Building economic data processing method, device and system based on privacy calculation

Abstract

The embodiment of the invention provides a building economic data processing method, device and system based on privacy calculation. The first party holds the first building economic data, and the second party holds the second building economic data. And the first party performs privacy processing on the first building economic data to obtain first secret-state building economic data. And the second party performs privacy processing on the second building economic data to obtain second secret-state building economic data. And processing the first encrypted building economic data and the second encrypted building economic data between the first party and the second party through data interaction based on privacy calculation according to a preset building economic data processing logic to obtain encrypted building comprehensive economic data.

Description

Technical Field

[0001] One or more embodiments of the present specification relate to the field of computer technology, and in particular, to a method, device, and system for processing building economic data based on privacy computing. Background Art

[0002] Building economic analysis refers to the process of systematically studying and evaluating the role, function, benefits and impact of building properties such as commercial buildings, office buildings and complexes in the city in economic development. This concept emphasizes that buildings are not only providers of physical space, but also important carriers for promoting regional economic development, promoting industrial upgrading, increasing local finances and improving the image of the city. Building economic analysis requires joint data processing using building economic data held by multiple parties. Building economic data is private data related to enterprises in the building and cannot be sent to the outside in plain text.

[0003] Therefore, it is hoped that there will be an improved solution that can achieve building economic data processing between multiple parties without leaking the original building economic data. Summary of the invention

[0004] One or more embodiments of this specification describe a method, device and system for processing building economic data based on privacy computing, so as to realize the processing of building economic data between multiple parties without leaking the original building economic data. The specific technical solution is as follows.

[0005] In a first aspect, an embodiment provides a building economic data processing method based on privacy computing, which is performed by a first party holding first building economic data and a second party holding second building economic data, and includes:

[0006] The first party performs privacy processing on the first building economic data to obtain first confidential building economic data;

[0007] The second party performs privacy processing on the second building economic data to obtain second confidential building economic data;

[0008] The first party and the second party, through data interaction based on privacy computing, process the first confidential building economic data and the second confidential building economic data according to the preset building economic data processing logic to obtain confidential building comprehensive economic data.

[0009] In a second aspect, an embodiment provides a method for processing building economic data based on privacy computing, which is performed by a first party holding first building economic data, and the first party is in communication connection with a second party holding second building economic data; the method includes:

[0010] Performing privacy processing on the first building economic data to obtain first confidential building economic data;

[0011] Through data interaction based on privacy computing with the second party, the first confidential building economic data and the second confidential building economic data are processed in accordance with the preset building economic data processing logic to obtain confidential building comprehensive economic data; wherein, the second confidential building economic data is obtained after the second party performs privacy processing on the second building economic data.

[0012] In one implementation, the first party represents a building management party, the second party represents a relevant government department, the first building economic data includes several enterprises in the building, and the second building economic data includes economic data of multiple enterprises; the step of processing the first confidential building economic data and the second confidential building economic data includes:

[0013] Through data interaction based on privacy computing with the second party, the intersection of the several enterprises and the multiple enterprises is calculated to obtain several target enterprises, and according to the preset building economic data processing logic, the confidential building comprehensive economic data is determined based on the economic data of the several target enterprises.

[0014] In one embodiment, the first building economic data also includes the operation data of the several enterprises in the building; the step of determining the dense building comprehensive economic data based on the economic data of the several target enterprises includes:

[0015] Based on the operation data, selecting target enterprises that are operating normally from the plurality of target enterprises;

[0016] Based on the sum of the economic data of the normally operating target enterprises, the comprehensive economic data of the dense building is determined.

[0017] In one implementation, the step of performing privacy processing on the first building economic data includes:

[0018] Through data interaction based on privacy computing with several enterprises in the building, the operating data of each enterprise is privacy processed, so that the building management party can obtain the first confidential building economic data.

[0019] In one embodiment, the method further comprises:

[0020] The first prediction model is trained using the first building economic data to obtain first gradient data, and the encrypted first gradient data is sent to a third party, so that the third party aggregates the encrypted first gradient data and the encrypted second gradient data sent by the second party to obtain aggregated gradient data; the second gradient data is obtained by the second party using the second building economic data to train a second prediction model, and the aggregated gradient data is used to update the parameters of the first prediction model and the second prediction model.

[0021] In one implementation, the first building economic data includes a plurality of data; the first party performs privacy processing on the first building economic data, including:

[0022] The first building economic data is privacy processed based on a differential privacy algorithm.

[0023] In one embodiment, the first party and the second party are trusted computing nodes in a trusted computing platform; the first party performs privacy processing on the first building economic data, comprising:

[0024] The first building economic data is privacy-processed in its own trusted execution environment.

[0025] In one embodiment, the first party and the second party are nodes in a blockchain network, and the first party, performing privacy processing on the first building economic data, comprises:

[0026] Submitting a first transaction for invoking a first smart contract to the blockchain network, wherein the input parameters include the first building economic data;

[0027] Based on the first transaction, the operating logic of the first smart contract is executed to perform privacy processing on the first building economic data to obtain first confidential building economic data.

[0028] In one embodiment, the first party and the second party represent any one of the following entities:

[0029] Building managers, relevant government departments, energy suppliers, building lessors, and building owners;

[0030] The first building economic data and the second building economic data include one or more of the following data:

[0031] Basic information of several enterprises in the building, rental management data, operation data, energy consumption data, financial data, market performance data, and tax data.

[0032] In one embodiment, the data interaction based on privacy computing includes: data interaction based on secure multi-party computing MPC.

[0033] In a third aspect, an embodiment provides a building economic data processing system based on privacy computing, the system comprising a first party holding first building economic data and a second party holding second building economic data executing;

[0034] The first party is used to perform privacy processing on the first building economic data to obtain first confidential building economic data;

[0035] The second party is used to perform privacy processing on the second building economic data to obtain second confidential building economic data;

[0036] The first party and the second party are used to process the first confidential building economic data and the second confidential building economic data according to the preset building economic data processing logic through data interaction based on privacy computing to obtain confidential building comprehensive economic data.

[0037] In a fourth aspect, an embodiment provides a building economic data processing device based on privacy computing, which is deployed in a first party holding first building economic data, and the first party is communicatively connected with a second party holding second building economic data; the device includes:

[0038] A privacy processing module, configured to perform privacy processing on the first building economic data to obtain first confidential building economic data;

[0039] The privacy interaction module is configured to process the first confidential building economic data and the second confidential building economic data with the second party through data interaction based on privacy computing in accordance with a preset building economic data processing logic to obtain confidential building comprehensive economic data; wherein the second confidential building economic data is obtained after the second party performs privacy processing on the second building economic data.

[0040] In a fifth aspect, an embodiment provides a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute any one of the methods in the first aspect and the second aspect.

[0041] In a sixth aspect, an embodiment provides a computing device, comprising a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, the method described in any one of the first aspect and the second aspect is implemented.

[0042] In the method and device provided in the embodiments of this specification, multiple parties holding different building economic data respectively perform privacy processing on their respective building economic data to obtain corresponding confidential building economic data, and process the confidential building economic data of multiple parties according to the preset building economic data processing logic through data interaction based on privacy computing. This processing method can realize the processing of building economic data between multiple parties while protecting the original building economic data of multiple parties from being leaked, thereby ensuring that the privacy data of the enterprise is not leaked. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1 A schematic diagram of an implementation scenario of an embodiment disclosed in this specification;

[0045] Figure 2 A flowchart of a building economic data processing method based on privacy computing provided in an embodiment;

[0046] Figure 3 A flowchart of a building economic value assessment method in a certain application scenario;

[0047] Figure 4 A schematic block diagram of a building economic data processing system based on privacy computing provided in an embodiment;

[0048] Figure 5 A schematic block diagram of a building economic data processing device based on privacy computing provided in an embodiment. DETAILED DESCRIPTION

[0049] The solution provided in this specification is described below in conjunction with the accompanying drawings.

[0050] Figure 1 The following is a schematic diagram of an implementation scenario of an embodiment disclosed in this specification. It includes multiple participants such as participant A, participant B and participant C. Multiple participants A, B and C hold different building economic data 1, 2 and 3 respectively. Through data interaction based on privacy computing, joint processing of building economic data held by multiple parties is performed, thereby realizing building economic analysis, but the original building economic data held by each party is not disclosed. Figure 1 The multiple participants shown are only an example. In practical applications, the number of multiple participants can be 2 or more.

[0051] The building economic analysis based on privacy-preserving computing technology performed in the above embodiment is a method that can conduct in-depth analysis of the building's economic activities, operational efficiency, energy consumption, leasing conditions, value assessment and other dimensions under the premise of protecting data privacy. This analysis method utilizes the core principles of privacy computing to ensure that the original data is calculated and analyzed without being leaked, providing valuable information for building managers, investors, policy makers, etc.

[0052] Buildings include commercial buildings, office buildings, complexes, etc. The building mentioned here can be a single building or a park consisting of multiple buildings.

[0053] Building economic data includes economic data of several enterprises in the building. The types of economic data may include basic information of enterprises, lease management data, operation data, energy data, financial data, market performance data and tax data.

[0054] The basic information of an enterprise includes the enterprise name, unified social credit code, registered address, establishment date, and company size (number of employees, office area, etc.).

[0055] Lease management data includes data on office space in buildings leased by enterprises, such as rent, start and end dates, etc.

[0056] Operational data includes the operational data of enterprise employees and the operational data of enterprise equipment. The operational data of enterprise employees includes the number of employees who clocked in and out and the number of visitors during a specified period. The operational data of enterprise equipment includes the duration and frequency of use of specific equipment by the enterprise.

[0057] The energy data consumed by the enterprise includes data such as the consumption of air conditioning, heating, water, electricity, etc. within a preset period.

[0058] Financial data includes the company's turnover (i.e. sales data), profit status and balance sheet, etc. Market performance data includes the company's market share and industry ranking in the industry it is in. Tax data includes the company's actual tax payments, etc.

[0059] Building economic data refers to economic data related to the enterprises settled in the building. These economic data can reflect the economic situation of the building to a certain extent, and the economic situation of the building can reflect the economic situation of the region.

[0060] The above-mentioned various building economic data are usually in the hands of different entities. These entities include building managers, relevant government departments, energy suppliers, building lessors and building owners. Among them, building lessors can be understood as enterprises in buildings, and building owners are the property owners of the leased places in buildings. Building managers can be property management companies of buildings or management committees of parks. Relevant government departments can include tax departments, industrial and commercial management departments and investment promotion bureaus. Energy suppliers include water companies, power companies and heating companies.

[0061] Joint data processing using different types of building economic data held by multiple entities or parties can be used to conduct, but is not limited to, the following building economic analyses: tenant behavior and preference analysis, energy efficiency optimization, asset value assessment, leasing strategy formulation, and risk management and compliance assessment.

[0062] Tenant behavior and preference analysis: Through privacy computing, tenants' consumption habits, preference changes, length of stay, etc. can be analyzed without exposing personal or corporate sensitive information, helping building managers optimize tenant mix and improve user experience and satisfaction.

[0063] Energy efficiency optimization: Building energy consumption is an important part of corporate operating costs. Privacy computing allows energy consumption data to be securely shared between different buildings, and better practices for energy saving can be found through aggregated analysis without worrying about leaking detailed energy consumption details.

[0064] Asset valuation: By combining market data and building-specific information (such as location, age, maintenance records, etc.), privacy computing technology can provide more accurate asset valuation while protecting the privacy of owners, providing support for buying, selling, and leasing decisions.

[0065] Leasing strategy formulation: By analyzing the company's historical leasing data and market trends, privacy computing can help determine better leasing prices, lease structures, etc., while protecting the business secrets of tenants (i.e. companies) and owners, and promoting both parties to reach a fair and reasonable agreement.

[0066] Risk management and compliance assessment: For financial institutions and investors, privacy computing technology can help them assess the risks and compliance of building investments without directly accessing sensitive financial data, thereby enhancing the scientific nature and security of investment decisions.

[0067] When multiple parties jointly process different types of building economic data, they hope to do so while protecting their own privacy data from being leaked. In this regard, the embodiments of this specification provide a building economic data processing method based on privacy computing, which can realize the joint processing of building economic data between multiple parties under the premise of protecting the privacy data in the building economic data, thereby realizing the required building economic analysis.

[0068] Different building economic analyses involve different participants and require the use of different types of building economic data, which are implemented through different building economic data processing logic. For example, in urban development, "100 million yuan buildings" and "quasi-100 million yuan buildings" projects are usually regarded as key factors in promoting regional economic growth and enhancing urban competitiveness. 100 million yuan buildings refer to commercial buildings with a single tax revenue of over 100 million yuan, which usually gather modern service industries dominated by finance, technology, large and multinational companies. Quasi-100 million yuan buildings refer to commercial buildings with tax revenues approaching but not yet reaching 100 million yuan. They are potential buildings that are expected to develop into "100 million yuan buildings" in the future. By evaluating the asset value of a building, it can be determined whether the building belongs to a 100 million yuan building or a quasi-100 million yuan building. When evaluating the asset value of a building, it can be conducted jointly by the building management party and relevant government departments.

[0069] For example, when optimizing the energy efficiency of a building, it can be done jointly by building managers and energy suppliers.

[0070] Regardless of the specific building economic analysis scenario, different parties holding different building economic data will be involved. Through data interaction based on privacy computing between multiple parties, joint processing of building economic data is achieved to obtain processing results. Figure 2 The embodiment is described in detail. The first party A and the second party B are respectively set as privacy computing nodes in the privacy computing platform. Through the means of privacy computing, the first party A and the second party B respectively perform privacy-protected joint data processing based on the first building economic data and the second building economic data held by themselves to obtain confidential building comprehensive economic data, and perform building economic analysis based on the confidential building comprehensive economic data.

[0071] Different parties including the first party A and the second party B respectively execute the method of this embodiment through their own computing devices. The computing device can be implemented by any device, equipment, platform, equipment cluster, etc. with computing and processing capabilities.

[0072] Figure 2 A flowchart of a method for processing building economic data based on privacy computing is provided in an embodiment. The method is performed by a first party A holding first building economic data and a second party B holding second building economic data, and includes the following steps.

[0073] Step S210: The first party A performs privacy processing on the first building economic data to obtain first confidential building economic data.

[0074] The first building economic data is a type of building economic data held by the first party A. The specific types of the first building economic data may include, but are not limited to, one or more of the following data: basic information of several enterprises in the building, leasing management data, operation data, energy consumption data, financial data, market performance data, and tax data, etc.

[0075] The first building economic data is the economic data of several enterprises in a certain building, and is the original data held by the first party A. The first confidential building economic data is the confidential data obtained after the first party A performs privacy processing on the first building economic data. When the confidential data is sent to other parties, the other parties cannot decrypt the original data from the first confidential building economic data in an untrusted environment, nor can they infer relevant information of the original data from it.

[0076] When the first party A performs privacy processing on the first building economic data, a variety of privacy computing technologies can be used.

[0077] For example, when the first building economic data contains multiple data, such as operation data of multiple enterprises, the first party A can perform privacy processing on the first building economic data based on a differential privacy algorithm.

[0078] Differential Privacy (DP) is a technology that protects the privacy of individuals or individual buildings by adding random noise to the data, making it difficult to trace the analysis results to individuals. When building economic analysis requires determining the overall statistics of building economic data, such as average rental levels, total energy consumption, etc., applying differential privacy can ensure that the final published overall statistics will not disclose the specific information of any individual building participating in data sharing.

[0079] In one implementation, the first party A and the second party B may be configured as trusted computing nodes in a trusted computing platform, and both the first party A and the second party B include a trusted execution environment (TEE). When the first party A performs privacy processing on the first building economic data, it may perform privacy processing on the first building economic data in its own TEE.

[0080] Alternatively, the first party A can use the encryption algorithm included in the secure multi-party computation (MPC) to perform privacy processing on the first building economic data. MPC is a cryptographic technology that allows multiple participants to jointly perform calculations and analyses while maintaining the privacy of their respective data. The core idea is to use cryptographic technology and algorithms to encrypt the data of each participant and then perform calculations to ensure that each participant can only obtain the data and calculation results they need, but cannot obtain the original data of other participants. MPC may include algorithms such as homomorphic encryption, secret sharing, zero-knowledge proof, oblivious transmission and obfuscated circuits. For example, the first party A can perform privacy processing on the first building economic data through a homomorphic encryption algorithm, or use a secret sharing algorithm to split the first building economic data into a secret sharing form.

[0081] In specific applications, the first party A can choose to use different privacy processing methods according to different data processing scenarios.

[0082] Step S220: The second party B performs privacy processing on the second building economic data to obtain second confidential building economic data.

[0083] The second building economic data is a type of building economic data held by the second party B. The specific types of the second building economic data may include, but are not limited to, one or more of the following data: basic information of several enterprises in the building, leasing management data, operation data, energy consumption data, financial data, market performance data, and tax data, etc.

[0084] The second building economic data is the economic data of several enterprises in a certain building, and is the original data held by the second party B. The second confidential building economic data is the confidential data obtained after the second party B performs privacy processing on the second building economic data. When the confidential data is sent to other parties, the other parties cannot decrypt the original data from the second confidential building economic data in an untrusted environment, nor can they infer relevant information of the original data from it.

[0085] When the second party B performs privacy processing on the second building economic data, a variety of privacy computing technologies can be used.

[0086] For example, when the second building economic data contains multiple data, such as operation data of multiple enterprises, the second party B can perform privacy processing on the second building economic data based on a differential privacy algorithm.

[0087] In one embodiment, when the first party A and the second party B are configured as trusted computing nodes in a trusted computing platform, when the second party B performs privacy processing on the second building economic data, the second party B may perform privacy processing on the second building economic data in its own TEE.

[0088] Alternatively, the second party B may use the encryption algorithm included in the MPC to perform privacy processing on the second building economic data. For example, the second party B may perform privacy processing on the second building economic data through a homomorphic encryption algorithm, or split the second building economic data into a secret sharing form using a secret sharing algorithm.

[0089] In specific applications, the second party B can choose to use different privacy processing methods according to different data processing scenarios.

[0090] In step S230, the first party A and the second party interact with each other based on privacy computing and jointly process the first confidential building economic data and the second confidential building economic data according to the preset building economic data processing logic to obtain confidential building comprehensive economic data.

[0091] Among them, data interaction based on privacy computing may include data interaction based on MPC.

[0092] The preset building economic data processing logic may include finding intersections, sums, or statistical data of data. The preset building economic data processing logic may be set according to specific building economic analysis requirements. The building economic data processing logic may also be implemented through a building economic calculation model.

[0093] In the above embodiment, the first party A and the second party B respectively perform privacy processing on their respective building economic data. This privacy processing is not simply to encrypt the building economic data using a key, but is performed under the premise that the other party cannot decrypt the plaintext data from the ciphertext data. For example, each party can use a homomorphic encryption algorithm to encrypt their respective building economic data, and each party performs data processing based on the first secret state building economic data and the second secret state building economic data. The data processing process does not require decryption of the ciphertext data, and the ciphertext data can be directly processed accordingly. For example, the specified second secret state building economic data can be directly summed, and the obtained ciphertext is the encryption result after directly summing the plaintext second building economic data. Therefore, the building economic data processing method provided in the embodiment can well protect the economic data privacy of each party from being leaked, and can also realize building economic analysis.

[0094] In another embodiment of the present application, building economic analysis can evaluate the value of a building. In this scenario, the first party A represents the building management party, the second party B represents the relevant government department, the first building economic data includes several enterprises in the building, and the second building economic data includes economic data of multiple enterprises.

[0095] In step S230, when the first party A and the second party B process the first confidential building economic data and the second confidential building economic data through data interaction based on privacy computing, the following steps can be performed:

[0096] Through data interaction based on privacy computing, the first party A and the second party B calculate the intersection of several enterprises held by the first party A and multiple enterprises held by the second party B to obtain several target enterprises. According to the preset building economic data processing logic, the confidential building comprehensive economic data is determined based on the economic data of several target enterprises.

[0097] The relevant government departments hold more economic data of enterprises. By finding the intersection of several enterprises held by the first party A and multiple enterprises held by the second party B, we can get the enterprises held by both parties, that is, the target enterprises.

[0098] The first building economic data and the second building economic data owned by the first party A and the second party B respectively can be regarded as two data sets, and the first secret building economic data and the second secret building economic data are two encrypted data sets. The problem that needs to be solved now is that the two parties find the intersection of the two data sets without leaking the privacy data. Specifically, the Private Set Intersection (PSI) technology can be used to achieve this. The specific process can refer to the existing technology and will not be repeated here.

[0099] In one application scenario, the relevant government department holds a list of support policies, and the names of the companies in the building on the list of support policies can be obtained by finding the intersection of the two.

[0100] In one application scenario, the government supports the park, and all enterprises in the park are supported. In order to verify whether the enterprises in the park are operating normally, the building management party and the relevant government departments can screen out the target enterprises that are operating normally from the target enterprises through data interaction based on privacy computing, based on the operating data of several enterprises in the building included in the first building economic data, and then sum the economic data of the target enterprises that are operating normally through data interaction based on privacy computing to obtain the comprehensive economic data of the confidential building.

[0101] When screening target enterprises that are operating normally, the confidential operating data of several enterprises can be compared with the confidential threshold, so as to screen out enterprises whose plaintext operating data is greater than the plaintext threshold based on the confidential operating data and the confidential threshold as target enterprises that are operating normally.

[0102] The following is an explanation based on a specific example of this application scenario. Figure 3 The figure is a flowchart of a building economic value assessment method in a certain application scenario. It includes data sources, privacy calculations, and application scenarios. In the data source, the building management party holds the unified social credit codes of N enterprises in building X, and the relevant government departments hold the unified social credit codes of multiple enterprises (including the aforementioned N enterprises) and the corresponding tax data and sales data. In the privacy calculation stage, the two parties use the building economic calculation model to achieve the privacy intersection and summation of the data they own, and obtain the confidential total enterprise tax revenue and confidential total enterprise sales of building X, and can decrypt the plaintext data from the confidential total enterprise tax revenue and confidential total enterprise sales. Then, in this application scenario, based on the plaintext data, it can be determined that building X is a billion-yuan building or a quasi-billion-yuan building.

[0103] When the first party A represents the building management party, when the first party A performs privacy processing on the first building economic data in step S210, it can be that the first party A and several enterprises in the building interact with each other based on privacy computing, and the operating data of each enterprise is privacy processed respectively, so that the building management party obtains the first confidential building economic data.

[0104] In this way, each enterprise in the building does not need to send its own operating data in plain text to the building manager, but through the interaction of private data with the building manager, the building manager can obtain the first confidential building economic data.

[0105] For example, each enterprise can serve as a privacy computing node. The building management party holds the date of the required operating data, and each enterprise holds the operating data of multiple dates. The building management party and each enterprise obtain the confidential operating data of the required date through data interaction based on privacy computing, as the first confidential building economic data. Specifically, the date of the operating data required by the building management party can be used as a data set, and the operating data of multiple dates held by an enterprise can be used as a data set. With the help of PSI technology, the data sets are intersected, that is, the building management party obtains the confidential operating data of the enterprise on the corresponding date without the parties leaking their own data. In this way, the enterprise cannot know what date the building management law requires, so it cannot modify the data of the date privately; the building management party cannot know the plaintext operating data either.

[0106] In this embodiment, building managers and relevant government departments can jointly perform privacy data processing and corresponding building economic data processing to achieve building economic analysis based on their respective building economic data through privacy computing.

[0107] In another embodiment of the present application, building economic analysis can optimize the energy mode of the building. The first party A and the second party B can represent different building management parties. The first building economic data and the second building economic data contain energy consumption mode data of the corresponding building.

[0108] In step S230, when the first party A and the second party B process the first confidential building economic data and the second confidential building economic data through data interaction based on privacy computing, the building economic data processing logic is used to optimize the overall energy model based on the energy consumption pattern data contained in the first confidential building economic data and the second confidential building economic data.

[0109] In another embodiment of the present application, in order to achieve more complex and diversified functions, multiple parties can respectively use the building economic data they hold, jointly train the prediction model through federated learning technology, and realize the function of building economic analysis through the prediction model. The prediction model can also be called a building economic calculation model.

[0110] exist Figure 2 Based on the embodiment, the embodiment may further include a third party C, which is used to fuse the gradient data sent by multiple parties and assist multiple parties in jointly training the model. This embodiment may further include the following steps 1 to 5.

[0111] Step 1: The first party A may use the first building economic data to train the first prediction model to obtain the first gradient data, and send the encrypted first gradient data to the third party C.

[0112] Step 2: The second party B uses the second building economic data to train the second prediction model to obtain second gradient data, and sends the encrypted second gradient data to the third party C.

[0113] The first building economic data and the second building economic data have corresponding tags, which are related to the specific building economic analysis task. In the energy consumption optimization scenario, the first building economic data and the second building economic data contain energy consumption mode data of the building, and the tags are used to indicate the energy consumption mode, including high energy consumption mode or low energy consumption mode.

[0114] The first prediction model and the second prediction model can be implemented using existing neural network models, such as convolutional neural networks, recurrent neural networks, deep neural networks, etc. The first prediction model is used to determine the probability of the energy consumption pattern of the building based on the input first building economic data, and the second prediction model is used to determine the probability of the energy consumption pattern of the building based on the input second building economic data. The above process of obtaining the corresponding gradient data based on the training of the first prediction model and the second prediction model can be carried out in an existing manner and will not be described in detail.

[0115] Step 3, the third party C aggregates the encrypted first gradient data sent by the first party A and the encrypted second gradient data sent by the second party B to obtain aggregated gradient data, and feeds the aggregated gradient data back to the first party A and the second party B. The aggregated gradient data is used to update the parameters of the first prediction model and the second prediction model. The process of the third party C aggregating the encrypted first gradient data and the encrypted second gradient data can be performed in an existing manner and will not be described in detail.

[0116] Step 4: The first party A receives the aggregated gradient data sent by the third party C, and updates the first prediction model using the aggregated gradient data. The process of updating the first prediction model using the aggregated gradient data can be performed in an existing manner, and the specific process will not be described in detail.

[0117] Step 5: The second party B receives the aggregated gradient data sent by the third party C, and updates the second prediction model using the aggregated gradient data. The process of updating the second prediction model using the aggregated gradient data can be performed in an existing manner, and the specific process will not be described in detail.

[0118] In this embodiment, multiple parties use their own building economic data to jointly train corresponding prediction models through a federated learning model, which can make full use of the building economic data held by each party, improve the effectiveness of the prediction model, and thus use the prediction model to achieve more effective building economic analysis. In the energy consumption optimization scenario, each party can use its trained prediction model to optimize the energy consumption pattern of the building.

[0119] In another embodiment of the present application, blockchain technology can be used to provide a transparent and trustworthy environment for data sharing and transactions among multiple parties, ensuring that every link in the data flow is traceable, enhancing the trust of all participants, and providing support for the application of privacy computing.

[0120] In this embodiment, the first party A and the second party B are set as nodes in the blockchain network.

[0121] In step S210, when the first party A performs privacy processing on the first building economic data, it can submit a first transaction for invoking the first smart contract to the blockchain network, and its input parameters include the first building economic data. Then, the first party A executes the operation logic of the first smart contract based on the first transaction to perform privacy processing on the first building economic data and obtain the first confidential building economic data.

[0122] In step S220, when the second party B performs privacy processing on the second building economic data, it can submit a second transaction for invoking the second smart contract to the blockchain network, and its input parameters include the second building economic data. Then, the second party B executes the operation logic of the second smart contract based on the second transaction to perform privacy processing on the second building economic data and obtain the second confidential building economic data.

[0123] In step S230, the first party A submits a third transaction for invoking the third smart contract to the blockchain network, and its input parameter includes the first encrypted building economic data. The second party B submits a fourth transaction for invoking the third smart contract to the blockchain network, and its input parameter includes the second encrypted building economic data. Then, the first party A executes the third smart contract based on the third transaction, and the second party B executes the third smart contract based on the fourth transaction. When the third smart contract obtains the first encrypted building economic data and the second encrypted building economic data, it performs specified data processing on them to obtain the encrypted building comprehensive economic data.

[0124] This embodiment uses blockchain to store the joint data processing process and data results between multiple parties in the blockchain network, thereby ensuring that every link in the data flow is traceable, enhancing the trust between the participants and the privacy of the data.

[0125] In the above embodiments, the application of privacy computing technology in high-value building analysis is an innovative data processing and analysis method that aims to protect personal privacy and data security while deeply exploring the value of building-related data. The embodiments ensure that data is analyzed and used without leaking original information through the core principle of data availability and invisibility. The result data obtained from the analysis can provide valuable information for building managers, investors, policy makers, etc.

[0126] In this specification, the word "first" in terms such as first party, first building economic data, first dense building economic data, first gradient data, etc., and the corresponding "second" (if any) in the text, etc., are merely for the convenience of distinction and description and do not have any limiting meaning.

[0127] The foregoing describes certain embodiments of the present specification, and other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in an order different from that in the embodiments, and the desired results may still be achieved. In addition, the processes depicted in the accompanying drawings do not necessarily have to be performed in the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0128] Figure 4 A schematic block diagram of a building economic data processing system based on privacy computing provided in an embodiment. Figure 2 The system 400 includes a first party 410 holding first building economic data and a second party 420 holding second building economic data.

[0129] The first party 410 is used to perform privacy processing on the first building economic data to obtain first confidential building economic data. The second party 420 is used to perform privacy processing on the second building economic data to obtain second confidential building economic data. The first party 410 and the second party 420 are used to process the first confidential building economic data and the second confidential building economic data according to the preset building economic data processing logic through data interaction based on privacy computing to obtain confidential building comprehensive economic data.

[0130] In one implementation, the first party 410 represents the building management party, the second party 420 represents the relevant government department, the first building economic data includes several enterprises in the building, and the second building economic data includes economic data of multiple enterprises. The step of processing the first confidential building economic data and the second confidential building economic data includes: the first party 410 and the second party 420, through data interaction based on privacy computing, obtain several target enterprises by finding the intersection of several enterprises and multiple enterprises, and determine the confidential building comprehensive economic data based on the economic data of the several target enterprises according to the preset building economic data processing logic.

[0131] In one embodiment, the first building economic data also includes operation data of several enterprises in the building. The step of determining the dense building comprehensive economic data based on the economic data of several target enterprises includes: based on the operation data, selecting the target enterprises that are operating normally from the several target enterprises, and determining the dense building comprehensive economic data based on the sum of the economic data of the target enterprises that are operating normally.

[0132] In one embodiment, the first party 410 performs a step of privacy processing on the first building economic data, including: a number of enterprises in the building and the building management party perform privacy processing on the operating data of each enterprise through data interaction based on privacy computing, so that the building management party obtains the first confidential building economic data.

[0133] In one embodiment, the system 400 may further include a third party 430. The first party 410 is further used to train the first prediction model using the first building economic data to obtain first gradient data, and send the encrypted first gradient data to the third party 430. The second party 420 is further used to train the second prediction model using the second building economic data to obtain second gradient data, and send the encrypted second gradient data to the third party 430. The third party 430 is used to aggregate the encrypted first gradient data and the encrypted second gradient data to obtain aggregated gradient data, and the aggregated gradient data is used to update parameters of the first prediction model and the second prediction model.

[0134] In one implementation, the first building economic data includes multiple data; the first party 410 performs privacy processing on the first building economic data, including: performing privacy processing on the first building economic data based on a differential privacy algorithm.

[0135] In one embodiment, the first party 410 and the second party 420 are trusted computing nodes in a trusted computing platform; the first party 410 performs privacy processing on the first building economic data, including: performing privacy processing on the first building economic data in its own trusted execution environment.

[0136] In one embodiment, the first party 410 and the second party 420 are nodes in the blockchain network. The first party 410 is specifically used to submit a first transaction for invoking a first smart contract to the blockchain network, wherein the input parameters include first building economic data, and execute the operation logic of the first smart contract based on the first transaction to perform privacy processing on the first building economic data to obtain first confidential building economic data.

[0137] The above system embodiment corresponds to the method embodiment, and the specific description can refer to the description of the method embodiment part, which will not be repeated here. The system embodiment has the same technical effect as the corresponding method embodiment, and the specific description can refer to the corresponding method embodiment.

[0138] Figure 5 A schematic block diagram of a building economic data processing device based on privacy computing provided in an embodiment. Figure 2The method executed by the first party in the illustrated method embodiment corresponds to the method. The device 500 is deployed in the first party holding the first building economic data, and the first party is in communication connection with the second party holding the second building economic data. The device 500 includes: a privacy processing module 510 and a privacy interaction module 520. Among them, the privacy processing module 510 is configured to perform privacy processing on the first building economic data to obtain the first confidential building economic data. The privacy interaction module 520 is configured to process the first confidential building economic data and the second confidential building economic data with the second party through data interaction based on privacy calculation according to the preset building economic data processing logic to obtain confidential building comprehensive economic data. Among them, the second confidential building economic data is obtained after the second party performs privacy processing on the second building economic data.

[0139] In one implementation, the first party represents the building management party, the second party represents the relevant government department, the first building economic data includes several enterprises in the building, and the second building economic data includes economic data of multiple enterprises.

[0140] The privacy interaction module 520 is specifically configured as follows: through data interaction based on privacy computing with the second party, the intersection of several enterprises and multiple enterprises is obtained to obtain several target enterprises, and according to the preset building economic data processing logic, the confidential building comprehensive economic data is determined based on the economic data of several target enterprises.

[0141] In one implementation, the first building economic data also includes operation data of several enterprises in the building; and the privacy interaction module, when determining the confidential building comprehensive economic data based on the economic data of several target enterprises, includes:

[0142] Based on the operation data, select target enterprises that are operating normally from a number of target enterprises;

[0143] Based on the sum of the economic data of the normally operating target enterprises, the comprehensive economic data of the dense building is determined.

[0144] In one implementation, the privacy processing module 510 is specifically configured as follows:

[0145] Through data interaction based on privacy computing with several enterprises in the building, the operating data of each enterprise is privacy processed, so that the building management party can obtain the first confidential building economic data.

[0146] In one embodiment, the device 500 further includes: a model training module (not shown in the figure), configured to use the first building economic data to train the first prediction model to obtain first gradient data, and send the encrypted first gradient data to a third party, so that the third party aggregates the encrypted first gradient data and the encrypted second gradient data sent by the second party to obtain aggregated gradient data. The second gradient data is obtained by the second party training the second prediction model using the second building economic data, and the aggregated gradient data is used to update the parameters of the first prediction model and the second prediction model.

[0147] In one implementation, the first building economic data includes multiple data. The privacy processing module 510 is specifically configured to: perform privacy processing on the first building economic data based on a differential privacy algorithm.

[0148] In one implementation, the first party and the second party are trusted computing nodes in a trusted computing platform. The privacy processing module 510 is specifically configured to perform privacy processing on the first building economic data in its own trusted execution environment.

[0149] In one embodiment, the first party and the second party are nodes in a blockchain network, and the privacy processing module 510 is specifically configured as follows: submitting a first transaction for calling a first smart contract to the blockchain network, wherein the input parameters include first building economic data; executing the operating logic of the first smart contract based on the first transaction to perform privacy processing on the first building economic data to obtain first confidential building economic data.

[0150] The above device embodiments correspond to the method embodiments. For specific descriptions, please refer to the description of the method embodiments, which will not be repeated here. The device embodiments are obtained based on the corresponding method embodiments and have the same technical effects as the corresponding method embodiments. For specific descriptions, please refer to the corresponding method embodiments.

[0151] The present specification also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed in a computer, the computer is caused to execute Figures 1 to 3 Any of the methods described above.

[0152] The embodiment of the present specification also provides a computing device, including a memory and a processor, wherein the memory stores an executable code, and when the processor executes the executable code, Figures 1 to 3 Any of the methods described above.

[0153] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the storage medium and computing device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0154] Those skilled in the art should be aware that in one or more of the above examples, the functions described in the embodiments of the present invention may be implemented using hardware, software, firmware, or any combination thereof. When implemented using software, these functions may be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium.

[0155] The specific implementation methods described above further describe the purpose, technical solutions and beneficial effects of the embodiments of the present invention in detail. It should be understood that the above description is only a specific implementation method of the embodiments of the present invention and is not intended to limit the scope of protection of the present invention. Any modification, equivalent replacement, improvement, etc. made on the basis of the technical solution of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A method for processing building economic data based on privacy computing, performed by a first party holding first building economic data and a second party holding second building economic data, comprising: The first party performs privacy processing on the first building economic data to obtain first confidential building economic data; The second party performs privacy processing on the second building economic data to obtain second confidential building economic data; The first party and the second party, through data interaction based on privacy computing, process the first confidential building economic data and the second confidential building economic data according to the preset building economic data processing logic to obtain confidential building comprehensive economic data.

2. A method for processing building economic data based on privacy computing, performed by a first party holding first building economic data, the first party being in communication connection with a second party holding second building economic data; the method comprising: Performing privacy processing on the first building economic data to obtain first confidential building economic data; Through data interaction based on privacy computing with the second party, the first confidential building economic data and the second confidential building economic data are processed in accordance with the preset building economic data processing logic to obtain confidential building comprehensive economic data; wherein, the second confidential building economic data is obtained after the second party performs privacy processing on the second building economic data.

3. According to the method of claim 1, the first party represents the building management party, the second party represents the relevant government department, the first building economic data includes several enterprises in the building, and the second building economic data includes economic data of multiple enterprises; the step of processing the first confidential building economic data and the second confidential building economic data comprises: Through data interaction based on privacy computing with the second party, the intersection of the several enterprises and the multiple enterprises is calculated to obtain several target enterprises, and according to the preset building economic data processing logic, the confidential building comprehensive economic data is determined based on the economic data of the several target enterprises.

4. According to the method of claim 3, the first building economic data also includes the operation data of the several enterprises in the building; the step of determining the dense building comprehensive economic data based on the economic data of the several target enterprises comprises: Based on the operation data, selecting target enterprises that are operating normally from the plurality of target enterprises; Based on the sum of the economic data of the normally operating target enterprises, the comprehensive economic data of the dense building is determined.

5. According to the method of claim 3, the step of performing privacy processing on the first building economic data comprises: Through data interaction based on privacy computing with several enterprises in the building, the operating data of each enterprise is privacy processed, so that the building management party can obtain the first confidential building economic data.

6. The method according to claim 2, further comprising: The first prediction model is trained using the first building economic data to obtain first gradient data, and the encrypted first gradient data is sent to a third party, so that the third party aggregates the encrypted first gradient data and the encrypted second gradient data sent by the second party to obtain aggregated gradient data; the second gradient data is obtained by the second party using the second building economic data to train a second prediction model, and the aggregated gradient data is used to update the parameters of the first prediction model and the second prediction model.

7. According to the method of claim 2, the first building economic data comprises a plurality of data; the first party, performing privacy processing on the first building economic data, comprises: The first building economic data is privacy processed based on a differential privacy algorithm.

8. The method according to claim 2, wherein the first party and the second party are trusted computing nodes in a trusted computing platform; the first party, performing privacy processing on the first building economic data, comprises: The first building economic data is privacy-processed in its own trusted execution environment.

9. According to the method of claim 2, the first party and the second party are nodes in a blockchain network, and the first party, performing privacy processing on the first building economic data, comprises: Submitting a first transaction for invoking a first smart contract to the blockchain network, wherein the input parameters include the first building economic data; Based on the first transaction, the operating logic of the first smart contract is executed to perform privacy processing on the first building economic data to obtain first confidential building economic data.

10. The method according to claim 2, wherein the first party and the second party represent any one of the following entities: Building managers, relevant government departments, energy suppliers, building lessors, and building owners; The first building economic data and the second building economic data include one or more of the following data: Basic information of several enterprises in the building, rental management data, operation data, energy consumption data, financial data, market performance data, and tax data.

11. The method according to claim 2, wherein the data interaction based on privacy computing comprises: Data interaction based on secure multi-party computing MPC.

12. A building economic data processing system based on privacy computing, the system comprising a first party holding first building economic data and a second party holding second building economic data executing; The first party is used to perform privacy processing on the first building economic data to obtain first confidential building economic data; The second party is used to perform privacy processing on the second building economic data to obtain second confidential building economic data; The first party and the second party are used to process the first confidential building economic data and the second confidential building economic data according to the preset building economic data processing logic through data interaction based on privacy computing to obtain confidential building comprehensive economic data.

13. A building economic data processing device based on privacy computing, deployed in a first party holding first building economic data, wherein the first party is in communication connection with a second party holding second building economic data; the device comprises: A privacy processing module, configured to perform privacy processing on the first building economic data to obtain first confidential building economic data; The privacy interaction module is configured to process the first confidential building economic data and the second confidential building economic data with the second party through data interaction based on privacy computing in accordance with a preset building economic data processing logic to obtain confidential building comprehensive economic data; wherein the second confidential building economic data is obtained after the second party performs privacy processing on the second building economic data.

14. A computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to execute the method according to any one of claims 1 to 11.

15. A computing device, comprising a memory and a processor, wherein the memory stores executable codes, and when the processor executes the executable codes, the method according to any one of claims 1 to 11 is implemented.

Images