WO2021169389A1 - Object information processing method and apparatus, electronic device, and readable storage medium - Google Patents

Abstract

An object information processing method and apparatus, a storage medium, and an electronic device. The object information processing method comprises: acquiring object requirement description information (S201); performing feature extraction with respect to the object requirement description information, so as to obtain feature data of multiple dimensions (S202); inputting the feature data of the multiple dimensions into a pre-trained object recommendation model so as to obtain recommended object information, wherein the object recommendation model is obtained by training, according to sample feature data of multiple dimensions and preset weights of the respective multiple dimensions, a preset model (S203); and outputting the recommended object information (S204). The invention reduces the amount of learning about a transaction platform for service providers, and reduces the workload of entering data and information associated with service content, thereby enabling the service providers to quickly and efficiently settle in the transaction platform.

Description

Object information processing method, device, electronic equipment and readable storage medium

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office, the application number is 202010125846.3, and the invention title is "Object Information Processing Methods, Devices, Electronic Equipment, and Readable Storage Media" on February 27, 2020, and its entire contents Incorporated in this application by reference.

Technical field

The embodiments of the present application relate to the field of data processing technology, and in particular, to an object information processing method, device, electronic device, and readable storage medium.

Background technique

With the development of e-commerce, people can easily access the Internet through various electronic devices to obtain various services, such as B2C (Business-to-Consumer, between service providers and consumers) that is common in daily life E-commerce) model, O2O (Online-to-Offline, online and offline e-commerce) model, which greatly facilitates and enriches people’s lives.

Among them, whether in the B2C model or the O2O model, service providers usually do not directly face consumers, but after entering various trading platforms, they will indirectly face consumers through the trading platforms for consumers to conduct service content. choose.

In related technologies, when a service provider enters a trading platform, it needs to enter data information related to the service content provided into the trading platform. For example, take a service provider’s entry into a takeaway trading platform as an example. If the service provider’s business item is catering, the service provider must at least provide the food delivery platform with information about the dishes provided to consumers in advance (such as vegetable varieties, prices, etc.) Etc.); if the service provider’s business project is floral art, the service provider at least needs to provide the outbound transaction platform with information about flowers (such as flower types, prices, etc.) that are provided to consumers in advance. As a result, service providers need to enter a large amount of data and information, and the efficiency of information processing is low.

Summary of the invention

The embodiments of this application provide an object information processing method, device, electronic equipment, and readable storage medium to assist service providers in entering the trading platform.

The first aspect of the embodiments of the present application provides an object information processing method, and the method includes:

Obtain the target demand description information;

Perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions;

Inputting feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, wherein the object recommendation model is based on sample feature data of multiple dimensions and preset weights of each of the multiple dimensions, Obtained by training the preset model;

Output the recommended object information.

A second aspect of the embodiments of the present application provides an object information processing device, the device including:

The first information obtaining module is used to obtain the object requirement description information;

The first feature extraction module is used to perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions;

The input module is used to input feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, wherein the object recommendation model is based on sample feature data of multiple dimensions and each of the multiple dimensions The preset weight of is obtained by training the preset model;

The output module is used to output the recommended object information.

A third aspect of the embodiments of the present application provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps in the method described in the first aspect of the present application are implemented.

The fourth aspect of the embodiments of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements the method described in the first aspect of the present application when executed A step of.

The fifth aspect of the embodiments of the present application provides a computer program, including computer-readable code, which when the computer-readable code runs on a computing processing device, causes the computing processing device to execute the aforementioned night scene high dynamic range image Generation method.

Using the object information processing method provided by the embodiments of this application, the server (or server cluster) of the trading platform obtains a large amount of sample characteristic data, which not only characterizes the characteristics of the service provider’s service content, but also from a number of different perspectives. (That is, dimensions) reflect the characteristics of the service content; and assign preset weights to each dimension. Use sample feature data with multiple dimensions and preset weights assigned to each dimension to train the preset model, for example, to train a pre-selected neural network model to obtain an object recommendation model. The object recommendation model can be based on the input The service provider is prepared to provide characteristic data of the service content, and recommends to the service provider at least one specific service item that matches the service content it is preparing to provide and relevant information about each service item.

Using the technical solutions provided by the embodiments of this application, when a service provider is ready to settle in the trading platform, the server (or server cluster) of the trading platform obtains the relevant information about the service content that the service provider intends to provide, that is, after the object’s demand description information, the service provider Feature extraction is performed on the target demand description information of the supplier, and the characteristic information related to the service content and representing the service content from different angles is extracted from the description information, that is, the characteristic data of multiple dimensions is extracted from the target demand description information, and then The feature data of multiple dimensions extracted from the description information is input to the pre-trained object recommendation model. After the object recommendation model is processed, it outputs at least one specific service item and various service items that match the service content that the service provider intends to provide Relevant information, that is, the recommended object information.

Using the technical solution provided by the embodiments of this application, it is only necessary to save the recommended object information output by the object recommendation model in the data center of the transaction platform, thereby helping the service provider to enter some data information related to the service content that it is preparing to provide into the transaction In the platform, the amount of learning for service providers to learn the trading platform and the workload of entering data and information related to the service content are reduced to assist service providers to quickly and efficiently complete the work of entering the trading platform.

The above description is only an overview of the technical solution of the present invention. In order to understand the technical means of the present invention more clearly, it can be implemented in accordance with the content of the specification, and in order to make the above and other objectives, features and advantages of the present invention more obvious and understandable. In the following, specific embodiments of the present invention are specifically cited.

Description of the drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

Figure 1 is an exemplary illustration of the relationship between service providers, trading platforms, and consumers in the B2C model and the O2O model;

FIG. 2 is a flowchart of an object information processing method proposed by an embodiment of the present application;

FIG. 3 is a flowchart of training a preset model in an object information processing method proposed by an embodiment of the present application;

FIG. 4 is a flowchart of an object information processing method proposed by another embodiment of the present application;

FIG. 5 is a flowchart of an object information processing method proposed by another embodiment of the present application;

FIG. 6 is a flowchart of a method for processing object information proposed by still another embodiment of the present application;

FIG. 7 is an application schematic diagram of an object information processing method proposed by an embodiment of the present application;

FIG. 8 is a schematic diagram of an object information processing device proposed by an embodiment of the present application;

Fig. 9 schematically shows a block diagram of a computing processing device for executing the method according to the present invention; and

Fig. 10 schematically shows a storage unit for holding or carrying program codes for implementing the method according to the present invention.

Specific embodiment

In order to make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

With the development of e-commerce, more and more service providers choose to settle in third-party trading platforms to provide consumers with more diverse and personalized services conveniently and efficiently through third-party trading platforms. For example, various service providers in the catering industry mainly provide catering services to consumers through platforms such as Meituan Waimai, Ele.me Takeout, and Word of Mouth, which provide catering takeaway services. For example, various service providers engaged in commodity retailing mainly provide catering services to consumers. Platforms such as Jingdong Mall, Tmall Mall and Taobao Mall that provide retail services sell goods to consumers. For example, various service providers that provide accommodation and transportation services mainly settle in platforms that provide travel services such as Ctrip, Qunar and Fliggy. Provide consumers with travel services and so on.

When the service provider enters the trading platform, it needs to provide some necessary data information to the trading platform, especially the necessary data information related to the service content provided by the trading platform, so that the trading platform can standardize and integrate the data information of the service provider. Consumer display. For example, take a service provider’s entry into a takeaway trading platform as an example. If the service provider’s business item is catering, the service provider must at least provide the food delivery platform with information about the dishes provided to consumers in advance (such as vegetable varieties, prices, etc.) Etc.), if the service provider’s business project is floral art, the service provider at least needs to provide the outbound transaction platform with the flower information (such as flower type, price, etc.) pre-provided to consumers.

However, the inventor of this application found that if a service provider is entering the trading platform for the first time, for the service provider, because they are not familiar with the operation of the trading platform, not only need to spend time learning the operation of the trading platform, but also need It will be a very large and complicated task to enter data information related to the provided service content into the trading platform.

Furthermore, if the operator of the service provider (ie, the merchant) intends to provide consumers with the service content for the first time to engage in business, the merchant still needs to conduct a lot of market research and other work in the early stage to determine the service it intends to provide to consumers. Whether the specific service items of the content and its related information (such as prices, business hours, etc.) meet consumer expectations and/or market needs to a certain extent; the specific service content to be provided can be determined after the completion of the market research work The service items and related information, and then the service provider will enter the specific service items and related information of the service content they are going to provide into the third-party trading platform. The workload will be more large and complicated, and it will be difficult to complete in a short period of time. Especially for small and medium-sized service providers, due to the limitations of capital and human resources, the early entry work will become more difficult.

It can be seen that service providers face huge and complex workloads in related technologies, which makes it difficult for merchants to enter the trading platform. It is time-consuming and laborious to enter the trading platform in the early stage, which greatly affects the user experience of the service provider. This difficulty will be schematically illustrated in an example below.

As shown in Figure 1, it exemplarily shows the relationship between service providers, trading platforms, and consumers in the B2C model and the O2O model. The service providers enter the entrance ( For example, the mobile APP, desktop client, interactive web interface, etc. provided by the trading platform for service providers enter various trading platforms, and the trading platform stores the data information related to the services provided by each service provider to the data center (or In the database, or server cluster), the trading platform standardizes and integrates the information of these service providers, and then through the trading platform to provide consumers with the service selection entrance (such as the mobile platform provided by the trading platform for consumers) APP, desktop client, interactive web interface, etc.) are displayed to consumers, so that consumers can choose service providers that meet consumers' expectations and service content provided by service providers, and finally realize the provision of services for consumers.

As shown in Figure 1, the operator of the service provider 1 is preparing to engage in hot pot business for the first time (that is, the service content that the service provider 1 is going to provide), and it is a small business. The initial plan is to open a business near an office building in a certain business district. The Chongqing hot pot restaurant is about ㎡ (square meter), and the operator is preparing to enter the Meituan takeaway platform to provide services to consumers through a combination of online and offline methods. Then, the more traditional way to settle in is that the operator of the service provider 1 needs to conduct on-site visits and research in the business district after the site selection (that is, near the office building of a certain business district), for example, to obtain the main consumer groups , What are the main flavors and consumer price range and other information, after obtaining the necessary market information, combined with their own knowledge and business experience, sum up and analyze a set of menus. Limited to the difficulty and amount of information acquisition and the lack of professional knowledge, the operator of the service provider 1 summarizes and analyzes the various dishes in the menu (that is, the specific service items of the service content that the service provider 1 is prepared to provide) and information such as prices. To a large extent, it may not meet the consumer expectations and/or market needs of this business district.

After the operator of the service provider 1 has determined the menu, at least the menu needs to be entered into the data center of the Meituan takeaway platform before it can be stationed on the Meituan takeaway platform to provide services to consumers. The operator of the service provider 1 first needs to learn the operation method and process of menu entry on the Meituan takeaway platform. After being familiar with the operation process, they can enter the menu. There are mainly three traditional entry methods: the first one, the service provider The operator of 1 manually enters the dishes and related information in the menu one by one into the data center of the Meituan takeaway platform; in the second, the operator of the service provider 1 makes the menu into an Excel table, Then enter the menu into the data center of the Meituan takeaway platform through batch import; in the third type, the operator of the service provider 1 takes the menu as a photo of the menu and uploads it to the server (or server cluster) of the Meituan takeaway platform In this, the server (or server cluster) uses computer vision technology such as OCR recognition to recognize menu photos, and then stores the recognized menu recognition results in the data center. Among them, the first and second entry methods not only require a lot of work, but also have relatively low entry efficiency; the third method, although it improves entry efficiency to a certain extent, is limited to the limitations of OCR recognition technology and the recognition results are accurate. The rate is difficult to guarantee, resulting in large errors in the entered menu results.

It can be seen that the operator of the service provider 1 not only has a huge and complicated workload for determining the specific service items (menus) of the service content, but also has a larger and more complicated workload for entering information about specific service items into the trading platform.

In view of this, the inventor of this application proposes: whether the trading platform can recommend specific service items and related information that meet the expectations of the service provider to the service provider, and directly enter the recommended information into the data center of the trading platform, and the service provider no longer needs it Carry out the input of specific service items and related information to reduce the workload of service providers entering the trading platform in the initial stage.

To this end, this application proposes an inventive concept: to construct an object recommendation model that can recommend at least one service content that matches the service content that the service provider is prepared to provide based on the input feature information of the service content that the service provider is prepared to provide Specific service items and relevant information of each service item; when the service provider is ready to enter the trading platform, the characteristic information of the service content that the service provider is going to provide is input into the object recommendation model, and the recommendation model recommends the service provider and the service it is going to provide The specific service items that match the service content and the relevant information of each service item are used to assist service providers in completing the settlement of the trading platform.

It should be noted that, in the embodiments of the present application, the data center of the exemplary transaction platform refers to a database that stores data. The database for storing data can either be a storage medium and/or storage device built into the server (or server cluster) of the trading platform, or it can exist independently of the server (or server cluster) of the trading platform and be connected to the server (or server cluster). Or server cluster) data storage equipment, storage medium and/or storage device for communicating and data interaction.

It should also be noted that, in the embodiments of this application, the example service providers and trading platforms are all exemplary, and are not a limitation of this application. Those skilled in the art can directly and without doubt ascertain other industries' The relationship between service providers and corresponding trading platforms; at the same time, in all embodiments of this application, the focus will be on service providers engaged in the catering industry (such as service providers 1, 3, and 4 shown in Figure 1) as examples The description is made to fully disclose the technical solution to be claimed in this application.

It is understandable that the technical solutions of this application can be used in all walks of life to enter e-commerce trading platforms, especially B2C and O2O model trading platforms. Those skilled in the art can enter e-commerce trading platforms based on the service providers engaged in the catering industry described in this application. The process of the business transaction platform can directly and without any doubt determine the way and/or process for service providers in other industries to use the technical solutions of this application to enter the e-commerce transaction platform. Therefore, the use of other industries will not be repeated in this application. The specific process for the technical solution of this application to enter the e-commerce transaction platform will be illustrated in the following text, and these examples are not specific restrictions on this application.

The technical solution proposed by this application is described in detail below:

Referring to FIG. 2, FIG. 2 is a flowchart of an object information processing method proposed by an embodiment of the present application, which is applied to a server (or server cluster) of a trading platform. As shown in Figure 2, the method includes the following steps:

S201: Obtain target requirement description information.

In the embodiment of this application, the object refers to the specific item of the service content that the service provider intends to provide. It should be pointed out that the "service content" in this application should be in a broad sense, referring to various labor activities provided by service providers to meet the needs of enterprises, public organizations or other public in the form of equivalent exchange; both It means that service providers provide consumers with physical goods, such as clothing, jewelry, cosmetics, catering, flowers, daily necessities, electrical appliances, stationery, musical instruments, etc. It can also mean that service providers provide consumers with virtual goods or consumer services, such as digital Audio-visual products, digital audio-visual entertainment, software, education and training, etc. For example, if it is a service provider engaged in catering operations, then the object refers to the different dishes provided by the restaurant/restaurant; another example is a service provider engaged in flower art business, then the object refers to the various types of flowers sold by the flower shop Fresh flowers; another example is a service provider engaged in accommodation operations, then the object refers to the various different rooms provided by the hotel, and the above service providers provide physical goods. For example, for service providers engaged in digital audio-visual operations, the target refers to the various digital music, digital movies, and peripheral entertainment items provided by the audio-visual entertainment operators to consumers; for example, for service providers engaged in teaching-assisted training, the target refers to A variety of different training service projects provided by teaching and training institutions to consumers; another example is a service provider engaged in software development, then the target refers to a variety of software provided by software developers to consumers, and the above service providers What is provided is virtual goods or consumer services.

The target demand description information refers to the information describing the specific service content. Of course, the specific service content information obtained should match the expectations of the corresponding service provider, that is, the service content that the corresponding service provider intends to provide, such as The service providers in the catering industry usually cannot involve teaching and training, and the catering industry is also divided into many categories. For example, the main business items of the service providers 1, 3, and 4 described in Figure 1 are: hot pot, cakes And coffee. For example, the wish of service provider 1 as shown in Figure 1: "I plan to open a Chongqing hot pot restaurant of about 100 square meters (square meters) near an office building in a certain business district." The wish text of this service provider 1 is the service provider 1. The object requirement description information. The target demand description information reflects the various characteristics of the service content that the service provider intends to provide from multiple different dimensions to a certain extent. For example, the service provider 1’s target demand description information includes the characteristics of the three dimensions of the store location dimension (near the office building in the business district), the audience dimension (office workers/business people), and the store area dimension (100㎡). These three dimensions Characterizes the three characteristics of the service content (hot pot) that the service provider 1 intends to provide.

In a specific implementation, when a new service provider is ready to enter the trading platform, that is, when the server (or server cluster) of the trading platform receives the entry from the service provider (mobile APP and desktop client provided by the trading platform for service providers) , Interactive web interface, etc.) when the incoming request is entered, the server (or server cluster) of the trading platform returns to the service provider’s entry entry prompt information for instructing and guiding the service provider to enter the object’s demand description information. For example, the following is output in an output box on the display screen of the electronic device showing the service provider’s entrance: "Please tell me where and what kind of shop you are going to open. For example, I want to open a 100 shop near an office building. For a Chongqing hot pot restaurant about ㎡, of course, you can also enter the above information in the input box below", and you can also perform the following steps: check whether there is an acoustic energy conversion output device on the electronic equipment showing the service provider’s entry entrance, if so, Then, the content in the above output box can also be broadcast by voice to remind the service provider to input the description information of the object's requirement.

Figure 3 shows a schematic diagram of the operation of obtaining object demand description information in an object information processing method proposed by an embodiment of the present application. Take the APP in the mobile terminal 301 as an example. The service provider initiates a registration request to the server 302 (or server cluster) of the trading platform through the APP in the mobile terminal 301, and the server 302 (or server cluster) of the trading platform returns the above to the mobile terminal 301 Content: "Please tell me where and what kind of shop you are going to open. For example, I want to open a Chongqing hot pot restaurant of about 100 square meters near the office building. Of course, you can also enter the above information in the input box below" and move it The output box 3011 of the touch screen of the terminal shows the above content, and the input box 3012 is shown for the service provider to perform handwriting input, and the above content is played synchronously through the speaker of the mobile terminal 301. When the microphone of the mobile terminal 301 picks up the service provider The operator’s voice: “I’m going to open a 300㎡ Chongqing hot pot restaurant for middle-income people near the office building”, or the operator of the service provider detected the content entered in the input box 3012: “I want to be in the office building A 300㎡ Chongqing hot pot restaurant for middle-income people is opened nearby", the mobile terminal 301 responds to the voice of the operator of the service provider or the text entered in the input box 3012, and sends it (the picked up service) The voice of the operator of the merchant or the text content entered in the input box 3012) is sent to the server 302 (or server cluster) of the trading platform, so that the server of the trading platform completes the acquisition of the service provider’s object demand description information.

Of course, in other embodiments, it can also be based on the setting of dimensions, in a form of guidance, to assist service providers step by step in completing the input of object demand description information from various dimensions, where the dimensions include at least one of the following: business dimension, audience Dimension, object price dimension, object sales dimension, store location dimension, store area dimension. In the specific implementation, first show the following content in the output box 3011: "Please speak or enter in the input box below what kind of service you are going to open a store, such as a hot pot restaurant", and prompt the operator of the service provider to determine the service Content; after the operator of the service provider determines the service content, refresh the output box 3011 to show the following content: "Please speak or enter in the input box below where you are going to open a XX (hot pot) shop, for example, near an office building ", prompt the operator of the service provider to determine the store location dimension; after the operator of the service provider determines the store location dimension, refresh the output box 3011 again to show the following content: "Please speak or enter yours in the input box below The main audience groups of XX (hot pot) stores, such as high-income people", prompt the operator of the service provider to determine the audience dimension; the acquisition of other dimensions of information in the target demand description information is similar to the input method of the aforementioned dimension information, here It will not be repeated here; this guides the service provider step by step to complete the input of the object requirement description information, and then the mobile terminal 301 feeds back the acquired complete object requirement description information to the server 302. Of course, it is also possible that the mobile terminal 301 feeds back the information of one dimension in the object requirement description information to the server 302 every time it acquires, and there is no specific limitation in this application.

It should be noted that, in the embodiments of the present application, the service provider portal is exemplified by the APP in the mobile terminal 301 shown in FIG. 3 as an example, but this is not a limitation of the present application. Technicians can directly and without doubt determine the way in which service providers use the technical solutions of this application to enter the e-commerce trading platform through other service providers (desktop clients and interactive web interfaces provided by the trading platform for service providers) And/or process.

Continuing to refer to FIG. 2, after step S201 is performed, step S202 is continued to perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions.

With reference to FIG. 3, if the object requirement description information acquired by the server 302 (or server cluster) is the voice information or text information input by the operator of the service provider through the mobile terminal 301.

If the object requirement description information acquired by the server 302 (or server cluster) is voice information, then in step S202, the server 302 (or server cluster) performs the following steps S401 and S402. Correspondingly, step S202 includes:

S401: Convert the object requirement description information into a corresponding object requirement description text.

That is, the server 302 (or server cluster) converts the received voice message input by the service provider sent by the mobile device 301 into an object requirement description text in text form for feature extraction in the subsequent step S402. For example, the voice message "I want to open a 300m2 Chongqing hot pot restaurant near the office building for middle-income people" input by the service provider mentioned above is converted into text to generate a description of the target's needs.

In an optional embodiment, step S401 is implemented by the following sub-steps S501 and S502:

S501: Obtain an object requirement description voice.

That is, after the server 302 (or server cluster) receives the service request initiated by the mobile terminal 301, it returns to the mobile terminal 301 prompt information for instructing and guiding the operator of the service provider to input the object demand description information, and the service provider moves according to the prompt information. A voice message is input into the terminal 301, such as "I want to open a 300 square meters Chongqing hot pot restaurant near the office building for middle-income people" as mentioned above. The server receives the voice message and obtains the voice for describing the target's needs. Specifically, please refer to the relevant description of prompting the service provider to input the target demand description information in step S201 above, which will not be repeated here.

Then, step S502 is performed to perform voice recognition on the object requirement description voice to obtain the object requirement description text.

After the server 302 (or server cluster) receives the above-mentioned object demand description voice "I want to open a 300㎡ Chongqing hot pot restaurant for middle-income people near the office building", it performs speech recognition (Speech Recognition) and converts it In the form of text, the description text of the object requirement is generated. Among them, the speech recognition technology can refer to related technologies in the prior art.

After step S401 is performed, step S402 is then performed to analyze the semantics of the target requirement description text to obtain feature data that matches at least part of the multiple dimensions, and the multiple dimensions include at least one of the following : Business format dimension, audience dimension, object price dimension, object sales dimension, store location dimension, store area dimension.

The format dimension refers to the time when the service provider is ready to open business. For example, the usual business hours of hot pot restaurants are from 10 am to 23:00, while the usual business hours of flower shops are from 8 am to 16 pm, while the usual business hours of teaching and auxiliary training institutions are From Monday to Friday from 15:00 to 22:00 in the evening, from Saturday to Sunday from 8 am to 22:00 in the evening, digital music entertainment is mainly operated online, and its business The time can be 24 hours a day.

The audience dimension refers to the consumer groups that the service provider intends to provide. For example, the main consumer group for cosmetics is adult young women, and the main consumer group for stationery is students, etc., which will not be listed in detail. Moreover, consumers need different consumer goods due to their identity, social status, wealth and ability, etc. Taking automobiles as an example, Volkswagen and Buick are mainly for low- and middle-income groups, and Mercedes-Benz and Audi are mainly for high- and middle-income groups. Ferrari, Maserati, etc. are mainly for ultra-high-income groups. Therefore, the identity, social status, and spending power of consumers can also be considered when determining the audience dimensions, so that the service content provided by the service provider matches the corresponding consumer group.

The target price dimension refers to the price of each specific service item in the service content that the service provider intends to provide and/or the grade range to which the price belongs. For example, in the catering industry, the target price dimension refers to the price of each dish and/or the grade range to which the price belongs. Take the two famous Sichuan dishes of boiled cabbage and Mapo tofu as examples. Among them, boiled cabbage is due to the complexity of its production process. , Making its price breakdown higher than that of Mapo Tofu, which is more acceptable to the public. For another example, in the teaching assistant training industry, the target price dimension refers to the price of each different training course and/or the price range to which the price belongs. For example, the price of oriented training courses for business managers is usually higher than that for primary school students’ hobbies and hobbies. course.

The target sales dimension refers to the sales volume of each specific service item of the service content that the service provider intends to provide. For example, in the catering industry, the target sales dimension refers to the sales volume of various dishes. For example, the above-mentioned Mapo Tofu is more popular than the boiled cabbage because of its price. For another example, in the teaching assistant training, the audience of the above-mentioned hobby tutoring training courses for elementary school students is wider than that of the targeted training courses for business managers, making the hobby tutoring training courses for elementary school students more popular.

The store location dimension refers to the specific address of the new store that the service provider plans to open. For example, restaurants are mostly opened in the living area or near the business area, and the number of fast food restaurants in the business area is also significantly more than that in the living area. The teaching and auxiliary training institutions are mostly located in areas closer to the school.

The store area dimension refers to the store area of the physical store where the service provider plans to open a new store. In many industries, a physical business address is required to obtain a business license to permit service providers to operate registered businesses.

It should be noted that the above descriptions of dimensions are all exemplary introductions. Because of the various industries, they are different, complicated, and numerous, so I will not introduce them here. Therefore, the above examples are not exhaustive. It is not a restriction on this application. When the above six dimensions correspond to different industries, those skilled in the art can determine the characteristic data corresponding to each dimension after analyzing the data of the industry.

Semantic analysis is NLP (Natural Language Processing), which is a branch of natural speech processing technology. It uses NLP technology to perform semantic analysis on the object requirement description text generated in step S401 above, and extracts different information from the above object requirement description information. The dimension represents the characteristic data of the service content of the service provider. The NLP technology can refer to related technologies in the prior art.

The following is the description text of the above object requirements: "I want to open a 300 square meters Chongqing hot pot restaurant for middle-income people near the office building". After the above text is segmented and marked, the following sentence structure is obtained:

I want to be

Office building

Open a family

300㎡

Oriented towards

Middle-income people

of

Chongqing

hot pot

shop

After extraction, the following keywords were extracted: "near office building", "300 square meters", "middle-income group" and "hot pot"; among them, "hot pot" means that the service content that the service provider intends to provide is hot pot catering, and " "Near office building", "300㎡", and "for middle-income people" are represented by three dimensions: store location (near office building), audience (office workers/business people, middle-income people), and store area (300㎡) The content of the service (hot pot) that the service provider intends to provide to consumers.

If the object requirement description information acquired by the server 302 (or server cluster) is itself text information, in step S202, the server 302 (or server cluster) only needs to perform step S402. For details, please refer to the above related step S402 The related description of, I won’t repeat it here.

After step S202 is performed, step S203 is performed to input feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, where the object recommendation model is based on sample feature data of multiple dimensions and all The preset weights of each of the multiple dimensions are obtained by training the preset model.

The preset model can be a neural network model selected in advance, for example, a mathematical model obtained by a logistic regression operation according to a big data analysis technology.

In an exemplary embodiment, the object recommendation model can be implemented through the following steps S601 and S602.

S601: Obtain sample feature data of multiple dimensions corresponding to each group of object information in multiple groups of object information.

Multi-group object information refers to information describing objects in different groups. Taking the catering industry as an example, it is usually necessary to obtain the following three groups of information describing the object: the name of the dish and the corresponding price data (such as boiled fish & price, potato shreds & price, kung pao chicken & price...) , Merchant (ie service provider) data (Sichuan/Hunan/Lu/Guangdong, business district/living area, Chongqing hot pot/Chinese food/Western food, store area...), sales data (best seller: boiled fish..., slow-moving: potatoes Silk). Taking teaching assistant training as an example, it is usually necessary to obtain the following three groups of description object information: course type and corresponding price data (basic piano/intermediate/advanced course & price, swimming elementary/intermediate/advanced course & price, Calligraphy elementary/intermediate/advanced courses & prices), training institution data (prenatal education/children/youth/adults, online/offline, near school/business district, number of students that can accommodate....), sales data (best-selling: piano lessons) ......, slow sales: calligraphy class......). Similar to other industries, those skilled in the art can make flexible settings according to actual needs.

The sample feature data of multiple dimensions corresponding to each group of object information refers to the sample feature data of different dimensions in the information describing the object of each group. For example, the above-mentioned group "Dish Name and Corresponding Price Data" describes the object information including the sample feature data of the "object price dimension" where the object is the dish; the group "Merchant Data" describes the object information including the object is The sample feature data of the dishes' "store location dimension", "store area dimension", and "audience dimension (cuisine can indirectly determine the corresponding audience)"; the group "sales data" describes the object's information including the object is the dish Sample characteristic data of the "object sales dimension".

In specific implementation, a large number of relevant data information of the service content provided by different service providers are captured in advance from the cloud network, and big data analysis is performed on these data information to obtain a large amount of information that characterizes the service provider from multiple different dimensions. Various sample characteristic data of the characteristics of the service content. Taking the catering industry as an example, the takeaway platform collects all the data of the merchants that have settled on the platform in advance, sorts them according to the above groups, and then conducts big data analysis to extract each group of dish information from multiple groups of dish information Corresponding sample feature data in multiple dimensions.

S602. Use the sample feature data of multiple dimensions corresponding to each of the multiple sets of object information as a training set, and combine the preset weights of each of the multiple dimensions to train a preset model to obtain the object recommendation model .

After acquiring the sample feature data of the multiple dimensions, it needs to be analyzed again to determine the initial weight of the sample feature data of each dimension, that is, the preset weight. For example, in the catering industry, whether the dishes sell well relative to the location of the store will more directly affect the profitability of the merchant. That is to say, in the catering industry, the weight of the object sales dimension is usually greater than the store location dimension. In the flower industry, the distance between the shop location and the flower grower and the predetermined market (such as the living area) relative to the size of the shop area will be more directly related to the freshness of the flowers and the computing cost, which is directly related to the profit of the merchant. In the floral industry, the weight of the store location dimension is usually greater than the store area dimension. The preset weights of the sample feature data of each dimension of specific service items (ie objects) corresponding to the service content provided by service providers in different industries can be determined by performing big data analysis on the sample feature data of the above multiple dimensions.

However, the preset weights of each dimension determined above are usually not accurate. At this time, the sample feature data of the multiple dimensions corresponding to the above multiple sets of object information are used as the training set, and the preset weights of each of the multiple dimensions are combined to train the preset model, so that the preset model is The preset weights of each dimension are continuously optimized iteratively, and finally the object recommendation model is obtained.

Recommended object information refers to the recommendation result output by the object recommendation model that matches the object demand description information provided by the service provider after the sample data is input into the object recommendation model, that is, at least one specific service item that matches the content of the service to be provided And the relevant information of each service item. Taking the catering industry as an example, it means that the object recommendation model outputs various recommended dishes and recommended prices based on the characteristic data in the description information of the merchant’s dish demand, such as the recommended dish information shown in Table 1 below. Taking the teaching assistant training industry as an example, it means that the target recommendation model outputs various recommended courses and recommended pricing according to the characteristic data in the description information of the course requirements of the merchants. For example, composition training for grades one to three in elementary school: RMB 80.00 per hour, composition training for grades four to six in elementary school: RMB 100.00 per hour, and so on.

After training the object recommendation model, when the service provider enters the trading platform, after obtaining the feature data of multiple dimensions, input the feature data of multiple dimensions into the object recommendation model. After the object recommendation model is processed, it becomes the service provider Output the recommended object information that matches the service content that the service provider intends to provide. For example, the above feature data "near office building", "300㎡", "middle-income group" and "hot pot", after inputting the object recommendation model, output the recommended object (dish) information shown in Table 1 below:

Table 1, Recommended dishes information

Butter Spicy Base (Serves 2)	￥85.00 Yuan
Qingyou Spicy Base (Serves for 2)	￥75.00 Yuan
Tomato base (for 2 people)	￥45.00 yuan
Bacteria soup base (for 2 people)	￥55.00 Yuan
Fat cow	￥19.00 yuan
Fat sheep	￥22.00 yuan
Hairy belly	￥29.00 yuan
......	......

S204: Output the recommended object information.

That is, the output results of the recommendation model are fed back to the service provider's entry portal or stored in the data service center of the trading platform.

When outputting the recommended object information to the service provider’s entry portal, the output results (objects) can be sorted and output based on different dimensions. For example, in the catering industry, the target sales dimension is used as a reference for sorting, that is, according to the highest sales volume. To the lower order, output the recommended dishes information, and put the dishes with better sales in the higher position in the queue.

Using the technical solution of this application, the server (or server cluster) of the trading platform obtains a large amount of sample characteristic data, which not only characterizes the characteristics of the service provider’s service content, but also reflects from multiple different angles (ie dimensions) The characteristics of the service content; and assign preset weights to each dimension. Use sample feature data with multiple dimensions and preset weights assigned to each dimension to train the preset model, for example, to train a pre-selected neural network model to obtain an object recommendation model. The object recommendation model can be based on the service provider The relevant information of the service content to be provided is recommended, and at least one specific service item and relevant information of each service item that match the service content to be provided are recommended to the service provider.

When a service provider is ready to enter the trading platform, the server (or server cluster) of the trading platform obtains the relevant information of the service content that the service provider is going to provide, that is, the object demand description information, and then extracts the characteristics of the service provider’s object demand description information. Extract feature information related to the service content from the description information and characterize the service content from different perspectives, that is, extract feature data of multiple dimensions from the description information of the target demand, and then extract the multiple dimensions of the service content from the description information. The characteristic data is input to the pre-trained object recommendation model, and after the object recommendation model is processed, it outputs at least one specific service item and related information of each service item that matches the service content that the service provider intends to provide, that is, recommendation object information.

Using the technical solution provided by this application, it is only necessary to save the recommended object information output by the object recommendation model in the data center of the trading platform, thereby helping the service provider to enter some data information related to the service content that it is preparing to provide into the trading platform , To reduce the amount of learning for service providers to learn the trading platform and the workload of entering data information related to the service content, so as to assist service providers to quickly and efficiently complete the work of entering the trading platform.

A diagram of an object information processing method proposed in another embodiment of the present application is applied to a server (or server cluster) of a trading platform. The method includes the following steps:

S701: Obtain target requirement description information.

It is the same as the above step S201, please refer to step S201 for details.

S702: Perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions.

It is the same as the above step S202, please refer to step S202 for details.

S703: Perform feature extraction on the object requirement description information to obtain respective designated weights for multiple dimensions.

The designated weight is based on the business operated by the service provider (that is, the service content provided), and the designated weight for the characteristic data of each dimension in the description information of the service provider.

Different service providers may have different levels of attention to different dimensions. For example, although the service providers 1, 3, and 4 described in Figure 1 are all operating in the catering industry, the service provider 4’s business is coffee. Service providers 1 who operate hot pot and service providers 3 who operate cakes prefer to open their stores near business districts, especially in areas where there are more young office workers, that is, they pay more attention to the dimension of store locations. For the object of this kind of special service content, feature extraction can be performed on the object requirement description information to obtain the respective designated weights of multiple dimensions.

It should be pointed out that the above steps S702 and S703 can be executed in parallel or serially; in serial execution, step S702 can be executed first, and then step S703; of course, step S703 can be executed first, and then step S702 can be executed. ；

S704. Inputting feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information includes: inputting feature data of multiple dimensions and respective specified weights of the multiple dimensions into the pre-trained object recommendation model to obtain Recommended object information; wherein the object recommendation model is obtained by training a preset model based on sample feature data of multiple dimensions and preset weights of each of the multiple dimensions.

Then, the feature data of multiple dimensions and the respective specified weights of the multiple dimensions are input into the pre-trained object recommendation model. The object recommendation model will consider the new influence factor, namely the specified weight, when processing the object recommendation model. The recommended information is more in line with the expectations of the service provider.

The processing process of the object recommendation model is similar to the above step S203, please refer to step S203 for details.

S705: Output the recommended object information.

Similar to the above step S204, please refer to step S204 for details.

Referring to FIG. 5, FIG. 5 is a flowchart of an object information processing method proposed by another embodiment of the present application, which is applied to a server (or server cluster) of a trading platform. As shown in Figure 5, the method includes the following steps:

S801: Obtain target requirement description information.

It is the same as the above step S201, please refer to step S201 for details.

S802: Perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions.

It is the same as the above step S202, please refer to step S202 for details.

S803: Input feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, where the object recommendation model is based on sample feature data of multiple dimensions and presets of each of the multiple dimensions. The weight is obtained by training the preset model.

It is the same as the above step S203, please refer to step S203 for details.

S804: Output the recommended object information.

It is the same as the above step S204, please refer to step S204 for details.

S805: Obtain object adjustment information, where the object adjustment information is used to add new object information to the recommended object information or delete part of the object information from the recommended object information.

There may still be some errors in the recommended target information, and there are some gaps between the service provider’s expectations and the service provider’s manual correction. Take the catering industry as an example. For example, information about dishes expected by certain service providers is missing, or information about dishes not expected by service providers is superfluous, which needs to be manually added or deleted by the operator of the service provider. Of course, the “manual "It should be in a broad sense. It can also be the same as when obtaining object description information, through voice commands or clicking the operation buttons shown on the screen to perform corresponding operations to enter adjustment information into the service provider’s entry entrance to instruct to add new objects. The information is added to the recommended object information, or part of the object information is deleted from the recommended object information.

The adjustment information should also be broad. For example, it can also be used to assign alias information to the object. For example, in catering, "minced meat vermicelli" is also called "ant on the tree", "red bean milk tea" is also called "Acacia milk tea", etc., or used for adjustment Information about the price of the object, such as adjusting the price of dishes.

S806: Adjust the recommended object information according to the object adjustment information.

After obtaining the above adjustment information, according to the specific adjustment instructions in the adjustment information, corresponding adjustment operations are performed on the recommended object information, such as adding missing objects, deleting redundant objects, assigning aliases, and so on.

S807: Enter the adjusted recommended object information into the object database.

The adjusted recommendation object information will be stored in the data center of the trading platform. For example, the adjusted recommended dishes are stored in the data center of the food delivery platform.

Referring to FIG. 6, FIG. 6 is a flowchart of an object information processing method proposed by another embodiment of the present application, which is applied to a server (or server cluster) of a trading platform. As shown in Figure 6, the method includes the following steps:

S901: Obtain target requirement description information.

It is the same as the above step S801, please refer to step S801 for details.

S902: Perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions.

It is the same as the above step S802, please refer to step S802 for details.

S903. Input feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, where the object recommendation model is based on sample feature data of multiple dimensions and presets for each of the multiple dimensions. The weight is obtained by training the preset model.

It is the same as the above step S803, please refer to step S803 for details.

S904: Output the recommended object information.

It is the same as the above step S804, please refer to step S804 for details.

S905: Obtain object adjustment information, where the object adjustment information is used to add new object information to the recommended object information or delete part of the object information from the recommended object information.

It is the same as the above step S805, please refer to step S805 for details.

S906: Adjust the recommended object information according to the object adjustment information.

It is the same as the above step S806, please refer to step S806 for details.

S907: Enter the adjusted recommended object information into the object database.

It is the same as the above step S807, please refer to step S807 for details.

S908: Obtain new sample feature data of multiple dimensions generated based on the adjusted information of the recommended object.

Perform big data analysis on the above-mentioned recommended object information (or adjusted recommended object information) last stored in the data center to generate new sample feature data. Its implementation is similar to step S601, except that the data used in step S601 is sample data generated after analysis of data captured by the cloud, and the data used in step S908 is the information of the recommended object (or the adjusted information of the recommended object) ) The new sample feature data generated has higher accuracy. Therefore, for the implementation of step S908, please refer to step S601, which will not be repeated here.

S909: Update the object recommendation model according to the generated new sample feature data in multiple dimensions.

The new sample feature data used is fed back to the object recommendation model, so that the object recommendation model performs in-depth learning, so that the object recommendation model is more optimized and has better generalization performance.

Referring to FIG. 7, FIG. 7 is an application schematic diagram of an object information processing method proposed by an embodiment of the present application, which is applied to a server (or server cluster) of a trading platform. As shown in Figure 7, the method includes the following steps:

S1001: Obtain sample feature data in multiple dimensions corresponding to each group of dish information in the multiple sets of dish information.

That is, grab a large number of relevant data of catering service providers from the cloud, divide them into dish data groups, merchant data groups, and sales data groups, and perform multi-dimensional feature extraction on each group of data to generate sample data and store it in the sample database.

S1002: Use the sample feature data of multiple dimensions corresponding to each of the multiple sets of dish information as a training set, and combine the preset weights of each of the multiple dimensions to train a preset model to obtain the dish recommendation model .

The data analysis platform calls the sample data in the sample database, performs feature analysis, obtains sample feature data of multiple dimensions and preset weights of each of the multiple dimensions, and trains the preset model.

When a merchant needs to settle in the food delivery trading platform:

S1003: Obtain the voice of the demand description of the dish.

The service provider enters the entrance voice picking up the service provider’s “dish demand description voice” described in the form of speech, and records the voice into the smart engine.

S1004: Perform voice recognition on the cuisine requirement description voice to obtain the cuisine requirement description text.

S1005: Analyze the semantics of the dish requirement description text to obtain feature data that matches at least a part of the multiple dimensions, and the respective designated weights of the multiple dimensions, the multiple dimensions including at least one of the following: Business format dimension, audience dimension, dish price dimension, dish sales dimension, store location dimension, store area dimension.

S1006: Input the feature data of multiple dimensions and the respective designated weights of the multiple dimensions into a pre-trained dish recommendation model to obtain recommended dish information.

S1007: Output the recommended dish information.

S1008: Obtain dish adjustment information, where the dish adjustment information is used to add new dish information to the recommended dish information, or delete part of the dish information from the recommended dish information.

S1009: Adjust the recommended dish information according to the dish adjustment information.

S1010: Enter the adjusted recommended dish information into the dish database (data center).

S1011: Obtain new sample feature data in multiple dimensions generated based on the adjusted recommended dish information.

S1012: Update the dish recommendation model according to the generated new sample feature data in multiple dimensions.

The above steps have been described in detail in the above embodiments, and will not be repeated here. For details, please refer to the above corresponding embodiments.

Based on the same inventive concept, an embodiment of the present application provides an object information processing device. Referring to FIG. 8, FIG. 8 is a schematic diagram of an object information processing device provided by an embodiment of the present application. As shown in FIG. 8, the device 110 includes:

The first information obtaining module 111 is used to obtain the object requirement description information;

The first feature extraction module 112 is configured to perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions;

The input module 113 is configured to input feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, where the object recommendation model is based on sample feature data of multiple dimensions and each of the multiple dimensions The preset weight of the dimension is obtained by training the preset model;

The output module 114 is configured to output the recommended object information.

Optionally, the first feature extraction module includes:

A conversion unit, configured to convert the object requirement description information into a corresponding object requirement description text;

The semantic analysis unit is used to analyze the semantics of the target demand description text to obtain feature data that matches at least part of the multiple dimensions, and the multiple dimensions include at least one of the following: a format dimension, an audience dimension , Object price dimension, object sales dimension, store location dimension, store area dimension.

Optionally, the conversion unit includes:

The voice acquisition subunit is used to obtain the target demand description voice;

The voice recognition subunit is used to perform voice recognition on the object requirement description voice to obtain the object requirement description text.

Optionally, the device further includes:

An adjustment information obtaining module, configured to obtain object adjustment information, the object adjustment information being used to add new object information to the recommended object information, or to delete part of the object information from the recommended object information;

An adjustment module, configured to adjust the recommended object information according to the object adjustment information;

The entry module is used to enter the adjusted recommended object information into the object database.

Optionally, the device further includes:

The second feature extraction module is configured to obtain new sample feature data of multiple dimensions generated based on the adjusted information of the recommended object;

The update module is used to update the object recommendation model according to the generated new sample feature data in multiple dimensions.

Optionally, the first feature extraction module is further configured to perform feature extraction on the object requirement description information to obtain respective designated weights for multiple dimensions;

The input module is also used to input feature data of multiple dimensions and the respective specified weights of the multiple dimensions into a pre-trained object recommendation model to obtain recommended object information.

Optionally, the device further includes:

The third feature extraction module is used to obtain sample feature data of multiple dimensions corresponding to each group of object information in the multiple groups of object information;

The training module is configured to use the sample feature data of multiple dimensions corresponding to each of the multiple sets of object information as a training set, and combine the preset weights of each of the multiple dimensions to train the preset model to obtain the Object recommendation model.

As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the part of the description of the method embodiment. The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments can be referred to each other.

The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments. Those of ordinary skill in the art can understand and implement it without creative work.

The various component embodiments of the present invention may be implemented by hardware, or by software modules running on one or more processors, or by a combination of them. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all of the functions of some or all of the components in the computing processing device according to the embodiments of the present invention. The present invention can also be implemented as a device or device program (for example, a computer program and a computer program product) for executing part or all of the methods described herein. Such a program for realizing the present invention may be stored on a computer-readable medium, or may have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

Based on the same inventive concept, another embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the method described in any of the foregoing embodiments of the present application are implemented. .

Based on the same inventive concept, another embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The steps in the method described in the embodiment.

For example, FIG. 9 shows an electronic device that can implement the method according to the present invention, such as a computing processing device. The electronic device traditionally includes a processor 1010 and a computer program product in the form of a memory 1020 or a computer-readable medium. The memory 1020 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read Only Memory), EPROM, hard disk, or ROM. The memory 1020 has a storage space 1030 for executing program codes 1031 of any method steps in the above methods. For example, the storage space 1030 for program codes may include various program codes 1031 respectively used to implement various steps in the above method. These program codes can be read from or written into one or more computer program products. These computer program products include program code carriers such as hard disks, compact disks (CDs), memory cards, or floppy disks. Such a computer program product is usually a portable or fixed storage unit as described with reference to FIG. 10. The storage unit may have storage segments, storage spaces, etc. arranged similarly to the memory 1020 in the computing processing device of FIG. 9. The program code can be compressed in a suitable form, for example. Generally, the storage unit includes computer-readable code 1031', that is, code that can be read by a processor such as 1010, which, when run by a computing processing device, causes the computing processing device to execute the method described above. The various steps.

Those skilled in the art should understand that the embodiments of the embodiments of the present application may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present application may adopt the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

The embodiments of this application are described with reference to the flowcharts and/or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of this application. It should be understood that each process and/or block in the flowchart and/or block diagram, and the combination of processes and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded processors, or other programmable data processing terminal equipment to generate a machine, so that instructions executed by the processor of the computer or other programmable data processing terminal equipment A device for realizing the functions specified in a flow or multiple flows in the flowchart and/or a block or multiple blocks in the block diagram is generated.

These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal equipment to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including the instruction device. The instruction device implements the functions specified in one process or multiple processes in the flowchart and/or one block or multiple blocks in the block diagram. These computer program instructions can also be loaded on a computer or other programmable data processing terminal equipment, so that a series of operation steps are executed on the computer or other programmable terminal equipment to produce computer-implemented processing, so that the computer or other programmable terminal equipment The instructions executed above provide steps for implementing functions specified in a flow or multiple flows in the flowchart and/or a block or multiple blocks in the block diagram.

The above provides a detailed introduction to the object information processing method, device, storage medium, and electronic equipment provided by the present application. Specific examples are used in this article to illustrate the principles and implementations of the present application. The description of the above embodiments is only It is used to help understand the methods and core ideas of this application; at the same time, for those of ordinary skill in the art, according to the ideas of this application, there will be changes in the specific implementation and the scope of application. In summary, this The content of the description should not be construed as a limitation on this application.

The “one embodiment”, “an embodiment” or “one or more embodiments” referred to herein means that a specific feature, structure, or characteristic described in combination with the embodiment is included in at least one embodiment of the present invention. In addition, please note that the word examples "in one embodiment" here do not necessarily all refer to the same embodiment.

In the instructions provided here, a lot of specific details are explained. However, it can be understood that the embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures, and technologies are not shown in detail, so as not to obscure the understanding of this specification.

In the claims, any reference signs placed between parentheses should not be constructed as a limitation to the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of multiple such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims listing several devices, several of these devices may be embodied in the same hardware item. The use of the words first, second, and third, etc. do not indicate any order. These words can be interpreted as names.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions recorded in the foregoing embodiments are modified, or some of the technical features thereof are equivalently replaced; these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (11)

1. An object information processing method, characterized in that the method includes:

   Obtain the target demand description information;

   Perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions;

   Inputting feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, wherein the object recommendation model is based on sample feature data of multiple dimensions and preset weights of each of the multiple dimensions, Obtained by training the preset model;

   Output the recommended object information.
2. The method according to claim 1, characterized in that, performing feature extraction on the object requirement description information to obtain feature data of multiple dimensions, comprising:

   Converting the object requirement description information into a corresponding object requirement description text;

   Analyze the semantics of the target demand description text to obtain feature data that matches at least some of the multiple dimensions. The multiple dimensions include at least one of the following: business format dimension, audience dimension, target price dimension, object Sales dimension, store location dimension, store area dimension.
3. The method according to claim 2, wherein converting the object requirement description information into a corresponding object requirement description text comprises:

   Obtain the voice of object demand description;

   Perform voice recognition on the object requirement description voice to obtain the object requirement description text.
4. The method according to claim 1, wherein after outputting the recommended object information, the method further comprises:

   Obtaining object adjustment information, where the object adjustment information is used to add new object information to the recommended object information, or delete part of the object information from the recommended object information;

   Adjust the recommended object information according to the object adjustment information;

   Enter the adjusted recommended object information into the object database.
5. The method according to claim 4, characterized in that, after entering the adjusted information of the recommended object into the object database, the method further comprises:

   Obtaining new sample feature data of multiple dimensions generated based on the adjusted information of the recommended object;

   The object recommendation model is updated according to the generated new sample feature data in multiple dimensions.
6. The method according to claim 1, wherein the method further comprises:

   Perform feature extraction on the object requirement description information to obtain respective designated weights for multiple dimensions;

   Input the feature data of multiple dimensions into the pre-trained object recommendation model to obtain recommended object information, including:

   The feature data of multiple dimensions and the respective specified weights of the multiple dimensions are input into a pre-trained object recommendation model to obtain recommended object information.
7. The method according to any one of claims 1 to 6, characterized in that, before inputting feature data of multiple dimensions into a pre-trained object recommendation model, the method further comprises:

   Obtain sample feature data of multiple dimensions corresponding to each group of object information in multiple groups of object information;

   Using the sample feature data of the multiple dimensions corresponding to each of the multiple sets of object information as a training set, and combining the preset weights of each of the multiple dimensions, the preset model is trained to obtain the object recommendation model.
8. An object information processing device, characterized in that the device includes:

   The first information obtaining module is used to obtain the object requirement description information;

   The first feature extraction module is used to perform feature extraction on the object requirement description information to obtain feature data of multiple dimensions;

   The input module is used to input feature data of multiple dimensions into a pre-trained object recommendation model to obtain recommended object information, wherein the object recommendation model is based on sample feature data of multiple dimensions and each of the multiple dimensions The preset weight of is obtained by training the preset model;

   The output module is used to output the recommended object information.
9. A computer-readable storage medium with a computer program stored thereon, wherein the program is executed by a processor to implement the steps in the method according to any one of claims 1-7.
10. An electronic device comprising a memory, a processor, and a computer program stored on the memory and capable of running on the processor, wherein the processor implements the method according to any one of claims 1-7 when executed step.
11. A computer program comprising computer readable code, which when the computer readable code runs on a computing processing device, causes the computing processing device to execute the object information processing method according to any one of claims 1-7 .

Images