CN110895781A - Dish type recommendation method and device, electronic equipment and storage medium - Google Patents
Dish type recommendation method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110895781A CN110895781A CN201911039652.5A CN201911039652A CN110895781A CN 110895781 A CN110895781 A CN 110895781A CN 201911039652 A CN201911039652 A CN 201911039652A CN 110895781 A CN110895781 A CN 110895781A
- Authority
- CN
- China
- Prior art keywords
- dish
- classification
- request
- merchant
- classification data
- 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
Images
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
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
The application discloses a dish type recommendation method, belongs to the technical field of computers, and is beneficial to improving the efficiency of obtaining dish types corresponding to dish names. The dish category recommendation method disclosed by the embodiment of the application comprises the following steps: responding to a classification request of at least one target dish of a current merchant, and acquiring dish classification data matched with the classification request; determining a dish category of the at least one target dish through dish classification data matched with the classification request; and taking the determined dish type as the recommended dish type of the corresponding target dish. According to the dish classification recommending method disclosed by the embodiment of the application, when the merchant has dish classification data, the dish to be classified currently is subjected to classification prediction based on the dish classification data owned by the merchant, and when the merchant does not have the dish classification data, the dish classification data based on the platform is subjected to classification prediction on the dish to be classified currently, so that the accuracy of dish classification prediction can be improved.
Description
Technical Field
The embodiment of the application relates to the technical field of computers, in particular to a dish category recommendation method and device, electronic equipment and a computer-readable storage medium.
Background
In the application fields of take-out, online ordering or dish browsing and the like, the electronic menu comprises dish classification information. In the prior art, there are two ways to generate dish classification information. First, the merchant manually sets category information, for example, when inputting dish information into an electronic menu, at the same time sets dish classification information. Secondly, acquiring the dish type information by combining image recognition and manual setting, for example, when the platform collects relevant information of dishes through an OCR recognition technology, for some dishes lacking the dish type information, the server of the platform issues the relevant information of the dishes to the client of the merchant in a dish list form, and the merchant manually sets the type information of the dishes one by one and then returns the type information to the platform server.
When the category information of dishes is obtained in the prior art, for merchants with more dishes, the merchants need to frequently perform manual operation, and manpower is consumed; in addition, the dish type is manually set one by one, and omission or setting errors are likely to occur. Therefore, the method for acquiring the dish category information in the prior art at least has the defect of low efficiency.
Disclosure of Invention
The embodiment of the application provides a dish type recommendation method which is beneficial to improving the efficiency of obtaining dish type information.
In order to solve the above problem, in a first aspect, an embodiment of the present application provides a method for recommending a category of dishes, including:
in response to a classification request for at least one target dish of a current merchant, obtaining dish classification data matched with the classification request, wherein the dish classification data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name;
determining a dish category of the at least one target dish through dish classification data matched with the classification request;
and taking the determined dish type as a recommended dish type of the corresponding target dish.
In a second aspect, an embodiment of the present application provides a device for recommending a category of dishes, including:
the dish classification data acquisition module is used for responding to a classification request of at least one target dish of a current merchant and acquiring dish classification data matched with the classification request, wherein the dish classification data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name;
a dish type determining module, configured to determine a dish type of the at least one target dish according to dish classification data matched with the classification request;
and the dish type recommending module is used for taking the determined dish type as the recommended dish type of the corresponding target dish.
In a third aspect, an embodiment of the present application further discloses an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, the method for recommending a dish category according to the embodiment of the present application is implemented.
In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, performs the steps of the dish category recommendation method disclosed in the embodiments of the present application.
The dish category recommendation method disclosed in the embodiment of the application obtains dish category data matched with a classification request by responding to the classification request of at least one target dish of a current merchant, wherein the dish category data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name; then, determining the dish category of the at least one target dish through dish classification data matched with the classification request; and the determined dish type is used as the recommended dish type of the corresponding target dish, so that the efficiency of obtaining the dish type corresponding to the dish name is improved. According to the dish category recommendation method disclosed by the embodiment of the application, when a merchant has dish classification data, dish categories to be classified currently are subjected to classification prediction based on the own dish classification data of the merchant, and when the merchant does not have the dish classification data, dish categories to be classified currently are subjected to classification prediction based on the dish classification data of a platform, so that the accuracy of dish classification prediction can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is a flowchart of a dish category recommendation method according to a first embodiment of the present application;
FIG. 2 is a schematic diagram of a new dish category interface in the first embodiment of the present application;
FIG. 3 is a schematic diagram illustrating a display effect of recommended dish categories according to a first embodiment of the present application;
fig. 4 is a schematic structural diagram of a dish category recommendation device according to a third embodiment of the present application;
fig. 5 is a second schematic structural diagram of a dish type recommending apparatus according to a third embodiment of the present application;
fig. 6 is a schematic structural diagram of a dish category recommendation device according to a fourth embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Example one
As shown in fig. 1, a method for recommending a category of dishes disclosed in an embodiment of the present application includes: step 110 to step 130.
Wherein the dish classification data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the name of the dish and the dish type corresponding to the name of the dish.
In some embodiments of the present application, the classification request for at least one target dish of the current merchant may be: and generating a recommended dish type request corresponding to the target dish when the dish type of the at least one target dish is displayed. For example, after a merchant a uploads a menu image of the merchant a to a platform through a merchant client of the platform, a server of the platform obtains a name of a dish of each dish on the menu image through an OCR recognition technology, but there is no dish category information of each dish on the menu image, the server may generate a classification request according to the dish name obtained through recognition, where the classification request includes a plurality of names of dishes and merchant information (such as a merchant identifier) of the merchant a. In other embodiments of the present application, the server of the platform may further generate a classification request according to each identified dish name, where the classification request includes a dish name.
In other embodiments of the present application, the request for classifying at least one target dish of the current merchant may be: and generating a new dish type request corresponding to a target dish when the dish type of the target dish is added. For example, when a merchant B edits an electronic menu through a merchant client of a platform, and when the merchant B wants to add a category of a new dish, the merchant B may select a target dish in an electronic menu editing interface, and then trigger an operation of adding the category of the dish, as shown in fig. 2, the merchant client acquires a name of the dish of the target dish, generates a new dish category request according to the acquired name of the dish of the target dish, and sends the new dish category request to a server of the platform. The new dish category request at least carries the dish name of the target dish and merchant information (such as a merchant identifier) of the merchant B.
After the server of the platform receives different classification requests, in the embodiment of the application, the server adopts different dish classification data determination methods for different classification requests.
For example, when the sorting request for at least one target dish of the current merchant may be: when a new dish category request corresponding to a target dish is generated when a dish category of the target dish is added, acquiring dish category data matched with a category request in response to the category request for at least one target dish of a current merchant, wherein the dish category request comprises: and responding to the new dish classification request, and acquiring platform dish classification data as dish classification data matched with the classification request. Taking the merchant B to edit the electronic menu again, for example, regarding the target dish "eel mixed rice", if the merchant B can select "eel mixed rice", clicking a "new category" button of the electronic menu editing interface to trigger a new classification operation; then, the platform client logged in by the merchant B generates a new dish type request, and sends the new dish type request to a server of the platform, where the new dish type request at least includes: the dish name of the target dish is eel mixed rice, and the platform client side identification is obtained. And after receiving the new dish classification request, the server of the platform takes dish classification data of other merchants on the platform as reference data for classifying the target dish.
The electronic menu data center of the platform can store a large amount of dish data of merchants according to the electronic menus uploaded by the merchants, wherein the dish data comprise: the name of the dish, the category of the dish, the characteristic label of the dish, the price of the dish, the description information of the dish and the like. The dish type is an abstract of names, namely the dish type reflects the common characteristics of the names of the dishes. For example, the dish names below the classification of the rice with clay pot all contain the rice with clay pot; the names of dishes under the classification of the rice served with rice. For the classification of dishes, most merchants have the same judgment standard, and different merchants may enter different dishes into the same classification, so that classification information among different merchants can be shared. If the merchant A records the classification from the eggplant-covered rice dish to the covered rice dish, and the merchant B records the classification from the tomato-egg covered rice dish to the covered rice dish, the category can also record the classification from the tomato-egg covered rice dish to the covered rice dish. Therefore, the dish classification data of other merchants on the platform can be used as a reference to determine the dish category corresponding to a certain dish name of a certain merchant.
For another example, when the sorting request for at least one target dish of the current merchant may be: when a recommended dish category request corresponding to the target dish is generated when the dish category of the at least one target dish is displayed, acquiring dish category data matched with the classification request in response to the classification request for the at least one target dish of the current merchant, wherein the method comprises the following steps: and responding to the recommended dish category request, and acquiring the dish category data of the current merchant as dish category data matched with the category request. Taking a scenario that a merchant A uploads a menu image to a server of a platform to generate an electronic menu as an example, the server generates a classification request according to the dish names of all target dishes in the menu image, wherein the classification request carries a list of the dish names of the target dishes and merchant information of the current merchant. At this time, the server first obtains the dish classification data matched with the merchant information according to the merchant information carried in the classification request, that is, the dish classification data of the current merchant. The dish classification data described in the embodiment of the present application includes: the name of the dish and the category of the dish corresponding to each name of the dish.
In some embodiments of the present application, if none of the existing dishes in the dish data of merchant a has dish category information, obtaining the merchant dish category data of the current merchant will return a failure.
In some embodiments of the present application, null data may be returned if obtaining the merchant dish classification data for the current merchant fails.
In some preferred embodiments of the present application, if the acquisition of the merchant dish classification data of the current merchant fails, the server may further acquire dish classification data of a platform for reference with dish classification of merchant a. For example, after acquiring the merchant dish classification data of the current merchant as dish classification data matched with the classification request in response to the recommended dish classification request, the method further includes: and in response to failure of acquiring the merchant dish classification data of the current merchant, acquiring platform dish classification data as dish classification data matched with the classification request. That is, when the merchant a does not have its own dish classification data (e.g., when none of the dish names of the merchant a has a corresponding dish category), the server may use the dish classification data of other merchants on the platform as a reference for predicting the dish category corresponding to the dish name of the merchant a.
And 120, determining the dish category of the at least one target dish according to the dish classification data matched with the classification request.
In some embodiments of the present application, the determining the dish category of the at least one target dish by the dish classification data matched with the classification request includes: predicting the names of the dishes carried in the classification request through a target dish classification model to obtain the dish categories corresponding to the names of the dishes; and the target dish classification model is obtained by training based on dish classification data matched with the classification request.
In some embodiments of the present application, the target dish classification model is trained based on dish classification data matched to the classification request. For example, the server of the platform trains the dish classification model of each merchant in advance according to the dish classification data of each merchant, and trains the dish classification model of the platform according to the dish classification data of all merchants on the platform.
That is, before determining the dish category of the at least one target dish through the dish classification data matched with the classification request, the method further includes: training a platform dish classification model according to platform dish classification data, and training a merchant dish classification model of the current merchant according to the dish classification data of the current merchant.
Each piece of dish data in the existing dish data of each merchant on the platform comprises: the dish name still includes: the method comprises the steps of selecting information such as dish types, prices, descriptions and pictures, selecting dish data including dish names and dish types from dish data of each merchant, and constructing training samples of merchant dish classification models and platform dish classification models. When the method is implemented specifically, the name of a dish is used as the input of a dish classification model, the category of the dish is used as the output of the dish classification model, the dish name text is vectorized and expressed, the matrix data is converted into matrix data, the matrix data is input to a neural network, and the dish classification model is obtained through training. When a dish name is input into the dish classification model, the dish type possibly corresponding to the dish name can be predicted.
In some embodiments of the present application, training a platform dish classification model according to platform dish classification data comprises: acquiring a plurality of groups of dish information of at least one merchant in the platform, wherein each group of dish information comprises a dish type corresponding to a dish name; and training the platform dish classification model according to a plurality of groups of dish information of at least one merchant. Training the platform dish classification model according to a plurality of groups of dish information of at least one merchant, wherein the training comprises the following steps: and for each group of the dish information, taking the dish name in the group of the dish information as the input of a platform dish classification model, taking the dish type in the group of the dish information as the output of the platform dish classification model, and training the platform dish classification model. That is, when training the platform dish classification model, a training sample is constructed based on dish data of some or all of the merchants on the platform (the dish data includes a dish name and a dish category).
In some embodiments of the present application, the step of training the dish classification model of the current merchant according to the dish classification data of the current merchant includes: acquiring a plurality of groups of dish information of the current merchant, wherein each group of dish information comprises a dish type corresponding to a dish name; and training a merchant dish classification model of the current merchant according to the plurality of groups of dish information. Training the platform dish classification model according to a plurality of groups of dish information of the current merchant, wherein the training step comprises the following steps: and for each group of dish information, taking the dish name in the group of dish information as the input of a platform dish classification model, taking the dish type in the group of dish information as the output of the platform dish classification model, and training the merchant dish classification model. That is, when the merchant dish classification model is trained, a training sample is constructed based on the dish data of each merchant (the dish data includes a dish name and a dish type), and the merchant dish classification model of each merchant is trained.
Further, the step of predicting the names of the dishes carried in the classification request through a target dish classification model to obtain the dish categories corresponding to the names of the dishes comprises the following steps: when platform dish classification data are obtained and serve as dish classification data matched with the classification requests, dish names carried in the classification requests are respectively identified through a platform dish classification model, and dish categories corresponding to the dish names are obtained, wherein the platform dish classification model is a dish classification model obtained in advance according to platform dish classification data training. For example, when the merchant a does not have the dish classification data of its own, the merchant dish classification model of the merchant a cannot be trained in advance, and then, when the dishes are classified, the platform dish classification model trained according to the platform dish classification data is selected to perform dish classification prediction on each dish name to be classified, so as to obtain a dish class possibly corresponding to each dish name.
For another example, when a merchant B newly creates a dish classification, the newly created dish classification is certainly not the existing dish classification of the merchant B, and therefore, a platform dish classification model trained according to platform dish classification data is selected to perform dish classification prediction on a dish name targeted by field dish classification operation, so as to obtain a dish classification possibly corresponding to the dish name.
In some embodiments of the present application, the step of predicting, by a target dish classification model, each dish name carried in the classification request to obtain a dish category corresponding to each dish name further includes: when the merchant dish classification data is obtained and serves as dish classification data matched with the classification request, dish names carried in the classification request are respectively identified through a merchant dish classification model, and dish categories corresponding to the dish names are obtained, wherein the merchant dish classification model is a dish classification model obtained in advance according to dish classification data of the current merchant through training. For example, the dish data of the merchant a includes dish data with both the name of the dish and the category of the dish, so that a training sample can be constructed in advance according to the dish data of the merchant a with both the name of the dish and the category of the dish, and a merchant dish classification model of the merchant a can be trained. And then, when the merchant A completes the electronic menu of the merchant A, inputting the dish name of the target dish of the merchant A into the merchant dish classification model of the merchant A for prediction to obtain the dish category of the target dish.
And step 130, taking the determined dish type as a recommended dish type of the corresponding target dish.
And then, taking the dish type corresponding to each dish name predicted by the dish classification model as the recommended dish type of the target dish of the dish name, and displaying. For example, in an electronic menu editing interface, under each dish name, "recommended dish category: and prompting information in the XXX character form is used as the recommended dish type of the corresponding dish, so that the efficiency of the merchant in setting the dish type is improved. The presentation effect of the recommended dish category is shown in fig. 3.
The dish category recommendation method disclosed in the embodiment of the application obtains dish category data matched with a classification request by responding to the classification request of at least one target dish of a current merchant, wherein the dish category data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name; then, determining the dish category of the at least one target dish through dish classification data matched with the classification request; and the determined dish type is used as the recommended dish type of the corresponding target dish, so that the efficiency of obtaining the dish type corresponding to the dish name is improved. According to the dish category recommendation method disclosed by the embodiment of the application, when a merchant has dish classification data, dish categories to be classified currently are subjected to classification prediction based on the own dish classification data of the merchant, and when the merchant does not have the dish classification data, dish categories to be classified currently are subjected to classification prediction based on the dish classification data of a platform, so that the accuracy of dish classification prediction can be improved.
Example two
In other embodiments of the present application, when determining the category of the target dish, the method may further be implemented by means of dish name similarity comparison. For example, the obtaining of dish classification data matching with a classification request in response to the classification request for at least one target dish of a current merchant comprises: and responding to the classification request, and acquiring platform dish classification data as dish classification data matched with the classification request. Wherein the classification request comprises: and displaying the dish type of the at least one target dish to generate a recommended dish type request corresponding to the target dish, or adding a new dish type request corresponding to the target dish when the dish type of the target dish is added.
A specific obtaining manner of a recommended dish category request corresponding to the target dish generated when the dish category of the at least one target dish is displayed is shown in the first embodiment, and details are not repeated in this embodiment.
For a specific obtaining manner of a new dish category request corresponding to a target dish generated when a dish category of the target dish is added, refer to embodiment one, and this embodiment is not described again.
After acquiring the classification request for the target dish, in this embodiment, the server of the platform acquires the platform dish classification data as the dish classification data matched with the classification request, and is used as the reference data for classifying the target dish. The platform dish classification data comprises merchant dish classification data of at least one merchant on the platform (namely, part or all of the merchants on the platform comprise dish names and dish categories), and each merchant dish classification data comprises: the name of the dish and the dish type corresponding to the name of the dish.
Correspondingly, the determining the dish category of the at least one target dish through the dish classification data matched with the classification request comprises: and for each target dish, respectively carrying out similarity calculation on the dish name of the target dish and the dish name in each dish classification data matched with the classification request, and determining the dish type of the target dish according to the similarity calculation result. For example, the platform dish classification data includes dish classification data of merchant B, dish classification data of merchant C, and dish classification data of merchant D, and there are M pieces of platform dish classification data, and when the dish category of the target dish P1 to be classified is predicted, the similarity between the dish name in the M pieces of platform dish classification data of the platform and the dish name of the target dish P1 may be calculated, and then, according to the calculation result of the similarity, the dish category in the piece of platform dish classification data with the highest similarity is selected as the dish category of the target dish P1.
In some embodiments of the present application, the performing similarity calculation on the dish name of the target dish and each of the dish names in the dish classification data matched with the classification request, and determining the dish category of the target dish according to a similarity calculation result includes: determining a word segmentation sequence corresponding to the dish name by performing word segmentation processing on the dish name; acquiring a word segmentation sequence corresponding to each dish name in each dish classification data matched with the classification request; respectively calculating the editing distance between the word segmentation sequence corresponding to each dish name in the dish classification data matched with the classification request and the word segmentation sequence corresponding to the dish name of the target dish; and determining the dish category to which the dish name belongs, which is included in a piece of dish classification data matched with the classification request and corresponds to the minimum editing distance, as the dish category of the target dish.
For example, the server of the platform first performs word segmentation processing on the dish names included in each platform dish classification data to obtain word segmentation sequences corresponding to the dish names in each platform dish classification data, and stores the word segmentation sequences corresponding to the dish names in the server. When the server receives a classification request for a target dish, the server also performs word segmentation on the dish name of the target dish carried in the classification request (for example, performs word segmentation on the dish name of the target dish P1), so as to obtain a word segmentation sequence to be matched. And then, respectively calculating the editing distance between the word segmentation sequence corresponding to each dish name in each platform dish classification data and the word segmentation sequence to be matched. The edit distance can be used for quantitative measurement of the difference degree between two character strings, the measurement mode is to change one character string into another character string by at least how many times of processing is needed, and the smaller the edit distance is, the higher the similarity of the two character strings is. Therefore, the dish name most similar to the dish name of the target dish P1 can be known by the edit distance. And finally, determining the dish category to which the dish name belongs in the platform dish classification data corresponding to the minimum editing distance, and taking the dish category as the dish category of the target dish.
And finally, taking the determined dish type as the recommended dish type of the corresponding target dish.
And then, taking the dish type corresponding to the dish name determined by the dish name similarity calculation as the recommended dish type of the target dish of the dish name, and displaying.
According to the dish category recommendation method disclosed by the embodiment of the application, platform dish category data is obtained as dish category data matched with a classification request by responding to the classification request of at least one target dish of a current merchant; then, determining the dish type of the at least one target dish according to similarity calculation through the platform dish classification data; and the determined dish type is used as the recommended dish type of the corresponding target dish, so that the efficiency of obtaining the dish type corresponding to the dish name is improved. According to the dish category recommendation method disclosed by the embodiment of the application, the similarity between dish names is calculated based on a performance-efficient editing distance algorithm, meanwhile, the phrase attributes of the dish names are fully considered, the fuzzy classification relation between the dishes can be quantized, and the fuzzy classification relation can be quickly fed back to the dish category recommendation of the target dish, so that the manual input cost is reduced, and the dish category editing efficiency is improved.
EXAMPLE III
The vegetable type recommendation device disclosed in the embodiment of the present application, as shown in fig. 4, the device includes:
a dish classification data obtaining module 410, configured to, in response to a classification request for at least one target dish of a current merchant, obtain dish classification data matching the classification request, where the dish classification data includes: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name;
a dish category determining module 420 for determining a dish category of the at least one target dish by dish classification data matched with the classification request;
and the dish type recommending module 430 is configured to take the determined dish type as a recommended dish type of the corresponding target dish.
In some embodiments of the present application, the classification request comprises: a recommended dish type request corresponding to the target dish is generated when the dish type of the at least one target dish is displayed, or a new dish type request corresponding to the target dish is generated when the dish type of the target dish is added; as shown in fig. 5, the dish classification data acquisition module 410 further includes:
a first dish classification data obtaining sub-module 4101, configured to, in response to the recommended dish classification request, obtain the merchant dish classification data of the current merchant as dish classification data matched with the classification request;
a second dish classification data obtaining sub-module 4102, configured to, in response to the new dish classification request, obtain platform dish classification data as dish classification data matched with the classification request.
In some embodiments of the application, after the acquiring the merchant dish classification data of the current merchant in response to the recommended dish classification request as dish classification data matched with the classification request, as shown in fig. 5, the dish classification data acquiring module 410 further includes:
a third dish classification data obtaining sub-module 4103, configured to, in response to failure of obtaining the merchant dish classification data of the current merchant, obtain platform dish classification data as dish classification data matched with the classification request.
Accordingly, as shown in fig. 5, the dish category determining module 420 further includes:
a first dish category determining submodule 4201, configured to predict, through a target dish classification model, each dish name carried in the classification request, respectively, to obtain a dish category corresponding to each dish name; and the target dish classification model is obtained by training based on dish classification data matched with the classification request.
The dish category recommendation device disclosed in the embodiment of the application acquires dish category data matched with a classification request by responding to the classification request of at least one target dish of a current merchant, wherein the dish category data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name; then, determining the dish category of the at least one target dish through dish classification data matched with the classification request; and the determined dish type is used as the recommended dish type of the corresponding target dish, so that the efficiency of obtaining the dish type corresponding to the dish name is improved. The utility model provides a vegetable classification recommending device, when the trade company has vegetable classification data, carry out classification prediction to the current vegetable of waiting to classify based on trade company's own vegetable classification data, when the trade company does not have vegetable classification data, vegetable classification data based on the platform carries out classification prediction to the current vegetable of waiting to classify, can promote vegetable classification prediction's accuracy, show the trade company through the vegetable classification that obtains with the prediction, reduce the trade company when perfecting the vegetable classification of a plurality of vegetables, the work load of manual edition vegetable classification, promote the efficiency that the vegetable classification set up.
Example four
In other embodiments of the present application, the dish type recommending apparatus may further perform a means of comparing similarity of names of dishes when determining the dish type of the target dish. For example, the classification request includes: a recommended dish category request corresponding to the target dish is generated when the dish category of the at least one target dish is displayed; as shown in fig. 6, the dish classification data acquisition module 410 further includes:
a fourth dish classification data acquisition sub-module 4104, configured to acquire, in response to the classification request, platform dish classification data as dish classification data matched with the classification request;
accordingly, the dish category determining module 420 further includes:
and a second dish type determining sub-module 4202, configured to, for each target dish, perform similarity calculation on the dish name of the target dish and each dish name in the dish classification data matched with the classification request, and determine the dish type of the target dish according to a similarity calculation result.
In some embodiments of the present application, the performing similarity calculation on the dish name of the target dish and each of the dish names in the dish classification data matched with the classification request, and determining the dish category of the target dish according to a similarity calculation result includes:
determining a word segmentation sequence corresponding to the dish name by performing word segmentation processing on the dish name;
acquiring a word segmentation sequence corresponding to each dish name in each dish classification data matched with the classification request;
respectively calculating the editing distance between the word segmentation sequence corresponding to each dish name in the dish classification data matched with the classification request and the word segmentation sequence corresponding to the dish name of the target dish;
and determining the dish category to which the dish name belongs, which is included in a piece of dish classification data matched with the classification request and corresponds to the minimum editing distance, as the dish category of the target dish.
The dish category recommendation device disclosed in the embodiment of the application acquires platform dish category data serving as dish category data matched with a classification request by responding to the classification request of at least one target dish of a current merchant; then, determining the dish type of the at least one target dish according to similarity calculation through the platform dish classification data; and the determined dish type is used as the recommended dish type of the corresponding target dish, so that the efficiency of obtaining the dish type corresponding to the dish name is improved. The dish category recommendation device disclosed by the embodiment of the application calculates the similarity between dish names based on a performance-efficient editing distance algorithm, and meanwhile fully considers the phrase attributes of the dish names, so that the fuzzy classification relation between the dishes can be quantized, and the fuzzy classification relation can be quickly fed back to the dish category recommendation of the target dish, thereby reducing the manual input cost and improving the dish category editing efficiency.
The device for recommending vegetable categories disclosed in the embodiment of the present application is used to implement the method for recommending vegetable categories described in the first embodiment or the second embodiment of the present application, and specific implementation manners of each module of the device are not described again, and reference may be made to specific implementation manners of corresponding steps in the method embodiments.
Correspondingly, the application also discloses an electronic device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the computer program to realize the dish type recommendation method according to the first embodiment or the second embodiment of the application. The electronic device can be a PC, a mobile terminal, a personal digital assistant, a tablet computer and the like.
The present application also discloses a computer-readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the dish category recommendation method according to the first or second embodiment of the present application.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The method and the device for recommending the dish category provided by the application are introduced in detail, a specific example is applied in the method to explain the principle and the implementation mode of the application, and the description of the embodiment is only used for helping to understand the method and the core idea of the application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Claims (10)
1. A method for recommending a category of dishes, comprising:
in response to a classification request for at least one target dish of a current merchant, obtaining dish classification data matched with the classification request, wherein the dish classification data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name;
determining a dish category of the at least one target dish through dish classification data matched with the classification request;
and taking the determined dish type as a recommended dish type of the corresponding target dish.
2. The method of claim 1, wherein the classification request comprises: a recommended dish type request corresponding to the target dish is generated when the dish type of the at least one target dish is displayed, or a new dish type request corresponding to the target dish is generated when the dish type of the target dish is added; the step of acquiring dish classification data matched with a classification request in response to the classification request for at least one target dish of a current merchant comprises the following steps:
responding to the recommended dish category request, and acquiring the dish category data of the current merchant as dish category data matched with the category request;
and responding to the new dish classification request, and acquiring platform dish classification data as dish classification data matched with the classification request.
3. The method of claim 1, wherein the step of obtaining the merchant food classification data for the current merchant in response to the recommended food category request as food classification data matching the classification request further comprises, after the step of obtaining the merchant food classification data for the current merchant:
and in response to failure of acquiring the merchant dish classification data of the current merchant, acquiring platform dish classification data as dish classification data matched with the classification request.
4. The method of claim 2 or 3, wherein the step of determining the dish category of the at least one target dish from the dish classification data matched to the classification request comprises:
predicting the names of the dishes carried in the classification request through a target dish classification model to obtain the dish categories corresponding to the names of the dishes; and the target dish classification model is obtained by training based on dish classification data matched with the classification request.
5. The method of claim 1, wherein the classification request comprises: a recommended dish category request corresponding to the target dish is generated when the dish category of the at least one target dish is displayed; the step of acquiring dish classification data matched with a classification request in response to the classification request for at least one target dish of a current merchant comprises the following steps:
responding to the classification request, and acquiring platform dish classification data as dish classification data matched with the classification request;
the step of determining the dish category of the at least one target dish by the dish classification data matched with the classification request includes:
and for each target dish, respectively carrying out similarity calculation on the dish name of the target dish and the dish name in each dish classification data matched with the classification request, and determining the dish type of the target dish according to the similarity calculation result.
6. The method of claim 5, wherein the step of performing similarity calculation on the dish name of the target dish and the dish name in each of the dish classification data matched with the classification request, and determining the dish category of the target dish according to the similarity calculation result comprises:
determining a word segmentation sequence corresponding to the dish name by performing word segmentation processing on the dish name;
acquiring a word segmentation sequence corresponding to each dish name in each dish classification data matched with the classification request;
respectively calculating the editing distance between the word segmentation sequence corresponding to each dish name in the dish classification data matched with the classification request and the word segmentation sequence corresponding to the dish name of the target dish;
and determining the dish category to which the dish name belongs, which is included in a piece of dish classification data matched with the classification request and corresponds to the minimum editing distance, as the dish category of the target dish.
7. A device for recommending a category of dishes, comprising:
the dish classification data acquisition module is used for responding to a classification request of at least one target dish of a current merchant and acquiring dish classification data matched with the classification request, wherein the dish classification data comprises: platform dish classification data or merchant dish classification data of the current merchant, wherein the platform dish classification data comprises merchant dish classification data of at least one merchant on the platform, and each merchant dish classification data comprises: the dish name and the dish type corresponding to the dish name;
a dish type determining module, configured to determine a dish type of the at least one target dish according to dish classification data matched with the classification request;
and the dish type recommending module is used for taking the determined dish type as the recommended dish type of the corresponding target dish.
8. The apparatus of claim 7, wherein the classification request comprises: a recommended dish type request corresponding to the target dish is generated when the dish type of the at least one target dish is displayed, or a new dish type request corresponding to the target dish is generated when the dish type of the target dish is added; the dish classification data acquisition module further comprises:
a first dish classification data acquisition sub-module, configured to respond to the recommended dish classification request, acquire the merchant dish classification data of the current merchant, as dish classification data matched with the classification request;
and the second dish classification data acquisition sub-module is used for responding to the new dish classification request and acquiring platform dish classification data as dish classification data matched with the classification request.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the dish category recommendation method of any one of claims 1 to 6 when executing the computer program.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the dish category recommendation method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911039652.5A CN110895781A (en) | 2019-10-29 | 2019-10-29 | Dish type recommendation method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911039652.5A CN110895781A (en) | 2019-10-29 | 2019-10-29 | Dish type recommendation method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110895781A true CN110895781A (en) | 2020-03-20 |
Family
ID=69786396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911039652.5A Pending CN110895781A (en) | 2019-10-29 | 2019-10-29 | Dish type recommendation method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110895781A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111444345A (en) * | 2020-03-30 | 2020-07-24 | 拉扎斯网络科技（上海）有限公司 | Dish name classification method and device |
| WO2022137731A1 (en) * | 2020-12-25 | 2022-06-30 | Ｎｅｃプラットフォームズ株式会社 | Information processing device, system, method, and non-transitory computer-readable medium having program stored therein |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102222100A (en) * | 2011-06-21 | 2011-10-19 | 青岛海信传媒网络技术有限公司 | Method and device for commodity information access |
| CN103778205A (en) * | 2014-01-13 | 2014-05-07 | 北京奇虎科技有限公司 | Commodity classifying method and system based on mutual information |
| CN106296208A (en) * | 2015-06-05 | 2017-01-04 | 阿里巴巴集团控股有限公司 | A kind of commodity classification method and device |
| CN106815745A (en) * | 2016-12-30 | 2017-06-09 | 北京三快在线科技有限公司 | Vegetable recommends method and system |
| CN107657284A (en) * | 2017-10-11 | 2018-02-02 | 宁波爱信诺航天信息有限公司 | A kind of trade name sorting technique and system based on Semantic Similarity extension |
| CN107844790A (en) * | 2017-11-15 | 2018-03-27 | 上海捷售智能科技有限公司 | A kind of vegetable identification and POS and method based on image recognition |
| CN108256474A (en) * | 2018-01-17 | 2018-07-06 | 百度在线网络技术（北京）有限公司 | For identifying the method and apparatus of vegetable |
| CN108595418A (en) * | 2018-04-03 | 2018-09-28 | 上海透云物联网科技有限公司 | A kind of commodity classification method and system |
| CN109214409A (en) * | 2018-07-10 | 2019-01-15 | 上海斐讯数据通信技术有限公司 | A kind of vegetable recognition methods and system |
| US20190043033A1 (en) * | 2017-08-03 | 2019-02-07 | Toshiba Tec Kabushiki Kaisha | Point-of-sale terminal |
| CN109345179A (en) * | 2018-10-25 | 2019-02-15 | 珠海市视微网络科技有限公司 | A kind of commodity management system based on e-commerce platform |
| CN109740597A (en) * | 2018-12-29 | 2019-05-10 | 口碑(上海)信息技术有限公司 | The recognition methods of menu information and device |
| CN109872254A (en) * | 2018-11-30 | 2019-06-11 | 广州富港万嘉智能科技有限公司 | A kind of food plan reminding method, system and storage medium |
| CN110347801A (en) * | 2019-07-17 | 2019-10-18 | 安徽航天信息有限公司 | A kind of commodity classification codes match method and system |
-
2019
- 2019-10-29 CN CN201911039652.5A patent/CN110895781A/en active Pending
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102222100A (en) * | 2011-06-21 | 2011-10-19 | 青岛海信传媒网络技术有限公司 | Method and device for commodity information access |
| CN103778205A (en) * | 2014-01-13 | 2014-05-07 | 北京奇虎科技有限公司 | Commodity classifying method and system based on mutual information |
| CN106296208A (en) * | 2015-06-05 | 2017-01-04 | 阿里巴巴集团控股有限公司 | A kind of commodity classification method and device |
| CN106815745A (en) * | 2016-12-30 | 2017-06-09 | 北京三快在线科技有限公司 | Vegetable recommends method and system |
| US20190043033A1 (en) * | 2017-08-03 | 2019-02-07 | Toshiba Tec Kabushiki Kaisha | Point-of-sale terminal |
| CN107657284A (en) * | 2017-10-11 | 2018-02-02 | 宁波爱信诺航天信息有限公司 | A kind of trade name sorting technique and system based on Semantic Similarity extension |
| CN107844790A (en) * | 2017-11-15 | 2018-03-27 | 上海捷售智能科技有限公司 | A kind of vegetable identification and POS and method based on image recognition |
| CN108256474A (en) * | 2018-01-17 | 2018-07-06 | 百度在线网络技术（北京）有限公司 | For identifying the method and apparatus of vegetable |
| CN108595418A (en) * | 2018-04-03 | 2018-09-28 | 上海透云物联网科技有限公司 | A kind of commodity classification method and system |
| CN109214409A (en) * | 2018-07-10 | 2019-01-15 | 上海斐讯数据通信技术有限公司 | A kind of vegetable recognition methods and system |
| CN109345179A (en) * | 2018-10-25 | 2019-02-15 | 珠海市视微网络科技有限公司 | A kind of commodity management system based on e-commerce platform |
| CN109872254A (en) * | 2018-11-30 | 2019-06-11 | 广州富港万嘉智能科技有限公司 | A kind of food plan reminding method, system and storage medium |
| CN109740597A (en) * | 2018-12-29 | 2019-05-10 | 口碑(上海)信息技术有限公司 | The recognition methods of menu information and device |
| CN110347801A (en) * | 2019-07-17 | 2019-10-18 | 安徽航天信息有限公司 | A kind of commodity classification codes match method and system |
Non-Patent Citations (1)
| Title |
|---|
| 段雪梅 等: "双线性模型在中国菜分类中的应用", 《小型微型计算机系统》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111444345A (en) * | 2020-03-30 | 2020-07-24 | 拉扎斯网络科技（上海）有限公司 | Dish name classification method and device |
| WO2022137731A1 (en) * | 2020-12-25 | 2022-06-30 | Ｎｅｃプラットフォームズ株式会社 | Information processing device, system, method, and non-transitory computer-readable medium having program stored therein |
| JP2022102430A (en) * | 2020-12-25 | 2022-07-07 | Ｎｅｃプラットフォームズ株式会社 | Information processing apparatus, system, method, and program |
| JP7103683B2 (en) | 2020-12-25 | 2022-07-20 | Ｎｅｃプラットフォームズ株式会社 | Information processing equipment, systems, methods and programs |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107424043B (en) | Product recommendation method and device and electronic equipment | |
| US11875241B2 (en) | Aspect pre-selection using machine learning | |
| CN110750697B (en) | Merchant classification method, device, equipment and storage medium | |
| CN111400599A (en) | User group portrait generation method, device and system | |
| CN110929136A (en) | Personalized recommendation method and device | |
| CN116308684B (en) | Online shopping platform store information pushing method and system | |
| CN110929764A (en) | Picture auditing method and device, electronic equipment and storage medium | |
| US20230030560A1 (en) | Methods and systems for tagged image generation | |
| CN110895781A (en) | Dish type recommendation method and device, electronic equipment and storage medium | |
| US20170083574A1 (en) | Search system, search method, and program | |
| CN111626767A (en) | Resource data distribution method, device and equipment | |
| CN111654714B (en) | Information processing method, apparatus, electronic device and storage medium | |
| CN112783468A (en) | Target object sorting method and device | |
| CN107357847B (en) | Data processing method and device | |
| CN115907926A (en) | Commodity recommendation method and device, electronic equipment and storage medium | |
| WO2017219317A1 (en) | Information pushing method and device based on search content | |
| CN113327145A (en) | Article recommendation method and device | |
| CN111833085A (en) | Method and device for calculating price of article | |
| CN116151392B (en) | Training sample generation method, training method, recommendation method and device | |
| KR102605195B1 (en) | Method for providing information based on e-commerce and computing device for executing the same | |
| CN113343090B (en) | Method, apparatus, device, medium and product for pushing information | |
| CN116911913B (en) | Method and device for predicting interaction result | |
| CN111008331B (en) | Store-side display method and device, electronic equipment and storage medium | |
| US20220343388A1 (en) | Intelligent online platform for digitizing, searching, and providing services | |
| CN114329157A (en) | Related material acquisition method and device and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200320 |