vegetable sample is believed
Breath and vegetable vectorization matrix multiple, mapping obtain vector corresponding with the menu name of the vegetable sample, i.e. vegetable sample
Vegetable sample vector.
Step S302, user's sample attribute information of the multiple corresponding user's sample of ordering of behavior of ordering of acquisition and field of ordering
The scene sample attribute information of scape;By user's sample attribute information and scene sample attribute information MAP at corresponding sample attribute
Information vector.
Specifically, order user and scene of ordering are determined according to the corresponding data of ordering of each behavior of ordering.Optionally,
According to account or the payment account of ordering, user's sample attribute information is obtained into corresponding channel, for example, according to Alipay account
Obtain the corresponding user's representation data of the account.According to date of ordering, time and weather, the corresponding field of behavior of ordering is determined
Scape sample attribute information.Then, with vector characterization each the order corresponding user's sample attribute information of behavior and scene sample category
Property information, obtain the sample attribute information vector of corresponding behavior sample of ordering every time.
Step S303, according to repeatedly ordering behavior, to the degree of association of sample attribute information vector and vegetable sample vector into
Rower note, obtains degree of association annotation results.
Specifically, the degree of association marks, i.e. each corresponding user that orders of behavior that orders of mark is corresponding in the behavior of ordering
To the selection situation of each part vegetable sample under scene of ordering.And the order channel or not different due to the behavior source of ordering of acquisition
Same shop, then there may be the differences in vegetable name, correspondingly, to sample attribute information vector and vegetable sample vector
Before the degree of association is labeled, lower single vegetable that the behavior of ordering every time includes is carried out with vegetable sample corresponding, obtains lower list dish
The corresponding relationship of product and vegetable sample.Further according to corresponding relationship to the degree of association of sample attribute information vector and vegetable sample vector
It is labeled.
Further, according to corresponding relationship, by the vegetable sample vector for the vegetable sample for including in the behavior of ordering and the point
The degree of association of the corresponding sample attribute information vector of meal behavior is labeled as first degree of association;And/or it will not include in the behavior of ordering
The degree of association of vegetable sample vector sample attribute information vector corresponding with the behavior of ordering of vegetable sample be labeled as second
The degree of association.For example, order in behavior comprising lower single vegetable a, b and c, and a, b and c respectively with vegetable sample A, B and C couple
It answers, then by the degree of association of the vegetable sample vector of vegetable sample A, B and C sample attribute information vector corresponding with the behavior of ordering
It is labeled as 1 (first degree of association), and by the vegetable sample vector sample corresponding with the behavior of ordering of vegetable remaining in vegetable sample
The degree of association of this attribute information vector is labeled as 0 (second degree of association).
Vegetable sample vector and sample attribute information vector are input in degree of association training pattern, obtain by step S304
The degree of association of degree of association training pattern output exports result.
Specifically, using vegetable sample vector and sample attribute information vector as the training input number of degree of association training pattern
According to, and using corresponding sample attribute information vector and the degree of association annotation results of vegetable sample vector as degree of association training mould
The training output data of type.And in the training process, corresponding every group of trained input data (i.e. vegetable sample vector and sample attribute
Information vector), the degree of association output result of degree of association training pattern reality output can be obtained.
Step S305, using the loss between degree of association annotation results and degree of association output result, to degree of association training mould
Type is trained, and obtains related degree model.
Specifically, it is exported between result and degree of association annotation results by the degree of association that loss function calculates reality output
Loss, and adaptive learning is carried out according to the loss, until loss is reduced in default loss range, then deconditioning, obtains
Related degree model, training parameter, that is, related degree model model parameter when deconditioning.
More portions of vegetables to be recommended are ranked up from high to low according to association angle value, select from ranking results by step S205
The vegetable to be recommended for selecting the forward preset quantity that sorts is recommended as recommendation vegetable to target user.
Specifically, vegetable recommendation is carried out according to association angle value, makes the recommendation vegetable for recommending target user and target user
Customer attribute information and current recommendations scene scene properties information matching degree it is high, improving the same of recommendation efficiency
When, improve the experience of ordering of user.
For convenient for this implementation implementation process and using this implementation scheme technical effect understanding, below with one
A specific application example illustrates the specific implementation process of this implementation: the vegetable vectorization matrix and degree of association mould that training is obtained
Type is applied in meal ordering system, when detect user 1 entered in the A of shop by the meal ordering system order food when, get shop
The menu name for 50 portions of vegetables for including in the menu that paving A is provided is obtained by vegetable vectorization matrix by menu name vectorization
To the vegetable vector of corresponding 50 portions of vegetables;Meanwhile obtaining user's 1 according to the user account that user 1 inputs in meal ordering system
Gender, age and household register etc. customer attribute informations, and ordered time, festivals or holidays etc. according to current scene determination of ordering
Scene properties information, and each of these customer attribute informations and scene properties information item of information are mapped to corresponding category
Property information vector, if obtain totally 10 attribute informations to Li Ang;10 attribute information vector sums are corresponded to the dish of every portion of vegetable
Product vector, which is input in related degree model, to be predicted, obtain the vegetable vector and attribute information vector is associated with angle value;According to
Predict obtained 50 association angle value of 50 portions of vegetables of correspondence, therefrom determine corresponding 5 vegetables of maximum 5 associations angle value to
Amount, and determine the menu name of the corresponding 5 portions of vegetables of 5 vegetable vectors, then by the corresponding menu name of 5 portions of vegetables and
Its corresponding vegetable picture presentation the page where the A of shop predeterminated position, when ordering as user selection reference.It can
See, in the application example, recommends the process of vegetable with reference to the personalized factor and scene factor of user, so that recommending
The vegetable of user more meets the demand of ordering of user;And the process that the vegetable is recommended, the history independent of user are ordered row
For, so that it is few or without the user of data of ordering for data of ordering, it also can be carried out accurate vegetable and recommend.
According to vegetable recommended method provided in this embodiment, instructed according to the corresponding text feature information of menu name
Practice, obtains vegetable vectorization matrix;Using the vegetable vector of the available vegetable to be recommended of vegetable vectorization matrix, and
To the vegetable sample vector for training the vegetable sample of related degree model;Also, pass through the corresponding point of existing behavior of ordering
Meal data are trained, and obtain related degree model, are then believed using the attribute of trained related degree model prediction target user
Breath vector is associated with angle value with the vegetable vector of each part vegetable to be recommended, obtains target user in current recommendation scene to each
The preference of vegetable part to be recommended, rather than rely on target user history order behavior prediction target user to each part wait push away
The preference of vegetable is recommended, this mode can be accurate to arbitrary target user (including behavior of ordering target user not abundant)
Carry out the prediction of preference;Then, vegetable recommendation is carried out according to the obtained association angle value of prediction, makes to recommend target user's
Recommend the matching degree of the customer attribute information of vegetable and target user and the scene properties information of current recommendation scene high,
While improving recommendation efficiency, the experience of ordering of user is improved.
Fig. 4 shows the functional block diagram of vegetable recommendation apparatus according to an embodiment of the invention.As shown in figure 4, the dress
Setting includes: mapping block 401, obtains module 402, prediction module 403, recommending module 404;Optionally, the device further include: the
One training module 405, the second training module 406.
Mapping block 401, suitable for by the customer attribute information of target user and scene properties information MAP at corresponding category
Property information vector;
Module 402 is obtained, suitable for obtaining the vegetable vector of more portions of vegetables to be recommended;
Prediction module 403, suitable for the vegetable vector of described each part of attribute information vector sum vegetable to be recommended is input to instruction
In the related degree model got, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value;
Recommending module 404, suitable for selected from more portions of vegetables to be recommended according to the association angle value recommend vegetable recommend to
Target user.
In a kind of optional embodiment, described device further include:
First training module 405, suitable for constructing the training corpus of vegetable vectorization model according to vegetable knowledge mapping;From
Training sample corpus is obtained in the training corpus, by the stroke feature information input of the training sample corpus to initialization
Vegetable vectorization model in be trained;Vegetable vectorization matrix is obtained according to training result.
In a kind of optional embodiment, the acquisition module 402 is further adapted for:
The corresponding stroke feature information of the menu name of each part vegetable to be recommended is reflected according to the vegetable vectorization matrix
Penetrate into the vegetable vector of the vegetable to be recommended.
In a kind of optional embodiment, described device further include:
Second training module 406, suitable for obtaining the vegetable sample vector of more parts of vegetable samples;And acquisition is repeatedly ordered
The scene sample attribute information of user's sample attribute information of the corresponding user's sample of ordering of behavior and scene of ordering;By the use
Family sample attribute information and scene sample attribute information MAP are at corresponding sample attribute information vector;According to row of repeatedly ordering
To be labeled to the sample attribute information vector and the degree of association of the vegetable sample vector, obtaining degree of association mark knot
Fruit;The vegetable sample vector and the sample attribute information vector are input in degree of association training pattern, the degree of association is obtained
The degree of association of training pattern output exports result;Utilize the damage between the degree of association annotation results and degree of association output result
It loses, degree of association training pattern is trained, the related degree model is obtained.
In a kind of optional embodiment, second training module 406 is further adapted for:
According to the vegetable vectorization matrix by the corresponding stroke feature information MAP of the menu name of each part vegetable sample
At the vegetable vector of the vegetable sample.
In a kind of optional embodiment, second training module 406 is further adapted for:
Lower single vegetable that the behavior of ordering every time includes is carried out with vegetable sample corresponding, obtains lower single vegetable and vegetable sample
Corresponding relationship;
It is carried out according to the degree of association of the corresponding relationship to the sample attribute information vector and the vegetable sample vector
Mark.
In a kind of optional embodiment, second training module 406 is further adapted for:
It is according to corresponding relationship, the vegetable sample vector for the vegetable sample for including in the behavior of ordering is corresponding with the behavior of ordering
The degree of association of sample attribute information vector be labeled as first degree of association;
And/or by the vegetable sample vector for the vegetable sample not included in the behavior of ordering sample corresponding with the behavior of ordering
The degree of association of this attribute information vector is labeled as second degree of association.
In a kind of optional embodiment, the recommending module 404 is further adapted for: by more portions of vegetables to be recommended according to
Association angle value is ranked up from high to low, and the vegetable to be recommended of the forward preset quantity of selected and sorted is used as and pushes away from ranking results
Vegetable is recommended to recommend to target user.
It can refer to the description of corresponding steps in embodiment of the method about the specific structure and working principle of above-mentioned modules,
Details are not described herein again.
The embodiment of the present application provides a kind of nonvolatile computer storage media, and the computer storage medium is stored with
The vegetable recommended method in above-mentioned any means embodiment can be performed in an at least executable instruction, the computer executable instructions.
Fig. 5 shows a kind of structural schematic diagram for calculating equipment according to an embodiment of the present invention, the specific embodiment of the invention
The specific implementation for calculating equipment is not limited.
As shown in figure 5, the calculating equipment may include: processor (processor) 502, communication interface
(Communications Interface) 504, memory (memory) 506 and communication bus 508.
Wherein:
Processor 502, communication interface 504 and memory 506 complete mutual communication by communication bus 508.
Communication interface 504, for being communicated with the network element of other equipment such as client or other servers etc..
Processor 502 can specifically execute the correlation in above-mentioned vegetable recommended method embodiment for executing program 510
Step.
Specifically, program 510 may include program code, which includes computer operation instruction.
Processor 502 may be central processor CPU or specific integrated circuit ASIC (Application
Specific Integrated Circuit), or be arranged to implement the integrated electricity of one or more of the embodiment of the present invention
Road.The one or more processors that equipment includes are calculated, can be same type of processor, such as one or more CPU;It can also
To be different types of processor, such as one or more CPU and one or more ASIC.
Memory 506, for storing program 510.Memory 506 may include high speed RAM memory, it is also possible to further include
Nonvolatile memory (non-volatile memory), for example, at least a magnetic disk storage.
Program 510 specifically can be used for so that processor 502 executes following operation:
By the customer attribute information of target user and scene properties information MAP at corresponding attribute information vector;
Obtain the vegetable vector of more portions of vegetables to be recommended;
The vegetable vector of described each part of attribute information vector sum vegetable to be recommended is input to the degree of association mould that training obtains
In type, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value;
It is selected that vegetable is recommended to recommend to target user from more portions of vegetables to be recommended according to the association angle value.
In a kind of optional embodiment, program 510 can specifically be further used for so that processor 502 execute it is following
Operation:
The training corpus of vegetable vectorization model is constructed according to vegetable knowledge mapping;
Training sample corpus is obtained from the training corpus, the stroke feature information of the training sample corpus is defeated
Enter into the vegetable vectorization model of initialization and is trained;Vegetable vectorization matrix is obtained according to training result.
In a kind of optional embodiment, program 510 can specifically be further used for so that processor 502 execute it is following
Operation:
The corresponding stroke feature information of the menu name of each part vegetable to be recommended is reflected according to the vegetable vectorization matrix
Penetrate into the vegetable vector of the vegetable to be recommended.
In a kind of optional embodiment, program 510 can specifically be further used for so that processor 502 execute it is following
Operation:
Obtain the vegetable sample vector of more parts of vegetable samples;And the multiple corresponding user's sample of ordering of behavior of ordering of acquisition
The scene sample attribute information of this user's sample attribute information and scene of ordering;By user's sample attribute information and scene
Sample attribute information MAP is at corresponding sample attribute information vector;
According to behavior of repeatedly ordering, the degree of association of the sample attribute information vector and the vegetable sample vector is carried out
Mark, obtains degree of association annotation results;
The vegetable sample vector and the sample attribute information vector are input in degree of association training pattern, closed
The degree of association of connection degree training pattern output exports result;
Using the loss between the degree of association annotation results and degree of association output result, degree of association training pattern is carried out
Training, obtains the related degree model.
In a kind of optional embodiment, program 510 can specifically be further used for so that processor 502 execute it is following
Operation:
According to the vegetable vectorization matrix by the corresponding stroke feature information MAP of the menu name of each part vegetable sample
At the vegetable vector of the vegetable sample.
In a kind of optional embodiment, program 510 can specifically be further used for so that processor 502 execute it is following
Operation:
Lower single vegetable that the behavior of ordering every time includes is carried out with vegetable sample corresponding, obtains lower single vegetable and vegetable sample
Corresponding relationship;
It is carried out according to the degree of association of the corresponding relationship to the sample attribute information vector and the vegetable sample vector
Mark.
In a kind of optional embodiment, program 510 can specifically be further used for so that processor 502 execute it is following
Operation:
It is according to corresponding relationship, the vegetable sample vector for the vegetable sample for including in the behavior of ordering is corresponding with the behavior of ordering
The degree of association of sample attribute information vector be labeled as first degree of association;
And/or by the vegetable sample vector for the vegetable sample not included in the behavior of ordering sample corresponding with the behavior of ordering
The degree of association of this attribute information vector is labeled as second degree of association.
In a kind of optional embodiment, program 510 can specifically be further used for so that processor 502 execute it is following
Operation:
More portions of vegetables to be recommended are ranked up from high to low according to association angle value, selected and sorted is forward from ranking results
Preset quantity vegetable to be recommended as recommend vegetable recommend to target user.
Algorithm and display are not inherently related to any particular computer, virtual system, or other device provided herein.
Various general-purpose systems can also be used together with teachings based herein.As described above, it constructs required by this kind of system
Structure be obvious.In addition, the present invention is also not directed to any particular programming language.It should be understood that can use various
Programming language realizes summary of the invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this specification.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of the various inventive aspects,
Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the disclosed method should not be interpreted as reflecting the following intention: i.e. required to protect
Shield the present invention claims features more more than feature expressly recited in each claim.More precisely, as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, wherein each claim itself
All as a separate embodiment of the present invention.
Those skilled in the art will understand that can be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more devices different from this embodiment.It can be the module or list in embodiment
Member or component are combined into a module or unit or component, and furthermore they can be divided into multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it can use any
Combination is to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so disclosed
All process or units of what method or apparatus are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Benefit require, abstract and attached drawing) disclosed in each feature can carry out generation with an alternative feature that provides the same, equivalent, or similar purpose
It replaces.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments mean it is of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
Meaning one of can in any combination mode come using.
Various component embodiments of the invention can be implemented in hardware, or to run on one or more processors
Software module realize, or be implemented in a combination thereof.It will be understood by those of skill in the art that can be used in practice
Microprocessor or digital signal processor (DSP) come realize some in vegetable recommendation apparatus according to an embodiment of the present invention or
The some or all functions of person's whole component.The present invention is also implemented as one for executing method as described herein
Point or whole device or device programs (for example, computer program and computer program product).Such this hair of realization
Bright program can store on a computer-readable medium, or may be in the form of one or more signals.It is such
Signal can be downloaded from an internet website to obtain, and is perhaps provided on the carrier signal or is provided in any other form.
It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and ability
Field technique personnel can be designed alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between parentheses should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" located in front of the element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be through the same hardware branch
To embody.The use of word first, second, and third does not indicate any sequence.These words can be explained and be run after fame
Claim.
Claims (10)
1. a kind of vegetable recommended method, comprising:
By the customer attribute information of target user and scene properties information MAP at corresponding attribute information vector;
Obtain the vegetable vector of more portions of vegetables to be recommended;
The vegetable vector of described each part of attribute information vector sum vegetable to be recommended is input in the related degree model that training obtains,
Obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value;
It is selected that vegetable is recommended to recommend to target user from more portions of vegetables to be recommended according to the association angle value.
2. according to the method described in claim 1, wherein, before the vegetable vector for obtaining more portions of vegetables to be recommended, institute
State method further include:
The training corpus of vegetable vectorization model is constructed according to vegetable knowledge mapping;
Training sample corpus is obtained from the training corpus, extremely by the stroke feature information input of the training sample corpus
It is trained in the vegetable vectorization model of initialization;Vegetable vectorization matrix is obtained according to training result.
3. according to the method described in claim 2, wherein, the vegetable vector for obtaining more portions of vegetables to be recommended further wraps
It includes:
According to the vegetable vectorization matrix by the corresponding stroke feature information MAP of the menu name of each part vegetable to be recommended at
The vegetable vector of the vegetable to be recommended.
4. according to the method described in claim 2, wherein, the related degree model is obtained by following steps training:
Obtain the vegetable sample vector of more parts of vegetable samples;And the multiple corresponding user's sample of ordering of behavior of ordering of acquisition
The scene sample attribute information of user's sample attribute information and scene of ordering;By user's sample attribute information and scene sample
Attribute information is mapped to corresponding sample attribute information vector;
According to behavior of repeatedly ordering, the sample attribute information vector and the degree of association of the vegetable sample vector are marked
Note, obtains degree of association annotation results;
The vegetable sample vector and the sample attribute information vector are input in degree of association training pattern, the degree of association is obtained
The degree of association of training pattern output exports result;
Using the loss between the degree of association annotation results and degree of association output result, degree of association training pattern is instructed
Practice, obtains the related degree model.
5. according to the method described in claim 4, wherein, the vegetable sample vector for obtaining more parts of vegetable samples further wraps
It includes:
According to the vegetable vectorization matrix by the corresponding stroke feature information MAP of the menu name of each part vegetable sample at institute
State the vegetable vector of vegetable sample.
6. according to the method described in claim 4, wherein, described to the sample attribute information vector and the vegetable sample
Before the degree of association of vector is labeled, the method also includes:
Lower single vegetable that the behavior of ordering every time includes is carried out with vegetable sample corresponding, obtains pair of lower list vegetable and vegetable sample
It should be related to;
It is described that the sample attribute information vector and the degree of association of the vegetable sample vector are labeled specifically: according to institute
Corresponding relationship is stated to be labeled the sample attribute information vector and the degree of association of the vegetable sample vector.
7. according to the method described in claim 6, wherein, it is described according to the corresponding relationship to the sample attribute information to
It measures to be labeled with the degree of association of the vegetable sample vector and further comprises:
According to corresponding relationship, by the vegetable sample vector for the vegetable sample for including in the behavior of ordering sample corresponding with the behavior of ordering
The degree of association of this attribute information vector is labeled as first degree of association;
And/or by the vegetable sample vector for the vegetable sample not included in the behavior of ordering sample category corresponding with the behavior of ordering
The degree of association of property information vector is labeled as second degree of association.
8. a kind of vegetable recommendation apparatus, comprising:
Mapping block, suitable for by the customer attribute information of target user and scene properties information MAP at corresponding attribute information to
Amount;
Module is obtained, suitable for obtaining the vegetable vector of more portions of vegetables to be recommended;
Prediction module, suitable for the vegetable vector of described each part of attribute information vector sum vegetable to be recommended is input to what training obtained
In related degree model, obtain attribute information vector and the vegetable vector of each part vegetable to be recommended is associated with angle value;
Recommending module recommends vegetable to recommend to target to use suitable for being selected from more portions of vegetables to be recommended according to the association angle value
Family.
9. a kind of calculating equipment, comprising: processor, memory, communication interface and communication bus, the processor, the storage
Device and the communication interface complete mutual communication by the communication bus;
The memory executes the processor as right is wanted for storing an at least executable instruction, the executable instruction
Ask the corresponding operation of vegetable recommended method described in any one of 1-7.
10. a kind of computer storage medium, an at least executable instruction, the executable instruction are stored in the storage medium
Processor is set to execute such as the corresponding operation of vegetable recommended method of any of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811069258.1A CN109300059B (en) | 2018-09-13 | 2018-09-13 | Dish recommending method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811069258.1A CN109300059B (en) | 2018-09-13 | 2018-09-13 | Dish recommending method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109300059A true CN109300059A (en) | 2019-02-01 |
CN109300059B CN109300059B (en) | 2020-11-17 |
Family
ID=65167000
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811069258.1A Active CN109300059B (en) | 2018-09-13 | 2018-09-13 | Dish recommending method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109300059B (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110135646A (en) * | 2019-05-20 | 2019-08-16 | 梁志鹏 | The method, apparatus quickly served and storage medium are estimated in a kind of dining room |
CN110162694A (en) * | 2019-04-02 | 2019-08-23 | 莫毓昌 | A kind of recommender system and method based on pairs of correlation rule |
CN110782308A (en) * | 2019-09-20 | 2020-02-11 | 北京三快在线科技有限公司 | Pushing method and device for recommended package, electronic equipment and readable storage medium |
CN110837552A (en) * | 2019-09-30 | 2020-02-25 | 口口相传(北京)网络技术有限公司 | Diet information recommendation method and device |
CN111563788A (en) * | 2020-03-26 | 2020-08-21 | 口碑(上海)信息技术有限公司 | Object recommendation method and device |
CN111666418A (en) * | 2020-04-23 | 2020-09-15 | 北京三快在线科技有限公司 | Text regeneration method and device, electronic equipment and computer readable medium |
CN111695960A (en) * | 2019-03-12 | 2020-09-22 | 阿里巴巴集团控股有限公司 | Object recommendation system, method, electronic device and storage medium |
CN112016582A (en) * | 2019-05-31 | 2020-12-01 | 口口相传(北京)网络技术有限公司 | Dish recommending method and device |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103412936A (en) * | 2013-08-21 | 2013-11-27 | 五八同城信息技术有限公司 | Dish recommendation system based on data mining and cloud computing service |
CN104133901A (en) * | 2014-08-04 | 2014-11-05 | 上海巨浪信息科技有限公司 | Mobile catering dish ordering and intelligent recommending system |
CN106548006A (en) * | 2016-10-09 | 2017-03-29 | 浙江大学 | A kind of meals based on user's typical case's taste recommend method |
CN107590246A (en) * | 2017-09-15 | 2018-01-16 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
CN107665254A (en) * | 2017-09-30 | 2018-02-06 | 济南浪潮高新科技投资发展有限公司 | A kind of menu based on deep learning recommends method |
CN107862542A (en) * | 2017-09-28 | 2018-03-30 | 北京三快在线科技有限公司 | A kind of vegetable recommends method and apparatus |
CN107886348A (en) * | 2017-09-30 | 2018-04-06 | 厦门快商通信息技术有限公司 | A kind of artificial intelligence method of ordering and system |
CN107886400A (en) * | 2017-11-15 | 2018-04-06 | 维沃移动通信有限公司 | A kind of recipe recommendation method, device and mobile terminal |
CN108492144A (en) * | 2018-03-30 | 2018-09-04 | 百度在线网络技术(北京)有限公司 | Method and system, terminal and the computer readable storage medium that vegetable is recommended |
-
2018
- 2018-09-13 CN CN201811069258.1A patent/CN109300059B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103412936A (en) * | 2013-08-21 | 2013-11-27 | 五八同城信息技术有限公司 | Dish recommendation system based on data mining and cloud computing service |
CN104133901A (en) * | 2014-08-04 | 2014-11-05 | 上海巨浪信息科技有限公司 | Mobile catering dish ordering and intelligent recommending system |
CN106548006A (en) * | 2016-10-09 | 2017-03-29 | 浙江大学 | A kind of meals based on user's typical case's taste recommend method |
CN107590246A (en) * | 2017-09-15 | 2018-01-16 | 百度在线网络技术(北京)有限公司 | Method and apparatus for pushed information |
CN107862542A (en) * | 2017-09-28 | 2018-03-30 | 北京三快在线科技有限公司 | A kind of vegetable recommends method and apparatus |
CN107665254A (en) * | 2017-09-30 | 2018-02-06 | 济南浪潮高新科技投资发展有限公司 | A kind of menu based on deep learning recommends method |
CN107886348A (en) * | 2017-09-30 | 2018-04-06 | 厦门快商通信息技术有限公司 | A kind of artificial intelligence method of ordering and system |
CN107886400A (en) * | 2017-11-15 | 2018-04-06 | 维沃移动通信有限公司 | A kind of recipe recommendation method, device and mobile terminal |
CN108492144A (en) * | 2018-03-30 | 2018-09-04 | 百度在线网络技术(北京)有限公司 | Method and system, terminal and the computer readable storage medium that vegetable is recommended |
Non-Patent Citations (1)
Title |
---|
曾强: "个性化菜品推荐系统的研究与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111695960A (en) * | 2019-03-12 | 2020-09-22 | 阿里巴巴集团控股有限公司 | Object recommendation system, method, electronic device and storage medium |
CN110162694A (en) * | 2019-04-02 | 2019-08-23 | 莫毓昌 | A kind of recommender system and method based on pairs of correlation rule |
CN110135646A (en) * | 2019-05-20 | 2019-08-16 | 梁志鹏 | The method, apparatus quickly served and storage medium are estimated in a kind of dining room |
CN110135646B (en) * | 2019-05-20 | 2021-07-20 | 梁志鹏 | Restaurant pre-estimation quick serving method and device and storage medium |
CN112016582A (en) * | 2019-05-31 | 2020-12-01 | 口口相传(北京)网络技术有限公司 | Dish recommending method and device |
CN112016582B (en) * | 2019-05-31 | 2023-11-24 | 口口相传(北京)网络技术有限公司 | Dish recommending method and device |
CN110782308A (en) * | 2019-09-20 | 2020-02-11 | 北京三快在线科技有限公司 | Pushing method and device for recommended package, electronic equipment and readable storage medium |
CN110837552A (en) * | 2019-09-30 | 2020-02-25 | 口口相传(北京)网络技术有限公司 | Diet information recommendation method and device |
CN111563788A (en) * | 2020-03-26 | 2020-08-21 | 口碑(上海)信息技术有限公司 | Object recommendation method and device |
CN111666418A (en) * | 2020-04-23 | 2020-09-15 | 北京三快在线科技有限公司 | Text regeneration method and device, electronic equipment and computer readable medium |
CN111666418B (en) * | 2020-04-23 | 2024-01-16 | 北京三快在线科技有限公司 | Text regeneration method, device, electronic equipment and computer readable medium |
Also Published As
Publication number | Publication date |
---|---|
CN109300059B (en) | 2020-11-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109300059A (en) | Vegetable recommended method and device | |
US10049466B2 (en) | Color name generation from images and color palettes | |
CN104239571B (en) | It is a kind of to carry out using the method and apparatus recommended | |
US11727459B2 (en) | Search query-based replacement part interface | |
CN106294743A (en) | The recommendation method and device of application function | |
CN108648059A (en) | Order recommends method and device | |
CN110209926A (en) | Merchant recommendation method, device, electronic equipment and readable storage medium storing program for executing | |
CN105786875B (en) | Question and answer are provided to the method and apparatus of data search result | |
CN109214955A (en) | The generation method and device of food product set meal | |
CN109447749A (en) | Merchandise news input method and device | |
CN106096003A (en) | Data search method and client | |
CN109801138A (en) | The searching method and device of commodity picture | |
CN110069676A (en) | Keyword recommendation method and device | |
CN109919703A (en) | Store information processing method, device, system, storage medium and computer equipment | |
CN109684005A (en) | Component similarity determines method and device in graphical interfaces | |
CN110782308A (en) | Pushing method and device for recommended package, electronic equipment and readable storage medium | |
CN110363570A (en) | Classification methods of exhibiting, device, electronic equipment and storage medium in | |
CN105786910B (en) | Entry weighing computation method and device | |
CN109241455A (en) | A kind of methods of exhibiting and device of recommended | |
CN109426974A (en) | Competing product analysis method and device | |
US20190266648A1 (en) | Systems and methods for analyzing consumer products | |
CN110263973A (en) | Predict the method and device of user behavior | |
CN110515929A (en) | Books methods of exhibiting calculates equipment and storage medium | |
US11379898B2 (en) | Touch and pressure-based apparel image searching | |
CN108388555A (en) | Commodity De-weight method based on category of employment and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |