Gas operation method and internet of things system based on intelligent gas call center

Abstract

The embodiment of the specification provides a gas operation method and an Internet of Things system based on a smart gas call center. The method comprises the following steps: acquiring gas user call data of the call center; predicting demand information of different types of users based on the gas user call data, wherein the demand information at least comprises gas product demand and gas service demand; determining and pushing gas operation push features based on the demand information of different types of users, wherein the gas operation push features comprise push type features and push content features.

Description

TECHNICAL FIELD

[0001] The present specification relates to the field of gas operation, in particular to a gas operation method based on a smart gas call center and an Internet of Things system. BACKGROUND

[0002] At present, with the rapid development of the gas industry, the customer group faced by gas companies is becoming more and more complex, gas application scenarios are becoming more and more diversified, and gas smart operation is gradually implemented and applied. Among them, the gas call center is a key link for gas companies to obtain customer demand and provide gas services. The purpose of gas operation is to improve user satisfaction with gas use, so how to improve gas problem solving efficiency through the gas call center is a technical problem to be solved in the field.

[0003] In view of this, CN113283915A discloses a gas call center service processing method, system, terminal and storage medium, which obtains user demand on the client side online and matches dedicated service personnel to solve user demand problems online, which can improve the efficiency of solving gas problems for users, but does not involve predicting user demand and targeted pushing.

[0004] Therefore, it is necessary to provide a gas operation method based on a smart gas call center and an Internet of Things system, which accurately predicts or determines user demand based on limited customer data of the call center, so as to improve gas operation efficiency and timely and fully meet user demand and improve user satisfaction. SUMMARY

[0005] One or more embodiments of the present specification provide a gas operation method based on a smart gas call center. The method is implemented based on a gas operation Internet of Things system of a smart gas call center, and the method comprises: obtaining gas user call data of the call center; based on the gas user call data, predicting demand information of different types of users respectively; the demand information at least includes gas product demand and gas service demand; based on the demand information of the different types of users, determining gas operation push features and performing pushing; the gas operation push features include push type features and push content features.

[0006] The one or more embodiments of the specification provide a gas operation Internet of Things system based on a smart gas call center, the system comprising a smart gas user platform, a smart gas service platform, a smart gas management platform, a smart gas sensing network platform, and a smart gas object platform, the system being configured to: acquire gas user call data of the call center; predict demand information of different types of users based on the gas user call data; the demand information comprising at least gas product demand and gas service demand; determine gas operation push features based on the demand information of the different types of users and perform the push; the gas operation push features comprising push type features and push content features.

[0007] The one or more embodiments of the specification provide a gas operation device based on a smart gas call center, the device comprising at least one processor and at least one memory; the at least one memory being configured to store computer instructions; the at least one processor being configured to execute at least part of the computer instructions to implement a gas operation method based on a smart gas call center.

[0008] The one or more embodiments of the specification provide a computer readable storage medium, the storage medium storing computer instructions, when a computer reads the computer instructions in the storage medium, the computer executes a gas operation method based on a smart gas call center. BRIEF DESCRIPTION OF DRAWINGS

[0009] The specification will be further described in the manner of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same reference numbers represent the same structures, wherein:

[0010] Figure 1 is a platform structure diagram of a gas operation Internet of Things system based on a smart gas call center according to some embodiments of the specification;

[0011] Figure 2 is an exemplary flowchart of a gas operation method based on a smart gas call center according to some embodiments of the specification;

[0012] Figure 3 is an exemplary schematic diagram of determining predicted demand information according to some embodiments of the specification;

[0013] Figure 4 is an exemplary schematic diagram of determining a replenishment scheme according to some embodiments of the specification;

[0014] Figure 5A is an exemplary schematic diagram of determining gas operation push features according to some embodiments of the specification;

[0015] Figure 5B is an exemplary schematic diagram of dividing one or more push community sub-regions. DETAILED DESCRIPTION

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings needed to be used in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structure or operation.

[0017] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, sections or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0018] Unless the context clearly indicates otherwise, the words "one", "an", "a", and / or "the" do not mean to specify a single number, but can also include a plurality. Generally, the terms "comprise" and "include" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0019] Flowcharts are used in the present specification to illustrate the operations performed by the system according to the embodiments of the present specification. It should be understood that the preceding or subsequent operations are not necessarily performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can be added to these processes, or one or more steps of the operation can be removed from these processes.

[0020] Gas has been widely used as an energy source in factories, homes, etc. The rapid growth of gas users has brought many gas demands, and thus the problem of low efficiency of gas business solutions. CN113283915A only targets the business needs reported by the client from the client, identifies the business type, matches the dedicated business personnel for processing, and does not consider the potential needs of gas users. Therefore, in some embodiments of the present specification, the demand information of different types of users is predicted according to the gas user call data, and then the gas operation push features are determined and pushed, and the demand quantity information is predicted, and the inventory scheme is determined through the call center, which is beneficial to reasonably and accurately predict or determine user demand, and more forwardly improve the efficiency of gas operation.

[0021] Figure 1is a platform structure diagram of a gas operation Internet of Things system based on a smart gas call center according to some embodiments of the present specification.

[0022] In some embodiments, the gas operation Internet of Things system based on the smart gas call center 100 can be applied to a gas operation management system and used to perform a gas operation method based on a smart gas call center.

[0023] As shown in Figure 1 The gas operation Internet of Things system based on the smart gas call center 100 can include a smart gas user platform 110, a smart gas service platform 120, a smart gas management platform 130, a smart gas sensing network platform 140, and a smart gas object platform 150.

[0024] The smart gas user platform 110 is a user-oriented platform. In some embodiments, the smart gas user platform 110 is configured as a terminal device (such as a mobile phone, a tablet computer, etc.). In some embodiments, the smart gas user platform 110 includes a gas user sub-platform, a government user sub-platform, and a regulatory user sub-platform, for providing services to gas users according to gas user needs. In some embodiments, the gas user sub-platform provides gas use-related data and gas problem solutions for gas users; the government user sub-platform provides gas operation-related data for government users; and the regulatory user sub-platform regulates the operation of the entire Internet of Things system for regulatory users.

[0025] In some embodiments, the smart gas user platform 110 can interact downwardly with the smart gas service platform 120. In some embodiments, the gas user sub-platform corresponds to and interacts with the smart gas service sub-platform to obtain safe gas use services; the government user sub-platform corresponds to and interacts with the smart gas operation service sub-platform to obtain gas operation services; and the regulatory user sub-platform corresponds to and interacts with the smart gas regulatory service sub-platform to obtain safe regulatory needs services. For example, call information of a gas user is issued to the smart gas service sub-platform, or customer service feedback information uploaded by the smart gas service sub-platform is received; gas operation management information query instructions are issued to the smart gas operation service sub-platform, or gas maintenance management information uploaded by the smart gas operation service sub-platform is received.

[0026] The smart gas service platform 120 includes a smart gas service sub-platform, a smart gas operation service sub-platform, and a smart gas regulatory service sub-platform. In some embodiments, the smart gas service sub-platform corresponds to the gas user sub-platform to provide gas equipment-related information for gas users; the smart gas operation service sub-platform corresponds to the government user sub-platform to provide gas operation-related information for government users; and the smart gas regulatory service sub-platform corresponds to the regulatory user sub-platform to provide safe regulatory-related information for regulatory users.

[0027] In some embodiments, the smart gas service platform 120 interacts with the smart gas management platform 130 downward. For example, the smart gas service platform 120 issues a gas operation management information query instruction to the smart gas data center, or receives the operation management information uploaded by the smart gas data center. The smart gas service platform 120 can also interact with the smart gas user platform 110 upward. For example, the smart gas service platform 120 receives an operation management information query instruction issued by the government user sub-platform, or uploads operation management information to the government user sub-platform.

[0028] The smart gas management platform 130 is a platform for executing smart gas operation management, including a smart customer service management sub-platform, a smart gas data center, and a smart operation management sub-platform. The information interaction with the upper smart gas service platform 120 and the lower smart gas sensing network platform 140 is all through the smart gas data center. The smart customer service management sub-platform and the smart operation management sub-platform both interact with the smart gas data center bidirectionally.

[0029] The smart customer service management sub-platform includes modules such as revenue management, business management, installation management, customer service management, message management, and customer analysis management. The customer service management module is used for analyzing and replying to customer feedback information. The message management module is used for viewing customer call, consultation, feedback, complaint, and other information, which can be sent to the customer service management module for corresponding reply processing. The customer analysis management module is used for viewing customer feedback records such as consultation, complaint, report, and evaluation, customer-related information such as gas consumption, and equipment maintenance, and generating an analysis report of the customer.

[0030] The smart operation management sub-platform includes modules such as gas quantity procurement management, gas quantity reserve management, gas consumption scheduling management, purchase and sale difference management, pipe network engineering management, and comprehensive office management. The comprehensive office management module is used for overall planning of human resources, public resources, gas equipment, daily office, and administrative affairs of operation.

[0031] In some embodiments, the smart gas management platform 130 interacts with the smart gas sensing network platform 140 downward. For example, the smart gas management platform 130 issues an instruction for obtaining gas equipment related data to the smart gas sensing network platform, or receives gas equipment related data uploaded by the smart gas sensing network platform.

[0032] In some embodiments, the smart gas management platform 130 interacts with the smart gas service platform 120 upward. For example, the smart gas management platform 130 receives a query instruction for gas operation management information issued by the smart gas service platform, or uploads gas operation management information to the smart gas service platform.

[0033] In some embodiments, the gas operation Internet of Things system further comprises a smart gas sensing network platform 140. The smart gas sensing network platform 140 comprises a gas indoor equipment sensing network sub-platform and a gas pipeline equipment sensing network sub-platform. The gas indoor equipment sensing network sub-platform and the gas pipeline equipment sensing network sub-platform each comprise a network management, a protocol management, an instruction management, and a data analysis module. Among them, the gas indoor equipment sensing network sub-platform corresponds to the gas indoor equipment object sub-platform, and is used to obtain related data of indoor equipment; the gas pipeline equipment sensing network sub-platform corresponds to the gas pipeline equipment object sub-platform, and is used to obtain related data of pipeline equipment. In some embodiments, the sensing network platform can be configured as a communication network and a gateway.

[0034] In some embodiments, the smart gas sensing network platform 140 can interact with the smart gas object platform 150 downward. For example, receiving gas equipment related data uploaded by the smart gas object platform, and issuing a gas equipment related data obtaining instruction to the smart gas object platform. The smart gas sensing network platform 140 can also interact with the smart gas management platform 130 upward. For example, receiving a gas equipment related data obtaining instruction issued by the smart gas data center, and uploading gas equipment related data to the smart gas data center.

[0035] In some embodiments, the gas operation Internet of Things system further comprises a smart gas object platform 150. The smart gas object platform 150 comprises a gas indoor equipment object sub-platform and a gas pipeline equipment object sub-platform. Among them, the gas indoor equipment object sub-platform corresponds to the gas indoor equipment sensing network sub-platform, and the indoor equipment related data is uploaded to the smart gas data center through the gas indoor equipment sensing network sub-platform. The gas pipeline equipment object sub-platform corresponds to the gas pipeline equipment sensing network sub-platform, and the pipeline equipment related data is uploaded to the smart gas data center through the gas pipeline equipment sensing network sub-platform.

[0036] In some embodiments, the smart gas object platform 150 can interact with the smart gas sensing network platform 140 upward. For example, receiving a gas equipment related data obtaining instruction issued by the smart gas sensing network platform, and uploading gas equipment related data to the smart gas sensing network platform.

[0037] It should be noted that the above description of the gas operation Internet of Things system based on the smart gas call center and its internal modules is for convenience of description, and cannot limit the scope of the present application to the embodiments. It can be understood that, for those skilled in the art, after understanding the principle of the system, the modules can be combined arbitrarily, or connected with other modules to form a subsystem without departing from the principle. In some embodiments, Figure 1The disclosed intelligent gas user platform 110, intelligent gas service platform 120, intelligent gas management platform 130, intelligent gas sensing network platform 140, and intelligent gas object platform 150 can be different modules in a system, or one module can implement the functions of two or more of the above modules. For example, the modules can share a storage module, and each module can have its own storage module. Variations such as these are within the scope of the present specification.

[0038] Figure 2 is an exemplary flowchart of a gas operation method based on an intelligent gas call center according to some embodiments of the present specification. In some embodiments, the flow 200 can be performed by the intelligent gas management platform.

[0039] As shown in Figure 2 , the flow 200 includes the following steps.

[0040] Step 210, obtaining gas user call data of a call center.

[0041] The call center can refer to a customer service center for handling telephone inquiries from customers and recording and storing all call information. For example, a customer can call a gas call center for service, the gas call center handles the call, and stores the call information.

[0042] In some embodiments, a gas user can refer to a consumer who uses gas, including industrial gas users, commercial gas users, ordinary gas users, etc.

[0043] In some embodiments, call data can refer to call information of a gas user, and the call data can be classified into various types, including complaints, maintenance, consultations, gas ordering, etc. For example, consulting gas modification related issues, complaining about impure gas, etc.

[0044] In some embodiments, the intelligent gas management platform obtains the gas user call data of the call center through the intelligent gas data center.

[0045] Step 220, based on the gas user call data, respectively predicting demand information of different types of users.

[0046] In some embodiments, different types of users can refer to different types of gas users. According to different gas demands, gas users can be classified into various types, such as industrial gas users, commercial gas users, ordinary gas users, etc.

[0047] The demand information can reflect gas product demand and gas service demand. The gas product demand can refer to product demand related to gas or gas equipment, such as gas subscription demand, gas water heater purchase demand, etc. The gas service demand can refer to service demand related to gas or gas equipment, such as gas modification service demand, gas pipeline replacement demand, etc.

[0048] In some embodiments, the demand information of different types of users is different. For example, ordinary gas users have gas subscription demand; industrial gas users have high-calorific-value gas demand; commercial gas users have gas pipeline replacement demand, etc.

[0049] In some embodiments, the demand information can be predicted based on historical data by associating users of the same type, users with similar characteristics, and users in the same region. The historical data includes historical demand information. The smart gas management platform can make targeted predictions of demand information according to the historical demand information of each type of user. For example, in the historical demand information of ordinary gas users in a certain residential area, the proportion of gas subscription demand is large, so it can be predicted that the demand information of gas users in this area has a large probability of gas subscription demand; in the historical demand information of a certain chemical industrial gas user, the proportion of high-calorific-value gas demand is large, so it can be predicted that the demand information of the chemical enterprise has a large probability of high-calorific-value gas demand; in the historical demand information of a certain snack shop gas user in a commercial street, the proportion of gas pipeline replacement demand is large, so it can be predicted that the demand information of gas users in this commercial street has a large probability of gas pipeline replacement demand.

[0050] In some embodiments, the demand matching degree of users for different demands can also be determined based on different types of gas user call data, so as to predict the demand information of different types of users. For more information about the types of gas user call data and demand matching degree, please refer to Figure 3 and the related description.

[0051] Step 230, based on the demand information of different types of users, determine the gas operation push features and push.

[0052] The gas operation push feature can refer to the feature of gas operation promotion and publicity, which can include push type feature and push content feature. The push type feature can refer to the push form, such as telephone push, on-site push, email push, etc. The push content feature can refer to the content of the pushed product or service, such as push of gas stove, push of pipeline repair service, push of high-calorific-value gas, etc.

[0053] In some embodiments, the potential demand of different gas users can be determined according to the type, gas usage habit, region, similarity of indoor equipment, and the like of the gas users based on the historical demand information of different types of gas users, so as to determine the gas operation push features for different users.

[0054] In some embodiments, the gas user association graph can also be constructed based on the demand information of different types of users, so as to determine the gas operation push features. More details about the gas user association graph can be found in Figure 5A 、 Figure 5B and the related descriptions thereof.

[0055] In some embodiments of the present specification, the demand information of different types of users is predicted respectively by the gas user call data of the call center, and then the gas operation push features are determined for push, which is beneficial to accurately predict the user demand and provide services in time.

[0056] It should be noted that the above description of the process 200 is only for example and illustration, and does not limit the scope of the present specification. Various modifications and changes can be made to the process 200 by those skilled in the art under the guidance of the present specification. However, these modifications and changes are still within the scope of the present specification.

[0057] Figure 3 is an exemplary schematic diagram for determining the predicted demand information according to some embodiments of the present specification.

[0058] As shown in Figure 3 , the matching model 300 can include a feature extraction layer 320 and a determination layer 350.

[0059] In some embodiments, predicting the demand information of different types of users based on the gas user call data includes: classifying the gas user call data; determining the demand matching degree of users for different demands based on different types of gas user call data; and predicting the demand information of different types of users based on the demand matching degree.

[0060] In some embodiments, the smart gas management platform can classify the gas user call data according to the purpose of the gas users calling the gas call center for service. For specific details about the classification of the gas user call data, please refer to the related description of Figure 2 .

[0061] In some embodiments, based on different types of gas user call data, the demand matching degree of users for different demands can be determined by a matching model, that is, the demand matching degree can be determined based on the processing of the gas user call data corresponding to different types of users by the matching model.

[0062] In some embodiments, the demand matching degree can refer to a degree of matching of the demand information of different presets by the gas user. The preset demand information can include various demand information, such as gas stove purchase demand, gas water heater purchase demand, gas modification service demand, gas pipeline replacement demand, high-calorific-value gas demand, gas pressure regulator or filter purchase demand, and the like. In some embodiments, the demand matching degree can be expressed in percentage, and the larger the value, the higher the demand matching degree.

[0063] In some embodiments, the matching model 300 includes a feature extraction layer 320 and a determination layer 350. In some embodiments, the matching model 300 can refer to a machine learning model for determining the demand matching degree 360. In some embodiments, the matching model 300 can be a trained machine learning model. For example, the matching model 300 can include any one or combination of a neural network model, a convolutional neural network, or other custom model structure.

[0064] In some embodiments, the feature extraction layer 320 can be used to obtain gas call features 330. The feature extraction layer 320 can be a machine learning model, for example, the feature extraction layer 320 can include a convolutional neural network, a neural network, or the like. In some embodiments, the input of the feature extraction layer 320 can include gas user call data 310, and the output can include gas call features 330. The gas call features 330 can include user type, call type, call content, call time frequency, and the like. For example, the call content can include user feedback that the gas calorific value fluctuates greatly, user feedback that the pipeline leaks, and the like; the call type can include complaint, product consultation, fault maintenance, and the like; the call time frequency can refer to the time frequency of the gas user making a call, for example, calling once every five minutes. For specific content of the gas user call data 310, please refer to the related description of Figure 2 .

[0065] In some embodiments, the determination layer 350 can be used to determine the demand matching degree 360 of the user for different preset demand information. The determination layer 350 can be implemented by a machine learning model, such as a convolutional neural network, a neural network, or the like.

[0066] In some embodiments, the input of the determination layer 350 can include gas call features 330 and preset demand information 340. The output of the determination layer 350 can include the demand matching degree 360 of the user for different preset demand information. In some embodiments, when multiple preset demand information 340 is input at the same time, the demand matching degree 360 can be arranged in the corresponding order of the preset demand information 340 to form a sequence, where an element of the sequence corresponds to a preset demand information. For specific content of the demand information, please refer to the related description of Figure 2 .

[0067] In some embodiments, the matching model can be determined by jointly training the feature extraction layer 320 and the determination layer 350. In some embodiments, the first training sample of the matching model can include a large amount of historical gas user call data and historical preset demand information. The label of the first training sample can include the historical demand matching degree of the gas user for the demand information of different presets. In some embodiments, the first training sample and the first training label can be obtained based on the historical data of the smart gas call center.

[0068] In some embodiments, the gas call features 330 output by the feature extraction layer 320 can be used as input of the determination layer 350. The process of joint training can include: using the historical gas user call data in the first training sample as input of the feature extraction layer 320; using the gas call features 330 output by the feature extraction layer 320 as input of the determination layer 350 to obtain the output of the determination layer 350; constructing a loss function with the demand matching degree 360 output by the determination layer 350 and the first training label; iteratively updating based on the loss function until the loss function is less than a threshold, converges, or the training period reaches a threshold, etc. to obtain the trained feature extraction layer and determination layer.

[0069] In some embodiments, based on different types of gas user call data, the demand matching degree of the user for different demands can be determined based on historical experience. The demand matching degree can be manually evaluated according to experience.

[0070] In some embodiments of the present specification, the demand matching degree determined by the matching model can more reasonably and accurately determine the demand matching degree of the user for different demands, which is beneficial to improve the efficiency of gas operation.

[0071] In some embodiments, based on the demand matching degree 360, the demand information 380 of different types of users can include: determining at least one demand corresponding to the demand matching degree of the gas user satisfying the preset condition; in response to the existence of corresponding customer follow-up data 370 for at least one demand, combining the customer follow-up data 370 to predict the demand information 380 of different types of users.

[0072] In some embodiments, the demand matching degree satisfying the preset condition can refer to the demand matching degree ranked first after sorting the demands in descending order according to the size of the demand matching degree, or the demand matching degree ranked in the top n and exceeding a preset threshold. The preset threshold can be artificially set, for example, 80%.

[0073] In some embodiments, the user corresponding to the demand with a demand matching degree exceeding 80% and ranked in the top five can be selected to determine whether there is customer follow-up data.

[0074] The customer follow-up data can refer to evaluation feedback of the customer after accepting the service or using the product, such as customer satisfaction, whether the problem is solved (partially solved or completely solved), customer feedback, and the like.

[0075] In some embodiments, in a certain residential building gas user, the top five demands corresponding to the demand matching degree ranking are gas subscription demand, gas stove purchase demand, gas water heater purchase demand, filter replacement demand, and gas pipeline replacement demand, and the demand matching degrees are 92%, 88%, 85%, 75%, and 60%, respectively. For example, the demand corresponding to the demand matching degree exceeding the preset threshold of 80% includes the gas stove purchase demand. According to the follow-up data, it is found that multiple customers feedback that the A type gas stove purchased is very unsatisfactory, and the customers can choose to return the goods, return the goods, or purchase the gas stove from other channels, thereby predicting that the purchase demand of the A type gas stove will be less in the near future.

[0076] In some embodiments, in response to the demand corresponding to the demand matching degree satisfying the preset condition being the demand corresponding to the demand matching degree ranked first, the demand information is directly predicted without combining the customer follow-up data corresponding to the demand.

[0077] In some embodiments of the present specification, the demand corresponding to the demand matching degree satisfying the preset condition of the gas user is determined, and the customer follow-up data is combined, which can more accurately predict the demand information of different types of users, and is beneficial to improve the user satisfaction.

[0078] In some embodiments of the present specification, the demand matching degree is determined through the matching model, and the demand information is predicted based on the demand matching degree, which can accurately predict the user demand and perform targeted pushing.

[0079] Figure 4 is an exemplary schematic diagram of determining a stock preparation scheme according to some embodiments of the present specification.

[0080] In some embodiments, the intelligent gas management platform can also determine the stock preparation scheme through the call center, including: predicting the demand quantity information of the gas product and the gas service based on the demand information of different types of users; and determining the stock preparation scheme through the call center based on the predicted demand quantity information.

[0081] The stock preparation scheme refers to a related scheme for a gas company to prepare gas products in advance, improve the gas service level of gas service personnel, and the like based on the predicted user demand quantity information. For example, the stock preparation scheme can be to reserve a certain product type and a certain product quantity (for example, 10 gas stoves of a certain model, 20 gas pressure regulating valves, and the like). For another example, the stock preparation scheme can also be to train the gas service personnel for a certain gas service (for example, to train the gas service personnel for gas entry pipeline maintenance, and the like).

[0082] As shown in Figure 4 , the intelligent gas management platform can determine the replenishment plan based on the following steps.

[0083] Step 410, based on the demand information of different types of users, predict the demand quantity information about gas products and gas services.

[0084] Gas products refer to products related to gas use. For example, gas products can include gas stoves, gas water heaters, gas pressure regulating valves, gas filters, etc.

[0085] Gas services refer to services provided by gas companies to users related to gas. For example, gas services can include gas product replacement services, gas fault repair services, etc.

[0086] Demand quantity information refers to the demand quantity of gas users for the types and quantities of gas products (e.g., 1 gas stove, 1 gas pressure regulating valve, etc.) and the demand quantity of gas service types and times (e.g., 2 times of gas valve replacement service, 1 time of gas stove repair service, etc.). In some embodiments, gas service demand may also be related to gas product demand. For example, when providing gas stove repair service to users, new gas stoves may need to be replaced, etc.

[0087] In some embodiments, the intelligent gas management platform can predict the corresponding demand quantity information based on the demand information of different types of gas users. For example, when the demand information of a certain residential gas user is to repair a gas stove, the intelligent gas management platform can determine that the demand quantity information of the gas user is to repair the gas stove service 1 time. For example, when the demand information of a certain industrial gas user is to replace a gas valve, the intelligent gas management platform can determine that the demand quantity information of the gas user is to replace the gas valve service 1 time and 10 gas valve products, etc.

[0088] In some embodiments, the prediction of demand quantity information may also be related to demand matching degree. For example, when the demand matching degree of the gas stove is greater than a certain threshold (e.g., 90%, etc.), it can be predicted that the user has the demand to replace the gas stove, etc. For detailed description of demand matching degree, please refer to Figure 3 .

[0089] In some embodiments, demand quantity information is related to demand matching degree, which can improve the accuracy of demand quantity information prediction and better meet the demand of gas users for gas products and gas services.

[0090] In some embodiments, the demand information can also be predicted based on processing of the demand information of different types of users and the gas operation push features (including push type features, push content features) by a demand prediction model. The demand prediction model can be a machine learning model. For example, the demand prediction model can be a neural network model, a deep neural network model, a recurrent neural network model, or any combination thereof. For more information about the gas operation push features, please refer to Figure 2 .

[0091] As shown in FIG. 4, the demand prediction model 400 can include an embedding layer 412 and a demand prediction layer 414. Figure 4

[0092] The embedding layer 412 can be used to process the demand information of different types of users and the gas operation push features to obtain an embedding feature vector. In some embodiments, the input of the embedding layer 412 can include the demand information 411-1, the push type features 411-2, and the push content features 411-3, and the output can include the embedding feature vector 413. For more information about the demand information, the push type features, and the push content features, please refer to Figure 2 .

[0093] The embedding feature vector 413 is a vectorized representation of the user demand, the push type features, and the push content features. For example, the embedding feature vector 413 can be represented as (x, y, z); where x represents the user demand, y represents the push type features, and z represents the push content features.

[0094] The demand prediction layer 414 can be used to determine the demand information of the gas user. In some embodiments, the input of the demand prediction layer 414 can include the embedding feature vector 413, and the output can include the demand information 415. The demand information refers to the related information of the gas user's demand for the types and quantities of gas products, the types and times of gas services, etc. For example, the demand information can be that the gas user needs 1 gas water heater and 1 gas valve replacement service. For more information about the demand information, please refer to the relevant sections above.

[0095] In some embodiments, the demand prediction model can be determined by joint training of the embedding layer and the demand prediction layer.

[0096] ​In some embodiments, the second training sample of the demand quantity prediction model can include a plurality of sets of historical demand information, historical push type features, historical push content features, and / or the like corresponding to a plurality of sample gas users. Each set of historical demand information, historical push type features, and historical push content features corresponds to a sample gas user. The label of the second training sample can include historical demand quantity information of the sample gas user. In some embodiments, the second training sample and its label can be obtained based on historical data of the smart gas call center. For details of the training process of the demand quantity prediction model, please refer to Figure 3 the training process of the matching model in

[0097] In some embodiments, the smart gas management platform processes demand information of different types of users and gas operation push features through the demand quantity prediction model to determine demand quantity information of gas users, which can obtain more accurate results than manually predicting demand quantity information of gas users, improve efficiency, and save human resources.

[0098] At step 420, a replenishment plan is determined through the call center based on the predicted demand quantity information.

[0099] In some embodiments, the smart gas management platform determines a replenishment plan through the call center based on demand information of different types of gas users, which can predict the demand of gas users for types and quantities of gas products in advance, provide corresponding products to gas users, better meet user demand, improve user satisfaction, and improve gas operation efficiency.

[0100] Figure 5A is an exemplary schematic diagram of determining gas operation push features according to some embodiments of the present specification.

[0101] In some embodiments, the smart gas management platform can also determine gas operation push features based on demand information of different types of users, including: constructing a gas user association graph based on demand information of different types of users; and determining gas operation push features based on the gas user association graph. In some embodiments, the gas operation push features can be determined based on the following steps.

[0102] At step 510, a gas user association graph is constructed based on demand information of different types of users.

[0103] The gas user association graph can be used to reflect different gas users and the association relationship between different gas users, such as whether different gas users are users of the same community, etc. In some embodiments, the gas user association graph can include nodes and edges.

[0104] In Figure 5BIn the illustrated exemplary schematic diagram of dividing one or more push community sub-regions, a first gas user association graph 511 and a second gas user association graph 512 are included. The second gas user association graph 512 is generated after the first gas user association graph 511 is divided into push community sub-regions. Detailed descriptions of the push community sub-regions and the division thereof can be found in the relevant sections below.

[0105] In some embodiments, the nodes of the gas user association graph can reflect different gas users. For example, as shown in the first gas user association graph 511, the node a represents a gas user a, the node b represents a gas user b, and so on. Figure 5B

[0106] In some embodiments, the node features can include demand information of the gas users. Detailed descriptions of the demand information can be found in the relevant sections below. Figure 2

[0107] In some embodiments, the node features further include demand matching degrees of the gas users to gas products and gas services. Detailed descriptions of the demand matching degrees can be found in the relevant sections below. Figure 3

[0108] In some embodiments, the node features including the demand matching degrees can make the subsequently determined gas operation push features more in line with the demands of the users, thereby improving user satisfaction.

[0109] In some embodiments, the node features can further include customer revisit data. Detailed descriptions of the customer revisit data can be found in the relevant sections below. Figure 3

[0110] In some embodiments, the node features including the customer revisit data can enable adjustment of the push information of the users based on the feedback information of the users (for example, when a user feedbacks that a certain gas problem has been solved, the smart gas management platform can stop pushing the content related to the gas problem and determine whether to push other gas content based on the revisit data of the user), so as to obtain better push effects.

[0111] In some embodiments, the edges of the gas user association graph can reflect the association relationships between different gas users. For example, in the first gas user association graph 511, the lines connecting different nodes (such as the node a, the node b, and so on) are the edges of the gas user association graph. In some embodiments, the nodes having certain association relationships (for example, in the same community, in the same unit building, and so on) can be connected by edges. For example, the node a and the node b have an association relationship in the same community, and thus the node a and the node b can be connected by an edge; the node c and the node e do not have an association relationship, and thus the node c and the node e are not connected by an edge.​​​​

[0112] In some embodiments, the edge features of the gas user association graph can include association attributes of different nodes. The association attribute refers to the correlation degree of different nodes of the gas user association graph. For example, the association attribute can include the same community, the same unit building, the same gas pipeline, etc.

[0113] In some embodiments, the edges of the gas user association graph can also have edge weights. The edge weights can be determined based on the edge features. In some embodiments, the smart gas management platform can preset different edge features with different importance degrees in advance based on experience. For example, the importance degrees of different edge features can be represented by numerical values of 1-3, and the numerical value corresponding to the more important edge feature is larger (for example, the importance degree of the same unit building is 3, the importance degree of the same community but different unit buildings is 2, and the importance degree of the same gas pipeline is 1, etc.). For example, node a and node b are located in the same community, the same unit building, and the same gas pipeline, and the weight of the edge between node a and node b is 3+2+1=6. For another example, node b and node c belong to the same gas pipeline, but do not belong to the same community, and the weight of the edge between node b and node c is 1.

[0114] In some embodiments, the node features can also include the gas community features.

[0115] The gas community refers to a push community sub-region to which the gas user belongs. For example, as shown in FIG. 5, the gas user corresponding to node a belongs to the first push community sub-region 512-1. Figure 5B

[0116] The push community sub-region refers to a region composed of gas users with similar demand information (for example, all needing gas stove repair services, etc.) and having certain association relationships (for example, the same community, the same gas pipeline, etc.). For example, as shown in FIG. 5, the gas user association graph can be divided into one or more push community sub-regions. For example, the gas user association graph can be divided into the first push community sub-region 512-1, the second push community sub-region 512-2, etc. More details about determining the push community sub-region can be found in the relevant part later. Figure 5B

[0117] The gas community feature reflects the correlation degree (for example, belonging to the same community, having similar demand information, etc.) between multiple gas users. For example, as shown in FIG. 5, nodes a, b, and e belong to the first push community sub-region 512-1, and the gas community features of the gas users corresponding to nodes a, b, and e are the same, which can be represented as m1. Figure 5B

[0118] ​​​In some embodiments, the gas users belonging to the same push community sub-area have the same gas community features. Wherein, the same push community sub-area refers to the same push community sub-area to which the gas users belong. For example, the gas users a, b, and e all belong to the first push community sub-area 512-1.

[0119] In some embodiments, one or more push community sub-areas of the gas user association graph can be determined by a clustering method. In some embodiments, the smart gas management platform can randomly select multiple nodes in the gas user association graph as clustering centers, and cluster the gas user nodes in the gas user association graph by a clustering feature vector.

[0120] In some embodiments, the clustering feature vector can be constructed based on node features. Wherein, the elements in the clustering feature vector can correspond to the demand information of the gas user, the edge feature (i.e., converting the edge feature of the gas user association graph into the node feature, for example, the edge feature of the gas user a and the gas user b is that they belong to the same cell, which can be represented as: the gas user a is located in the p cell and the gas user b is located in the p cell after conversion into the node feature).

[0121] In some embodiments, the smart gas management platform can process the clustering feature vectors of each node in the gas user association graph by a clustering algorithm to obtain one or more clustering results composed of multiple nodes; and take the one or more clustering results as one or more push community sub-areas of the gas user association graph. Wherein, the clustering algorithm can include a variety of, for example, the clustering algorithm can include K-Means (K-means) clustering, density-based clustering method (DBSCAN), etc.

[0122] In some embodiments, the smart gas management platform can determine one or more push community sub-areas based on the gas user association graph, including: determining one or more push community sub-areas based on the gas user association graph by a preset algorithm. Wherein, the preset algorithm includes performing multiple rounds of iterations, and each round of iteration in the multiple rounds of iterations includes: determining the push community sub-area to which one or more gas user nodes in the gas user association graph belong.

[0123] In some embodiments, at least one round of iteration in the multiple rounds of iterations includes: calculating the increment of community joint degree; and determining the push community sub-area to which one or more gas user nodes in the gas user association graph belong based on the increment of community joint degree. Wherein, the community joint degree is related to the graph complexity of the gas user association graph and the connection between the gas user nodes.

[0124] Community cohesiveness is a measure of the quality of the division of the gas user association graph. In some embodiments, the greater the value of the community cohesiveness of a node, the greater the degree of association of the node with other nodes in the same push community sub-region to which the node belongs, and the more accurate the gas community characteristics of the gas user corresponding node to which the node belongs.

[0125] The graph complexity of the gas user association graph refers to the complexity of the gas user association graph. In some embodiments, the graph complexity of the gas user association graph can be determined based on the number of edges in the graph. For example, the more edges in the gas user association graph, the greater the graph complexity. The connection between gas user nodes refers to the connection between different nodes in the gas user association graph. For example, the connection between gas user nodes can be whether two nodes are connected, whether two nodes belong to the same push community sub-region, etc.

[0126] In some embodiments, the community cohesiveness Q can be determined based on the following formula (1):

[0127]

[0128] wherein m represents the number of edges in the gas user association graph; i, j represent node identifiers respectively; a ij represents an element of the adjacency matrix in the gas user association graph, which is used to represent the adjacent relationship between nodes, wherein when node i is adjacent to node j (i.e., there is an edge connecting node i and node j), a ij = 1, otherwise a ij = 0; k i , k j represent the degrees of node i and node j respectively, wherein the degree of a node refers to the number of edges associated with the node (for example, the degree of node a in the first gas user association graph 511 is 1, and the degree of node b is 4); γ i , γ j represent the push community sub-regions to which node i and node j belong respectively; the function δ(γ i , γ j ) represents whether the push community sub-regions to which node i and node j belong are the same, i.e., when γ i = γ j , δ(γ i , γ j ) = 1, otherwise δ(γ i , γ j ) = 0.

[0129] In some embodiments, the smart gas management platform can determine the push community sub-region to which one or more gas user nodes in the gas user association graph belong based on the following steps:

[0130] Step 1, all nodes in the gas user association graph are regarded as an independent push community sub-area, and the initial number of push community sub-areas is equal to the number of nodes.

[0131] Step 2, for any node (for example, node i) in the gas user association graph, it is sequentially assigned to the push community sub-area corresponding to the adjacent node (that is, node i and the adjacent node are divided into the same push community sub-area), and the community joint degree Q1 before assignment and the community joint degree Q2 after assignment are calculated to obtain the community joint degree increment AQ = Q1-Q2; node i is assigned to the adjacent node corresponding to AQ>0 and the maximum AQ, if the AQ corresponding to multiple adjacent nodes of node i is less than or equal to 0, node i remains unchanged (that is, node i is not assigned to other push community sub-areas).

[0132] Step 3, repeat step 2 until the belonging push community sub-area of all nodes no longer changes (that is, all nodes cannot be assigned to other push community sub-areas).

[0133] Step 4, the result obtained in step 3 is graph compressed, that is, nodes belonging to the same push community sub-area are compressed into a new node (that is, a push community sub-area, such as Figure 5B the first push community sub-area 512-1, the second push community sub-area 512-2, etc.), the weight of the edge between different new nodes is equal to the sum of the edge weights between the new nodes, and each new node increases an edge to itself (such as Figure 5B the edge 513-1 and the edge 513-2 in ), and the weight of the edge is the sum of the weights of the edges between the original nodes in the new node; and steps 1 to 3 are re-executed.

[0134] For example, as shown in Figure 5B , nodes a, b, and e belonging to the same push community sub-area are compressed into a new node (that is, the first push community sub-area 512-1), and nodes c, d, and f are compressed into a new node (that is, the second push community sub-area 512-2); wherein the weight of the edge connecting the first push community sub-area 512-1 to itself (that is, the edge 513-1) is the sum of the weight of the edge connecting nodes a and b (assuming the weight is 5) and the weight of the edge connecting nodes b and e (assuming the weight is 3), that is, the weight of the edge 513-1 is 8, and the weight of the edge 513-2 can be obtained in the same way; the weight of the edge connecting the first push community sub-area 512-1 and the second push community sub-area 512-2 is the sum of the weight of the edge connecting nodes b and c (assuming the weight is 1) and the weight of the edge connecting nodes b and f (assuming the weight is 1), that is, the weight of the edge between the first push community sub-area 512-1 and the second push community sub-area 512-2 is 2.

[0135] Step 5, repeat step 4 (i.e. multiple times of graph compression), until the community joint degree of the entire gas user association graph no longer changes, and the compressed new node is taken as the one or more push community sub-regions to which the nodes in the gas user association graph belong.

[0136] In some embodiments, the community joint degree can also be related to the connection weight. The connection weight refers to the weight of the demand for gas products and gas services determined based on the similarity of the demand matching degrees of different nodes in the gas user association graph. The connection weight between nodes with closer demand matching degrees is larger, and the maximum is 1. The calculation formula (2) of the community joint degree related to the connection weight is as follows:

[0137]

[0138] wherein w ij represents the connection weight between node i and node j. The closer the demand matching degree between node i and node j, the larger the value of w ij . The meanings of the remaining letters can be referred to the corresponding description in formula (1).

[0139] In some embodiments, determining the push community sub-regions based on the community joint degree can make the nodes in the push community sub-regions have a higher association degree, and the obtained gas community features of the nodes are more in line with the actual situation of the users; wherein the community joint degree related to the connection weight can further improve the accuracy of the divided push community sub-regions.

[0140] In some embodiments, the gas community features of the nodes in the gas user association graph can be determined by the following steps: determining one or more push community sub-regions based on the gas user association graph; determining the gas community features of the gas user nodes in the gas user association graph based on the one or more push community sub-regions.

[0141] For example, the smart gas management platform can determine the push community sub-regions of the gas user association graph based on the foregoing manner, and determine the gas community features of the gas user nodes in the gas user association graph based on the push community sub-regions. For example, gas user a, gas user b, and gas user e belong to the first push community sub-region 512-1, and their gas community features can be represented as m1; gas user c, gas user d, and gas user f belong to the second push community sub-region 512-2, and their gas community features can be represented as m2, and so on.

[0142] In some embodiments, the node features of the gas user nodes in the gas user association graph include the gas community features, which can make users with similar gas demands obtain similar gas operation push features, facilitate the push of corresponding gas products and services to the gas users, and obtain better push effect.

[0143] At step 520, the gas operation push feature is determined based on the gas user association graph.

[0144] In some embodiments, the gas operation push feature can be determined based on processing the gas user association graph by a graph neural network (GNN) model. For example, the smart gas management platform can input the gas user association graph into the GNN, and output the gas operation push feature corresponding to each node. In some embodiments, the graph neural network model can be trained based on a large number of third training samples with labels. Specifically, the third training samples with labels are input into the graph neural network model, and the parameters of the graph neural network model are updated by training until the loss function is less than a threshold, convergence, or the training period reaches a threshold, etc. to obtain the trained graph neural network model. In some embodiments, the third training samples can be a large number of historical gas user association graphs, and the labels of the third training samples can be the historical gas operation push features of each node in the historical gas user association graphs. In some embodiments, the third training samples and labels can be obtained based on historical data of the call center and the smart gas data center.

[0145] In some embodiments, based on the demand information of different types of users, the gas user association graph is constructed, and further based on the gas user association graph, the gas operation push feature is determined, which can make the gas products and gas services pushed to the users more in line with the needs of the users.

[0146] The foregoing detailed description has set forth various embodiments of the application via the use of specific terminology. As such, the description herein is not intended to limit the application, but rather to describe various embodiments of the application. It is to be understood that the foregoing description is not of a total scope of the application. Various modifications, improvements, and alterations of the described embodiments can be made by those skilled in the art without departing from the spirit and scope of the described embodiments. Such modifications, improvements, and alterations are intended to be part of this disclosure.

[0147] Also, the use of "an" or "one" to refer to an item of an embodiment should not be construed to supersede a meaning of "at least one" or "one or more." Further, some of the features of the described embodiments can be used to advantage without the corresponding use of other features. As such, the foregoing description shall not be construed to be limited to the described embodiments, but rather, is intended to be construed broadly, to include all such modifications, improvements, and alterations that fall within the spirit and scope of the described embodiments.

[0148] Furthermore, the order of the processing elements and sequences described in this specification are not intended to be construed as a limitation, unless specifically stated, but are included to provide a complete description of one or more embodiments of the present specification. Regardless of the particular sequence of processing elements and sequences, however, the description herein of a process should be understood to include any and all combinations of one or more elements of a process independently selected from each sequence. For example, although the system components described above can be implemented by hardware devices, they can also be implemented by software solutions, such as installing the described system on an existing server or mobile device.

[0149] Similarly, it is to be noticed that the term "comprising", used in the description, is not intended to exclude other elements or steps. It is to be understood that the description and the examples are intended to be illustrative, but not limiting, of the scope of the present specification. Thus, the scope of the present specification should be given by the appended claims, along with their full scope of equivalents, and not by an restricting interpretation of the description or the examples.

[0150] Some embodiments use numerical designations to describe components, quantities of attributes. It is to be understood that such numerical designations used in the description of embodiments are, in some examples, modified by the adjectives "about", "approximately", or "generally". Unless otherwise stated, "about", "approximately", or "generally" indicates that the stated numerical value is allowed ±20% variation. Accordingly, in some embodiments, numerical parameters in the description and claims are approximations, and are intended to be construed in a liberal manner, based on the knowledge that individual embodiments can have different requirements. In some embodiments, numerical parameters are provided as approximations, and are not to be construed in a limiting sense. Although numerical ranges and parameters setting forth the broadest scope of the embodiments are approximations, in some embodiments, the numerical values set forth in the specific examples are reported as precisely as practicable. The numerical values set forth in the specific examples are provided to be as precise as practicable.

[0151] Each patent, patent application, patent publication, and other material, articles, books, instructions, documents, that has been identified herein, is hereby incorporated herein by reference in its entirety for all purposes. Except in instances where the instant specification contradicts or contradicts away part of the incorporated material, articles, books, instructions, documents, the incorporated material, articles, books, instructions, documents are hereby incorporated by reference for all purposes. It is specifically intended that the description, definitions, and / or terminology used herein be interpreted as consistent with the description, definitions, and / or terminology used throughout the incorporated material, articles, books, instructions, documents.

[0152] Finally, it should be understood that the embodiments described herein are only given by way of example and that other modifications can occur to persons skilled in the art. Therefore, the scope of the present description is not intended to be limited to the embodiments described herein but is only limited by the claims that follow.

Claims

1. A gas operation method based on a smart gas call center, characterized in that, The method is implemented by a smart gas management platform based on a smart gas call center-based gas operation IoT system, and the method includes: Obtain gas user call data from the call center; Based on the gas user call data, the demand information of different types of users is predicted; the demand information includes at least gas product demand and gas service demand. Based on the demand information of the different types of users, gas operation push features are determined and pushes are made; the gas operation push features include push type features and push content features. The step of determining gas operation push notification characteristics and pushing them based on the demand information of the different types of users includes: Based on the demand information of the different types of users, a gas user association graph is constructed; the gas user association graph includes nodes and edges, the nodes include gas user nodes, and the edges include the association attributes between the gas users. Based on the gas user association graph, the gas operation push characteristics are determined; The node characteristics of the node include the characteristics of the gas community to which it belongs, and the characteristics of the gas community to which it belongs include: Based on the gas user association graph, one or more push community sub-regions are determined: Based on the one or more push community sub-regions, determine the gas community characteristics to which the node belongs; The step of determining one or more push community sub-regions based on the gas user association graph includes: The one or more push community sub-regions are determined by a preset algorithm, wherein the preset algorithm includes performing multiple iterations, and each iteration in the multiple iterations includes: Determine the one or more push community sub-regions to which one or more gas user nodes in the gas user association graph belong.

2. The method according to claim 1, characterized in that, The gas operation IoT system based on the smart gas call center includes a smart gas user platform, a smart gas service platform, a smart gas management platform, a smart gas sensor network platform, and a smart gas object platform. The smart gas user platform is used to: send gas user call data to the smart gas service platform, and receive customer service feedback information uploaded by the smart gas service platform; send gas operation management information query instructions to the smart gas service platform, and receive gas maintenance management information uploaded by the smart gas service platform; The smart gas service platform is used to: send the gas operation management information query instruction to the smart gas management platform, receive the gas operation management information uploaded by the smart gas management platform; receive the gas operation management information query instruction sent by the smart gas user platform, and upload the gas operation management information to the smart gas user platform; The intelligent gas management platform is used to: send instructions to the intelligent gas sensor network platform to obtain relevant data of gas equipment, and receive relevant data of gas equipment uploaded by the intelligent gas sensor network platform; Receive the gas operation management information query instruction issued by the smart gas service platform, and upload the gas operation management information to the smart gas service platform; The intelligent gas sensor network platform is used to: receive gas equipment-related data uploaded by the intelligent gas object platform, and send the instruction to obtain gas equipment-related data to the intelligent gas object platform; receive the instruction to obtain gas equipment-related data sent by the intelligent gas management platform, and upload the gas equipment-related data to the intelligent gas management platform; The smart gas object platform is used to: receive the instruction to acquire gas equipment-related data issued by the smart gas sensor network platform, and upload the gas equipment-related data to the smart gas sensor network platform.

3. The method according to claim 1, characterized in that, The prediction of demand information for different types of users based on the gas user call data includes: The gas user call data is classified; Based on the gas user call data of the different types of users, determine the degree of matching of the gas user's needs for different requirements; Based on the required matching degree, the required information of the different types of users is predicted.

4. The method according to claim 1, characterized in that, The method further includes: Based on the demand information of the different types of users, predict the demand information for gas products and gas services; Based on the predicted demand information, the call center determines the inventory preparation plan.

5. The method according to claim 1, characterized in that, At least one round of the multiple iterations includes: Calculate the increment of community cohesion; based on the increment of community cohesion, determine the one or more push community sub-regions to which the one or more gas user nodes in the gas user association graph belong, wherein the community cohesion is related to the graph complexity of the gas user association graph and the connection status between the one or more gas user nodes.

6. A gas operation IoT system based on a smart gas call center, characterized in that, The smart gas management platform of the gas operation IoT system based on a smart gas call center is configured to perform the following operations: Obtain gas user call data from the call center; Based on the gas user call data, the demand information of different types of users is predicted; the demand information includes at least gas product demand and gas service demand. Based on the demand information of the different types of users, gas operation push features are determined and pushes are made; the gas operation push features include push type features and push content features. The step of determining gas operation push notification characteristics and pushing them based on the demand information of the different types of users includes: Based on the demand information of the different types of users, a gas user association graph is constructed; the gas user association graph includes nodes and edges, the nodes include gas user nodes, and the edges include the association attributes between the gas users. Based on the gas user association graph, the gas operation push characteristics are determined; The node characteristics of the node include the characteristics of the gas community to which it belongs, and the characteristics of the gas community to which it belongs include: Based on the gas user association graph, one or more push community sub-regions are determined: Based on the one or more push community sub-regions, determine the gas community characteristics to which the node belongs; The step of determining one or more push community sub-regions based on the gas user association graph includes: The one or more push community sub-regions are determined by a preset algorithm, wherein the preset algorithm includes performing multiple iterations, and each iteration in the multiple iterations includes: Determine the one or more push community sub-regions to which one or more gas user nodes in the gas user association graph belong.

7. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions, which, when executed by a processor, implement the gas operation method based on a smart gas call center as described in any one of claims 1 to 5.

Images