CN113781082B - Method and device for correcting regional portrait, electronic equipment and readable storage medium - Google Patents
Method and device for correcting regional portrait, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN113781082B CN113781082B CN202011291786.9A CN202011291786A CN113781082B CN 113781082 B CN113781082 B CN 113781082B CN 202011291786 A CN202011291786 A CN 202011291786A CN 113781082 B CN113781082 B CN 113781082B
- Authority
- CN
- China
- Prior art keywords
- server
- area
- information
- screening
- region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Marketing (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Processing Or Creating Images (AREA)
Abstract
The disclosure provides a method and a device for correcting a regional portrait, electronic equipment and a computer-readable storage medium, and relates to the field of machine learning. The method for correcting the regional image comprises the following steps: sending the screening area information screened from the plurality of areas to a collaboration server to receive the information of the overlapping area sent by the collaboration server, wherein the information of the overlapping area is generated by the collaboration server according to the screening area information sent by a first server and the screening area information sent by a second server; determining an area to be corrected based on the information of the overlapped area; calling the information of the overlapped area, the cooperative server and the second server to execute interactive training operation so as to generate a correction model according to an interactive training result; and correcting the area to be corrected based on the correction model so as to correct the area images of the plurality of areas. Through the technical scheme, the description accuracy of the region portrait can be improved, and the reliability of subsequent utilization of the region portrait is further improved.
Description
Technical Field
The present disclosure relates to the field of machine learning technologies, and in particular, to a method and an apparatus for correcting a region portrait, an electronic device, and a computer-readable storage medium.
Background
The construction of the regional portrait has important significance for site selection, urban fine management and the like, but because the self operating characteristics of a single organization and the covered user population are limited, a certain target index of a certain region is difficult to accurately portray by utilizing data of a single enterprise side, so in order to obtain a more accurate regional portrait, data fusion is needed among the organizations to correct the target index in the region by combining with multi-party data.
However, if some data among enterprises cannot be shared, the accuracy of regional portrayal is seriously affected, which has great influence on the later urban service construction or business construction.
It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present disclosure, and thus may include information that does not constitute prior art known to those of ordinary skill in the art.
Disclosure of Invention
The present disclosure is directed to a method, an apparatus, an electronic device, and a computer-readable storage medium for correcting a region portrait, which overcome, at least to some extent, the problem of poor description accuracy of the region portrait in the related art.
Additional features and advantages of the disclosure will be set forth in the detailed description which follows, or in part will be obvious from the description, or may be learned by practice of the disclosure.
According to a first aspect of the present disclosure, there is provided a method for correcting a region portrait, including: sending screening area information screened from the plurality of areas to a collaboration server so as to receive information of an overlapping area sent by the collaboration server, wherein the information of the overlapping area is generated by the collaboration server according to the screening area information sent by the first server and the screening area information sent by the second server; determining an area to be corrected based on the information of the overlapped area; calling the information of the overlapped area, the cooperative server and the second server to execute interactive training operation so as to generate a correction model according to an interactive training result; and correcting the area to be corrected based on the correction model so as to correct the area images of the plurality of areas.
In one embodiment, the sending the screened area information screened from the plurality of areas to the collaboration server to receive the information of the overlapped area sent by the collaboration server includes: performing a screening operation on the plurality of areas based on a first screening rule to obtain first screening area information; and sending the first screening area information to a cooperation server, and receiving information of a first overlapping area sent by the cooperation server, wherein the information of the first overlapping area is used for representing an invalid area, and the information of the first overlapping area is generated by the cooperation server according to the first screening area information and second screening area information sent by a second server.
In one embodiment, the sending the screened area information screened from the plurality of areas to the collaboration server to receive the information of the overlapped area sent by the collaboration server further includes: deleting the first overlapped area in the plurality of areas to obtain a residual area; performing screening operation on the residual area based on a second screening rule to obtain first screening area information; and sending the second screening area information to a coordination server, and receiving second overlapping area information sent by the coordination server, wherein the second overlapping area information is used for representing a reliable area, and the second overlapping area information is generated by the coordination server according to the third screening area information and fourth screening area information sent by the second server.
In one embodiment, the modifying the region to be modified based on the modification model to modify the region images of the plurality of regions includes: inputting the regional characteristics of the region to be corrected into the correction model to output corrected target characteristics; replacing original target features in the area to be corrected with the corrected target features to update the target features of the areas and determine target indexes of the areas based on the updated target features of the areas; and correcting the region images of the plurality of regions based on the target index.
In one embodiment, the determining the target indices of the plurality of regions based on the updated target features of the plurality of regions comprises: performing clustering operation on the target characteristics of the plurality of regions, and obtaining a plurality of clustering centers and corresponding clustering clusters; sequencing the plurality of clustering centers, and correspondingly configuring each clustering center with one obtaining area; and matching the clustering clusters to the corresponding scoring areas to generate target indexes of the areas.
In one embodiment, the determining the target indices of the plurality of regions based on the updated target features of the plurality of regions comprises: inputting the target features of the multiple regions into a preset classification model, and outputting target indexes of the multiple regions according to the classification result of the corrected target features by the classification model, wherein historical target indexes are trained on the basis of a supervised learning mode to generate the classification model.
In an embodiment, the invoking of the interactive training operation between the overlap area information and the collaboration server and the second server to generate a modified model according to an interactive training result includes: receiving key information sent by the cooperative server; and calling the key information and the overlapped area information to carry out interactive encryption training of a federated learning model with the second server, and generating the correction model.
According to a second aspect of the present disclosure, there is provided a method for correcting a region portrait, including: respectively receiving screening area information sent by a first server and a second server; taking intersection of the information of the screening areas, generating information of an overlapping area, and sending the information of the overlapping area to the first server and the second server; and performing interactive training operation with the first server and/or the second server based on the overlapping area information, so that the first server and/or the second server generate a correction model according to an interactive training result, and correcting the respective areas to be corrected based on the correction model.
In one embodiment, the receiving the screening area information sent by the first server and the second server respectively includes: receiving first screening information sent by the first server and second screening information sent by the second server so as to obtain intersection of the first screening information and the second screening information; and receiving third screening information sent by the first server and fourth screening information sent by the second server so as to obtain intersection of the third screening information and the fourth screening information.
In one embodiment, the performing interactive training operations with the first server and/or the second server based on the coincidence area information comprises: and respectively sending key information to the first server and the second server so that the first server and/or the second server perform interactive encryption training of a federated learning model based on the key information.
According to a third aspect of the present disclosure, there is provided an area portrait correction apparatus including: the sending module is used for sending the screening area information screened from the plurality of areas to a collaboration server so as to receive the information of the overlapping area sent by the collaboration server, wherein the information of the overlapping area is generated by the collaboration server according to the screening area information sent by the first server and the screening area information sent by the second server; the determining module is used for determining an area to be corrected based on the information of the overlapped area; the interactive training module is used for calling the information of the overlapped area, executing interactive training operation between the cooperative server and the second server and generating a correction model according to an interactive training result; and the correction module is used for correcting the area to be corrected based on the correction model so as to correct the area images of the plurality of areas.
According to a fourth aspect of the present disclosure, there is provided an area portrait correction apparatus, comprising: the transmission module is used for respectively receiving screening area information sent by the first server and the second server; the processing module is used for taking intersection of the information of the screening areas and generating the information of the overlapped areas; a sending module, configured to send the information of the overlapping area to the first server and the second server; and the auxiliary training module is used for executing auxiliary interactive training between the first server and/or the second server based on the overlapping area information so as to enable the first server and/or the second server to generate a correction model according to an interactive training result, and correcting the respective area to be corrected based on the correction model.
According to a fifth aspect of the present disclosure, there is provided an electronic device comprising: a processor; and a memory for storing executable instructions for the processor; wherein the processor is configured to execute any one of the above methods for correcting a region image by executing the executable instructions.
According to a sixth aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of correcting a region representation of any one of the above.
According to the correction scheme of the regional portrait, the screening region information is sent to the cooperative server, the cooperative server receives the overlapping region information obtained by combining the screening region information of the first server and the screening region information of the second server, and the overlapping region information is determined, so that the overlapping regions can be removed from a plurality of regions to obtain the region to be corrected, and data fusion with the second server can be achieved.
Furthermore, a correction model is obtained based on the fusion data, and the region to be corrected is corrected by the correction model, so that on one hand, the description accuracy of the region portrait can be improved, and further the reliability of subsequent utilization of the region portrait is improved, and on the other hand, the collaboration server is used for assisting training in the whole interaction process, and therefore resource occupation of the collaboration server is reduced.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the disclosure. It is to be understood that the drawings in the following description are merely exemplary of the disclosure, and that other drawings may be derived from those drawings by one of ordinary skill in the art without the exercise of inventive faculty.
FIG. 1 is a schematic diagram illustrating a system for correcting a region representation according to an embodiment of the disclosure;
FIG. 2 is a flow chart illustrating a method for correcting a region representation according to an embodiment of the disclosure;
FIG. 3 is a flow chart illustrating another method for correcting a region profile according to an embodiment of the disclosure;
FIG. 4 is a flow chart illustrating a method for correcting a region profile according to another embodiment of the disclosure;
FIG. 5 is a flow chart illustrating a method for correcting a region profile according to another embodiment of the disclosure;
FIG. 6 illustrates an interactive schematic of a region representation correction scheme in accordance with an embodiment of the disclosure;
FIG. 7 is a schematic diagram illustrating an apparatus for correcting a region image according to an embodiment of the disclosure;
FIG. 8 is a schematic diagram of another device for correcting a region image according to an embodiment of the disclosure;
fig. 9 shows a schematic diagram of an electronic device in an embodiment of the disclosure.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Furthermore, the drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.
The scheme that this application provided obtains the correction model through setting up the fusion data to adopt the correction model to treat the portrayal of revising the region and revise, on the one hand, can improve the precision to the description of region portrayal, and then improve the reliability to the follow-up utilization of region portrayal, on the other hand, at whole interactive process, the server in coordination is used for assisting the training, thereby is favorable to reducing the resource occupation to the server in coordination.
For ease of understanding, the following first explains several terms referred to in this application.
Urban portraits (multi-factor portraits based on big data and machine learning) are SaaS products oriented to planning, real estate, retail and a plurality of GIS application industries. The innovation point of the method is that on one hand, machine learning calculation and interactive visualization are integrated, and the limitation of the traditional GIS on multi-dimensional/high-dimensional spatio-temporal data exploration and analysis is broken through from the technical method; on the other hand, a huge amount of city data, spark/elastic search big data processing engines, distributed computing, online data processing, online index computing and multi-factor mining analysis are integrated, an SaaS service which is extremely convenient, easy to use and powerful in function is created, the professional barrier of a GIS is broken, each user is enabled, and data acquisition, data processing and multi-factor spatial data mining become efficient and easy. Meanwhile, API/SDK secondary development is supported, and the system is easily accessed to the existing platform of the user.
Federal learning: when multiple data owners (e.g., enterprises) Fi (i =1\ … \ N) want to train a machine learning model in conjunction with their respective data Di, the conventional practice is to integrate the data into one side and train with data D = { Di \ i =1\ … \ N } and obtain model M _ sum. However, this solution is often difficult to implement due to legal issues relating to privacy and data security. To address this problem, we introduced federal learning. The federal learning means that the data owner Fi can perform model training to obtain a calculation process of the model M _ fed without giving own data Di, and can ensure that the difference between the effect V _ fed of the model M _ fed and the effect V _ sum of the model M _ sum is small enough, namely | V _ fed-V _ sum | < δ, where δ is an arbitrarily small positive value.
Multi-party loan: it means that a bad user has money to another financial institution after loan, and a lot of illegal actions may cause the whole financial system to be broken down. To discover such users, it is conventional practice for the financial institution to query a central database for user information, and each institution must upload all their user information, but this is equivalent to exposing all important user privacy and data security of the financial institution, which is not allowed under the GDPR. Under the federal learning mechanism, a central database does not need to be established, and any financial institution participating in the federal learning can send a new user query request to other institutions in the federal, and the other institutions answer questions of the user about local loan on the premise of not knowing specific information of the user. Therefore, the privacy and data integrity of the existing user at each financial institution can be protected, and the important problem of inquiring multi-party loan can be solved.
The scheme provided by the embodiment of the application relates to technologies such as network modeling and machine learning, and is specifically explained by the following embodiment.
FIG. 1 is a schematic structural diagram of a system for region portrait correction in an embodiment of the present disclosure, including a plurality of terminals 120 and a server cluster 140.
The terminal 120 may be a mobile terminal such as a mobile phone, a game console, a tablet Computer, an e-book reader, smart glasses, an MP4 (Moving Picture Experts Group Audio Layer IV) player, an intelligent home device, an AR (Augmented Reality) device, a VR (Virtual Reality) device, or a Personal Computer (Personal Computer), such as a laptop Computer and a desktop Computer.
Among them, the terminal 120 may be installed with an application program for providing correction of the area representation.
The terminals 120 are connected to the server cluster 140 through a communication network. Optionally, the communication network is a wired network or a wireless network.
The server cluster 140 is a server, or is composed of a plurality of servers, or is a virtualization platform, or is a cloud computing service center. The server cluster 140 is used to provide background services for a correction application that provides a region profile and a training application that provides a traffic prediction model. Optionally, the server cluster 140 undertakes primary computational work and the terminal 120 undertakes secondary computational work; alternatively, the server cluster 140 undertakes secondary computing work and the terminal 120 undertakes primary computing work; alternatively, the terminal 120 and the server cluster 140 perform cooperative computing by using a distributed computing architecture.
In some alternative embodiments, the server cluster 140 is used to store the region representation correction model and prediction method.
Alternatively, the clients of the applications installed in different terminals 120 are the same, or the clients of the applications installed on two terminals 120 are clients of the same type of application of different control system platforms. Based on different terminal platforms, the specific form of the client of the application program may also be different, for example, the client of the application program may be a mobile phone client, a PC client, or a World Wide Web (Web) client.
Those skilled in the art will appreciate that the number of terminals 120 described above may be greater or fewer. For example, the number of the terminals may be only one, or several tens or hundreds of the terminals, or more. The number of terminals and the type of the device are not limited in the embodiments of the present application.
Optionally, the system may further include a management device (not shown in fig. 1), and the management device is connected to the server cluster 140 through a communication network. Optionally, the communication network is a wired network or a wireless network.
Optionally, the wireless network or wired network described above uses standard communication techniques and/or protocols. The Network is typically the Internet, but may be any Network including, but not limited to, a Local Area Network (LAN), a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a mobile, wireline or wireless Network, a private Network, or any combination of virtual private networks. In some embodiments, data exchanged over a network is represented using techniques and/or formats including Hypertext Mark-up Language (HTML), extensible markup Language (XML), and the like. All or some of the links may also be encrypted using conventional encryption techniques such as Secure Socket Layer (SSL), transport Layer Security (TLS), virtual Private Network (VPN), internet protocol Security (IPsec). In other embodiments, custom and/or dedicated data communication techniques may also be used in place of, or in addition to, the data communication techniques described above.
The following describes the steps of the region profile correction method and the flow prediction model training method in this exemplary embodiment in more detail with reference to the drawings and examples.
FIG. 2 is a flowchart illustrating a method for correcting a region portrait according to an embodiment of the disclosure. The method provided by the embodiment of the present disclosure may be performed by any electronic device with computing processing capability, for example, the terminal 120 and/or the server cluster 140 in fig. 1. In the following description, the server cluster 140 is used as an execution subject for illustration.
The first server and the second server may be multiple, and the collaboration server interacts with the first server and the second server respectively.
As shown in fig. 2, the server cluster 140 is specifically a first server, and performs a method for correcting a region representation, including the following steps:
step S202, sending the information of the screening area screened from the plurality of areas to the collaboration server to receive the information of the overlapping area sent by the collaboration server, where the information of the overlapping area is generated by the collaboration server according to the information of the screening area sent by the first server and the information of the screening area sent by the second server.
The screening area information may specifically be an ID of the screening area.
The method includes the steps of performing grid division by adopting Geohash6, geohash7 or other division methods to generate a plurality of areas, specifically, preprocessing characteristic data such as area population, geographic position and POI (Point of interest) in an enterprise database to obtain area characteristics corresponding to an area grid, and taking area consumption and the like as target characteristics of the area to obtain an area portrait of the area based on the characteristics and the target characteristics.
In addition, the collaboration server may be understood as a collaboration platform, which is used to perform a filtering operation based on the filtering area information sent by the first server and the second server, and is used to assist the first server and/or the second server in model training under the condition that the safety of the feature information of the plurality of areas on the first server and/or the second server is ensured. The second server is specifically information for parameter exchange with the first server, and further enables the second server to execute the same processing procedure with the first server in combination with the cooperative operation of the cooperative server, where different regional characteristics of the same region may be stored in the first server and the second server.
Step S204, determining the area to be corrected based on the information of the overlapped area.
It can be understood by those skilled in the art that the filtering area information filtered from the plurality of areas may be executed based on a filtering rule, and the area information filtered by the filtering rule is subjected to intersection-taking operation on the cooperative server to determine the areas which are common to the first server and the second server and satisfy the filtering rule at the same time, and the purpose of the filtering is to obtain the areas which do not need to be corrected, so as to obtain the areas to be corrected by removing the areas which do not need to be corrected from the plurality of areas.
Specifically, the cooperative server performs a screening operation on screening area information sent by the first server and the second server to obtain overlapping area information, on one hand, an area to be corrected can be obtained by removing the overlapping area from a plurality of areas, and on the other hand, relevant data of the second server can be introduced based on the overlapping area information by obtaining the overlapping area information, so that data fusion between different servers is realized.
And step S206, calling the overlapping area information, the cooperative server and the second server to execute interactive training operation so as to generate a correction model according to an interactive training result.
The correction model may be generated by performing interactive training with the collaboration server and the second server using the overlapping area information as training data.
In step S208, the region to be corrected is corrected based on the correction model to correct the region images of the plurality of regions.
In this embodiment, the modification model is used to modify a target feature with poor reliability in a plurality of areas, such as a target feature of an area with low consumption index reliability, and the target feature may be consumption data.
A region representation may be understood as a representation generated based on characteristics such as regional demographics, geographic location, regional POI (Point of interest), regional consumption, etc.
In this embodiment, the multiple regions may include an overlapping region and a region to be corrected, the screening region information is sent to the cooperative server, the cooperative server receives overlapping region information obtained by combining the screening region information of the first server and the screening region information of the second server, and the overlapping region information is determined, so that not only the overlapping region can be removed from the multiple regions to obtain the region to be corrected, but also data fusion with the second server can be achieved.
Furthermore, a correction model is obtained based on the fusion data, and the region to be corrected is corrected by the correction model, so that on one hand, the description accuracy of the region portrait can be improved, and further the reliability of subsequent utilization of the region portrait is improved, and on the other hand, the collaboration server is used for assisting training in the whole interaction process, and therefore resource occupation of the collaboration server is reduced.
For example, the subsequent use of the region picture includes a business scenario such as a region address and advertisement placement in a later stage.
In one embodiment, sending the screened area information screened from the plurality of areas to the collaboration server to receive the information of the overlapped area sent by the collaboration server includes: performing a screening operation on the plurality of areas based on a first screening rule to obtain first screening area information; and sending the first screening area information to a coordination server, and receiving information of a first overlapping area sent by the coordination server, wherein the information of the first overlapping area is used for representing an invalid area, and the information of the first overlapping area is generated by the coordination server according to the first screening area information and second screening area information sent by a second server.
Taking the consumption index as an example, the reliable region refers to a region with the consumption index reliability of 0, that is, the consumption data of the region is considered to not reflect the real consumption level.
For example, the first filtering rule may be: population total < C & feature POI total < D & spending amount < E to screen out regions with a spending index of 0. And the first server and the second server respectively screen out areas with self characteristics meeting the first screening rule, and transmit the areas to the cooperative server.
In one embodiment, sending the screened area information screened from the plurality of areas to the collaboration server to receive the information of the overlapped area sent by the collaboration server further comprises: deleting a first overlapped area in the plurality of areas to obtain a remaining area; performing screening operation on the remaining area based on a second screening rule to obtain first screening area information; and sending the second screening area information to the collaboration server, and receiving second overlapping area information sent by the collaboration server, wherein the second overlapping area information is used for representing a reliable area, and the second overlapping area information is generated by the collaboration server according to the third screening area information and fourth screening area information sent by the second server.
Taking the consumption index as an example, the reliable region refers to a region with high reliability of the consumption index, that is, the consumption data of the region is considered to reflect the real consumption level.
The second filtering rule may be generated based on the above feature, and the second filtering rule may be set such that the area population is larger than the total of a & POI and larger than 0, and the consumption amount is larger than or equal to B, taking the consumption index of the area as an example, and the area with high reliability of the consumption index may be filtered based on the second filtering rule.
In the embodiment, the area with the consumption index of 0 is screened out firstly, the remaining area is the area with the consumption index of not 0, the areas with high index reliability and low index reliability are further divided, so that the area with high index reliability can be used for carrying out federal modeling, and then the target characteristics of the area with low index reliability are re-determined, so that the region portrait is revised.
In one embodiment, in step S206, generating the region to be corrected based on the information of the first overlapping region includes: and deleting the first overlapping area and the second overlapping area in the plurality of areas to obtain an area to be corrected.
In this embodiment, by setting the first filtering rule and the second filtering rule, a first overlapping area, that is, an invalid area, and a second overlapping area, that is, a reliable area, are determined in the plurality of areas, and the first overlapping area and the second overlapping area are removed from the plurality of areas, and the remaining area is an area with low reliability, that is, an area that needs to be corrected.
As shown in fig. 3, in an embodiment, the step S208 of correcting the area to be corrected based on the correction model to correct the area image of the area to be corrected includes:
step S302, inputting the regional characteristics of the region to be corrected into the correction model to output the corrected target characteristics.
And step S304, replacing the original target features in the area to be corrected with the corrected target features to update the target features of the plurality of areas.
Specifically, a trained model is used for deducing an area with low target index reliability, and target features of the area are obtained again to replace original inaccurate area target features.
Determining target indexes of the multiple regions based on the updated target features of the multiple regions, specifically comprising:
step S306, clustering operation is carried out on the target characteristics of the plurality of areas, and a plurality of clustering centers and corresponding clustering clusters are obtained.
Step S308, a plurality of clustering centers are sequenced, and each clustering center is correspondingly configured with an obtaining area.
Step S310, the cluster clusters are matched to the corresponding scoring areas to generate target indexes of a plurality of areas.
In step S312, the area images of the plurality of areas are corrected based on the target index.
In the embodiment, the modified consumption data are clustered into 5-10 classes, and the specific number is judged according to the scene or service requirements.
After the target features are clustered, the clustering centers are sorted from small to large, and the index data of the corresponding clustering clusters are sequentially matched to corresponding scoring areas, so that the final index score is between 0 and 100, and the target index score is the modified urban area target index. The obtained accurate image target index can be used for consumption analysis, regional labor consumption estimation and the like.
In one embodiment, determining the target indexes of the plurality of regions based on the updated target features of the plurality of regions may be further implemented by: inputting the target characteristics of the multiple regions into a preset classification model, and outputting a target index of the region to be corrected by the classification model according to the classification result of the corrected target characteristics, wherein the historical target index is trained based on a supervised learning mode to generate the classification model.
In one embodiment, invoking interactive training operations between the coincidence area information and the collaboration server and the second server to generate the modified model according to the interactive training result comprises: receiving key information sent by a cooperative server; and calling the key information and the overlapped area information to carry out interactive encryption training of the Federal learning model with the second server, and generating a correction model.
In this embodiment, when the index correction of the city region representation is performed, the index correction of the region representation can be performed by using federal learning without going out of the enterprise data. The method utilizes the region with high credibility of the portrait index to train the federal model and revise the index characteristics of the region with low credibility. Compared with the area portrait obtained by single data, the area portrait correction technology of the text can achieve the aim of describing more accurate city area portrait on the premise of protecting the safety of multi-data so as to serve the application scene of the later area portrait.
In addition, under the condition that the original data is not exported, a method based on multi-party data security privacy protection cross-domain modeling, such as multi-party security calculation, can be adopted to replace a federal learning algorithm.
Specifically, we take a scenario involving two data owners (i.e., a first server and a second server) as an example to introduce a system architecture for federated learning. The framework is extensible to scenarios involving multiple data owners. Suppose that a first server and a second server jointly train a machine learning model, and business systems of the first server and the second server respectively have relevant data of respective users. In addition, the second server also has label data that the model needs to predict. Due to data privacy protection and safety considerations, the first server and the second server cannot directly exchange data, and a federal learning system can be used for establishing a model. The federal learning system framework is composed of three parts.
A first part: the encrypted samples are aligned. Because the user groups of the two users are not completely overlapped, the system confirms the common users of the two users on the premise that the first server and the second server do not disclose respective data by using an encryption-based user sample alignment technology, and does not expose the users which are not overlapped with each other, so that the modeling is performed by combining the characteristics of the users.
A second part: and (5) training an encryption model. After the common user population is determined, the machine learning model can be trained using these data. In order to ensure the confidentiality of data in the training process, a third party is required to cooperate with a server to perform encryption training.
As shown in fig. 4, taking a linear regression model as an example, the training process includes:
step S402, the cooperative server distributes the public key to the first server and the second server for encrypting the data to be exchanged in the training process.
And S404, interacting the intermediate result used for calculating the gradient between the first server and the second server in an encrypted form.
Step S406, the first server and the second server respectively calculate based on the encrypted gradient values, and meanwhile the second server calculates loss according to the tag data and collects the result to the cooperative server.
In step S408, the collaboration server calculates the total gradient value from the summary result and decrypts the total gradient value.
Step S410, the cooperative server returns the decrypted gradients to the first server and the second server, respectively.
In step S412, the first server and the second server update the parameters of the respective models according to the gradient.
Step S414, iterate the above steps until the loss function converges to generate a modified model.
In the sample alignment and model training process, the data of the first server and the data of the second server are kept locally, and data privacy disclosure cannot be caused by data interaction in training. Thus, both parties are enabled to collaboratively train the model with the help of federal learning.
And a third part: and (4) effect excitation. The model effect obtained by the organization providing the data is better, and the model effect depends on the contribution of the data provider to the organization and others. The effects of these models are distributed to individual agencies on the federal mechanism for feedback and continue to encourage more agencies to join this data federation.
As shown in fig. 5, the server cluster 140 is specifically a cooperative server, and a method for correcting a region portrait according to another embodiment of the present disclosure includes:
step S502, respectively receiving the screening area information sent by the first server and the second server.
Step S504, the intersection of the information of the screening areas is taken to generate the information of the overlapping area.
Step S506, the information of the overlapping area is sent to the first server and the second server.
Step S508, performing an interactive training operation with the first server and/or the second server based on the overlapping area information, so that the first server and/or the second server generates a correction model according to an interactive training result, and corrects the respective areas to be corrected based on the correction model.
In this embodiment, at the cooperative server side, an intersection operation is performed on the screening area sent by the first server and the screening area sent by the second server to obtain area IDs simultaneously possessed by the first server and the second server, so as to determine an overlapping area in the plurality of areas, where the overlapping area may include a reliable area, and further, in a training process of a correction model, model training may be performed in combination with feature information of the overlapping area stored on the first server and feature information of the overlapping area stored on the second server, and a correction model is obtained based on the fusion data, so as to correct an image of the area to be corrected by using the correction model.
In one embodiment, the receiving the screening area information sent by the first server and the second server respectively comprises: receiving first screening information sent by a first server and second screening information sent by a second server so as to obtain intersection of the first screening information and the second screening information; and receiving third screening information sent by the first server and fourth screening information sent by the second server so as to obtain intersection of the third screening information and the fourth screening information.
In this embodiment, at the collaboration server, a first overlapping area, that is, an invalid area, in the multiple areas is determined by performing the first screening area and the second screening area, and a second overlapping area, that is, a reliable area, in the multiple areas is determined by performing the third screening area and the fourth screening area, so that the first overlapping area and the second overlapping area are removed from the multiple areas at the first server and the second server, and the remaining area is an area with low reliability, that is, an area that needs to be corrected.
In one embodiment, performing interactive training operations with the first server and/or the second server based on the coincidence area information includes: and respectively sending the key information to the first server and the second server so that the first server and/or the second server perform interactive encryption training of the federated learning model based on the key information.
In the embodiment, by sending the key information to the first server and the second server, the regional portrait correction technology of the text can be used for depicting more accurate urban regional portrait on the premise of protecting the security of multiple data, so as to serve the application scene of the later regional portrait.
The following further describes a modification scheme of the region representation of the present disclosure with reference to fig. 6, taking modifying the city region consumption index as an example, taking the consumption data as a target feature, and taking the consumption index as a target index.
Each organization (including but not limited to the first server 10 and the second server 20 as examples) performs grid division on the city area according to the Geohash7 (which may adopt the Geohash6 or other division methods), and performs preprocessing on data such as regional population, geographic location, regional consumption, regional POI and the like in the enterprise database to obtain characteristics corresponding to the regional grid, for example, obtains regional consumption characteristics by matching order data to addresses and consumption amounts. In combination with the data characteristics of the two servers, a first screening rule (population total <10& feature POI total <1& consumption amount < 100) is formulated to screen out the area with the consumption index of 0, and a second screening rule (population total >3 &poitotal >0& consumption amount > = 100) is formulated to screen out the area with high reliability of the consumption index, and the correction process is completed in combination with the cooperation server 30.
Step S602, the first server and the second server respectively screen out an area whose own characteristics satisfy the first screening rule, and transmit the area to the collaboration server.
Step S604, the collaboration server collects the area ID sets transmitted by each server, and then obtains the intersection of the ID sets to obtain the information of the first overlapping area.
In step S606, the collaboration server transmits the information of the first overlapping area to each server.
The intersection region is defined as a region with a consumption index of 0, and the remaining regions are regions with high and low consumption index credibility.
And combining the platform characteristics of the two parties, and considering that the consumption data of the partial area can reflect the real consumption level of the area.
In step S608, after deleting the area with the consumption index of 0, the first server and the second server respectively screen out the areas whose own characteristics satisfy the second screening characteristics, and transmit the areas to the collaboration server.
Step S610, the collaboration server collects the area ID sets transmitted by each server, and then obtains the intersection of the ID sets to obtain the information of the second overlapping area.
In step S612, the collaboration server transmits the information of the second overlapping area to each server.
And defining the intersection region as a region with high consumption index reliability.
In step S614, the area with the elimination index of 0 and high reliability is removed, and the remaining area is the area to be corrected with low consumption reliability.
And step S616, performing id alignment on the areas with low reliability and high reliability of both parties, performing federal modeling by using the areas with high reliability as training data, adjusting the parameters to train the model for multiple times, selecting proper parameters to train the optimal model to be used as a correction model, and storing the model to the local by the servers of both parties respectively.
In addition, different federal models (such as federal Boosting and federal forest) can be adjusted and selected for multiple times to select proper parameters to train the optimal model, and the platforms of the two parties respectively store the model to the local.
And step S618, deducing an area with low consumption index reliability by using the trained model, and obtaining the consumption data of the area again to replace the original inaccurate area consumption data.
And S620, clustering the corrected consumption data to obtain a corrected urban area consumption index.
Specifically, the general clustering can be 5-10 classes, and the specific number is judged according to the scene or the service requirement. After the consumption data are clustered, the clustering centers are sorted from small to large, and the index data of the corresponding clustering clusters are sequentially matched to corresponding scoring areas, so that the final index score is between 0 and 100, and the consumption index score is the modified urban area consumption index. The obtained accurate portrait consumption index can be used for consumption analysis, regional labor consumption estimation and the like.
It is to be noted that the above-mentioned figures are only schematic illustrations of the processes involved in the method according to an exemplary embodiment of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
A region image correction apparatus 700 according to this embodiment of the present invention is described below with reference to fig. 7. The area image correction apparatus 700 shown in fig. 7 is merely an example, and should not limit the function and the application range of the embodiment of the present invention.
The area image correction device 700 is expressed in the form of a hardware module. Components of the area representation correction apparatus 700 may include, but are not limited to: a transmission module 702, configured to send the screening area information screened from the multiple areas to the collaboration server, so as to receive information of an overlapping area sent by the collaboration server, where the information of the overlapping area is generated by the collaboration server according to the screening area information sent by the first server and the screening area information sent by the second server; a determining module 704, configured to determine an area to be corrected based on the information of the overlapping area; the interactive training module 706 is configured to invoke interactive training operations between the overlapping area information and the collaboration server and between the overlapping area information and the second server, so as to generate a correction model according to an interactive training result; and a correcting module 708, configured to correct the area to be corrected based on the correction model, so as to correct the area images of the plurality of areas.
A region image correction apparatus 800 according to this embodiment of the present invention is described below with reference to fig. 8. The area image correction apparatus 800 shown in fig. 8 is only an example, and should not bring any limitation to the function and the range of use of the embodiment of the present invention.
The area image correction device 800 is expressed in the form of a hardware module. Components of the region representation correction apparatus 800 may include, but are not limited to: a receiving module 802, configured to receive screening area information sent by a first server and a second server respectively; the processing module 804 is configured to take an intersection from the information of the screening regions to generate information of the overlapping region; a sending module 806, configured to send information of the overlapping area to the first server and the second server; an auxiliary training module 808, configured to perform auxiliary interactive training with the first server and/or the second server based on the overlapping area information, so that the first server and/or the second server generate a modification model according to an interactive training result, and modify the respective area to be modified based on the modification model.
An electronic device 900 according to this embodiment of the invention is described below with reference to fig. 9. The electronic device 900 shown in fig. 9 is only an example and should not bring any limitations to the function and scope of use of the embodiments of the present invention.
As shown in fig. 9, the electronic device 900 is embodied in the form of a general purpose computing device. Components of electronic device 900 may include, but are not limited to: the at least one processing unit 910, the at least one memory unit 920, and a bus 930 that couples various system components including the memory unit 920 and the processing unit 910.
Where the storage unit stores program code, which may be executed by the processing unit 1010, to cause the processing unit 910 to perform the steps according to various exemplary embodiments of the present invention described in the above-mentioned "exemplary methods" section of this specification. For example, the processing unit 1010 may perform steps S202, S204 to S210 as shown in fig. 2, and other steps defined in the correction method of the area representation of the present disclosure.
The storage unit 920 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 9201 and/or a cache storage unit 9202, and may further include a read only storage unit (ROM) 9203.
The electronic device 900 may also communicate with one or more external devices 960 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 900 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interface 950. Also, the electronic device 900 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network such as the Internet) via the network adapter 950. As shown, the network adapter 950 communicates with the other modules of the electronic device 900 over a bus 930. It should be appreciated that although not shown in the figures, other hardware and/or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.
In an exemplary embodiment of the present disclosure, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above-described method of the present specification. In some possible embodiments, aspects of the invention may also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps according to various exemplary embodiments of the invention described in the above-mentioned "exemplary methods" section of the present description, when the program product is run on the terminal device.
According to the program product for realizing the method, the portable compact disc read only memory (CD-ROM) can be adopted, the program code is included, and the program product can be operated on terminal equipment, such as a personal computer. However, the program product of the present invention is not limited in this regard and, in the present document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
Moreover, although the steps of the methods of the present disclosure are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order, or that all of the depicted steps must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions, etc.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (which may be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which may be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
Claims (14)
1. A method for correcting a region portrait, applied to a first server, includes:
sending screening area information screened from the plurality of areas to a collaboration server so as to receive information of an overlapping area sent by the collaboration server, wherein the information of the overlapping area is generated by the collaboration server according to the screening area information sent by the first server and the screening area information sent by the second server;
determining an area to be corrected based on the information of the overlapping area, specifically comprising: removing the overlapped areas from the plurality of areas to obtain the area to be corrected;
calling the information of the overlapped area, the cooperative server and the second server to execute interactive training operation so as to generate a correction model according to an interactive training result;
and correcting the area to be corrected based on the correction model so as to correct the area images of the plurality of areas.
2. The method for correcting a region portrait according to claim 1, wherein the sending the information on the selected region selected from the plurality of regions to the collaboration server to receive the information on the overlapped region sent by the collaboration server includes:
performing a screening operation on the plurality of areas based on a first screening rule to obtain first screening area information;
sending the first screening area information to a collaboration server, and receiving information of a first overlapping area sent by the collaboration server,
the information of the first overlapping area is used for representing an invalid area, and the information of the first overlapping area is generated by the cooperation server according to the first screening area information and second screening area information sent by the second server.
3. The method for correcting a region representation according to claim 2, wherein said transmitting the information on the selected region selected from the plurality of regions to the collaboration server to receive the information on the overlapped region transmitted from the collaboration server further comprises:
deleting the first overlapped area in the plurality of areas to obtain a remaining area;
performing screening operation on the residual area based on a second screening rule to obtain third screening area information;
sending the third screening area information to a collaboration server, and receiving information of a second rendezvous area sent by the collaboration server,
the information of the second overlapping area is used for representing a reliable area, and the information of the second overlapping area is generated by the cooperation server according to the third screening area information and fourth screening area information sent by the second server.
4. The method for correcting a region representation according to claim 1, wherein the correcting the region to be corrected based on the correction model to correct the region representations of the plurality of regions comprises:
inputting the regional characteristics of the region to be corrected into the correction model to output corrected target characteristics;
replacing the original target features in the region to be corrected with the corrected target features to update the target features of the plurality of regions;
determining target indices of the plurality of regions based on the updated target features of the plurality of regions;
and correcting the region images of the plurality of regions based on the target index.
5. The method of modifying a region representation of claim 4, wherein said determining the target indices of the plurality of regions based on the updated target features of the plurality of regions comprises:
performing clustering operation on the target characteristics of the plurality of regions, and obtaining a plurality of clustering centers and corresponding clustering clusters;
sequencing the plurality of clustering centers, and correspondingly configuring each clustering center with one obtaining area;
and matching the clustering clusters to the corresponding scoring areas to generate target indexes of the areas.
6. The method of modifying a region representation of claim 4, wherein said determining the target indices of the plurality of regions based on the updated target features of the plurality of regions comprises:
inputting the target characteristics of the plurality of regions into a preset classification model, outputting target indexes of the plurality of regions according to the classification result of the corrected target characteristics by the classification model,
wherein historical target indices are trained based on supervised learning to generate the classification model.
7. The method for correcting a region portrait according to any one of claims 1 to 6, wherein the invoking of the interactive training operation between the overlay region information and the collaboration server and the second server to generate a correction model according to the interactive training result includes:
receiving key information sent by the cooperative server;
and calling the key information and the overlapped area information to carry out interactive encryption training of a federated learning model with the second server, and generating the correction model.
8. A method for correcting a region portrait, which is applied to a collaboration server, includes:
respectively receiving screening area information sent by a first server and a second server;
taking intersection of the information of the screening areas to generate information of a coincidence area;
sending the information of the overlapping area to the first server and the second server so that the overlapping area is removed from a plurality of areas by the first server and/or the second server to obtain an area to be corrected;
and performing interactive training operation with the first server and/or the second server based on the overlapping area information, so that the first server and/or the second server generate a correction model according to an interactive training result, and correcting the respective areas to be corrected based on the correction model.
9. The method for correcting an area portrait according to claim 8, wherein the receiving the filtered area information transmitted from the first server and the second server respectively comprises:
receiving first screening information sent by the first server and second screening information sent by the second server so as to obtain intersection of the first screening information and the second screening information; and
and receiving third screening information sent by the first server and fourth screening information sent by the second server so as to obtain intersection of the third screening information and the fourth screening information.
10. The method for correcting a region portrait according to claim 8, wherein the performing interactive training operation with the first server and/or the second server based on the overlapping region information includes:
and respectively sending key information to the first server and the second server so that the first server and/or the second server perform interactive encryption training of a federated learning model based on the key information.
11. An apparatus for correcting a region image, adapted to a first server, comprising:
the transmission module is used for sending the screening area information screened from the plurality of areas to the collaboration server so as to receive the information of the overlapping area sent by the collaboration server, wherein the information of the overlapping area is generated by the collaboration server according to the screening area information sent by the first server and the screening area information sent by the second server;
a determining module, configured to determine, based on the information of the overlapping area, an area to be corrected, and specifically configured to: removing the overlapped areas from the plurality of areas to obtain the area to be corrected;
the interactive training module is used for calling the information of the overlapped area, executing interactive training operation between the cooperative server and the second server and generating a correction model according to an interactive training result;
and the correction module is used for correcting the area to be corrected based on the correction model so as to correct the area images of the plurality of areas.
12. An apparatus for correcting a region image, which is applied to a collaboration server, comprising:
the receiving module is used for respectively receiving screening area information sent by the first server and the second server;
the processing module is used for taking intersection of the information of the screening areas and generating the information of the overlapped areas;
the sending module is used for sending the information of the overlapped area to the first server and the second server so as to enable the first server and/or the second server to remove the overlapped area from a plurality of areas to obtain an area to be corrected;
and the auxiliary training module is used for executing auxiliary interactive training between the first server and/or the second server based on the overlapping area information so as to enable the first server and/or the second server to generate a correction model according to an interactive training result, and correcting the respective area to be corrected based on the correction model.
13. An electronic device, comprising:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to execute the method of correcting a region representation according to any one of claims 1 to 7 and/or the method of correcting a region representation according to any one of claims 8 to 10 via execution of the executable instructions.
14. A computer-readable storage medium on which a computer program is stored, the computer program, when being executed by a processor, implementing a method for correcting a region representation according to any one of claims 1 to 10.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011291786.9A CN113781082B (en) | 2020-11-18 | 2020-11-18 | Method and device for correcting regional portrait, electronic equipment and readable storage medium |
| PCT/CN2021/126483 WO2022105554A1 (en) | 2020-11-18 | 2021-10-26 | Region portrait correction method and apparatus, and electronic device and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011291786.9A CN113781082B (en) | 2020-11-18 | 2020-11-18 | Method and device for correcting regional portrait, electronic equipment and readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113781082A CN113781082A (en) | 2021-12-10 |
| CN113781082B true CN113781082B (en) | 2023-04-07 |
Family
ID=78835317
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011291786.9A Active CN113781082B (en) | 2020-11-18 | 2020-11-18 | Method and device for correcting regional portrait, electronic equipment and readable storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN113781082B (en) |
| WO (1) | WO2022105554A1 (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109902742A (en) * | 2019-02-28 | 2019-06-18 | 深圳前海微众银行股份有限公司 | Sample complementing method, terminal, system and medium based on encryption transfer learning |
| CN110210626A (en) * | 2019-05-31 | 2019-09-06 | 京东城市(北京)数字科技有限公司 | Data processing method, device and computer readable storage medium |
| KR20190106942A (en) * | 2019-08-30 | 2019-09-18 | 엘지전자 주식회사 | Artificial device and method for controlling the same |
| CN110797124A (en) * | 2019-10-30 | 2020-02-14 | 腾讯科技（深圳）有限公司 | Model multi-terminal collaborative training method, medical risk prediction method and device |
| CN111325267A (en) * | 2020-02-18 | 2020-06-23 | 京东城市(北京)数字科技有限公司 | Data fusion method, device and computer readable storage medium |
| CN111582508A (en) * | 2020-04-09 | 2020-08-25 | 上海淇毓信息科技有限公司 | Strategy making method and device based on federated learning framework and electronic equipment |
| CN111666576A (en) * | 2020-04-29 | 2020-09-15 | 平安科技（深圳）有限公司 | Data processing model generation method and device and data processing method and device |
| CN111860868A (en) * | 2020-07-27 | 2020-10-30 | 深圳前海微众银行股份有限公司 | Training sample construction method, device, equipment and computer readable storage medium |
| CN111949859A (en) * | 2019-05-16 | 2020-11-17 | Oppo广东移动通信有限公司 | User portrait updating method and device, computer equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3477527A1 (en) * | 2017-10-31 | 2019-05-01 | Twinpeek | Privacy management |
| CN111935156B (en) * | 2020-08-12 | 2022-06-14 | 科技谷（厦门）信息技术有限公司 | Data privacy protection method for federated learning |
-
2020
- 2020-11-18 CN CN202011291786.9A patent/CN113781082B/en active Active
-
2021
- 2021-10-26 WO PCT/CN2021/126483 patent/WO2022105554A1/en active Application Filing
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109902742A (en) * | 2019-02-28 | 2019-06-18 | 深圳前海微众银行股份有限公司 | Sample complementing method, terminal, system and medium based on encryption transfer learning |
| CN111949859A (en) * | 2019-05-16 | 2020-11-17 | Oppo广东移动通信有限公司 | User portrait updating method and device, computer equipment and storage medium |
| CN110210626A (en) * | 2019-05-31 | 2019-09-06 | 京东城市(北京)数字科技有限公司 | Data processing method, device and computer readable storage medium |
| KR20190106942A (en) * | 2019-08-30 | 2019-09-18 | 엘지전자 주식회사 | Artificial device and method for controlling the same |
| CN110797124A (en) * | 2019-10-30 | 2020-02-14 | 腾讯科技（深圳）有限公司 | Model multi-terminal collaborative training method, medical risk prediction method and device |
| CN111325267A (en) * | 2020-02-18 | 2020-06-23 | 京东城市(北京)数字科技有限公司 | Data fusion method, device and computer readable storage medium |
| CN111582508A (en) * | 2020-04-09 | 2020-08-25 | 上海淇毓信息科技有限公司 | Strategy making method and device based on federated learning framework and electronic equipment |
| CN111666576A (en) * | 2020-04-29 | 2020-09-15 | 平安科技（深圳）有限公司 | Data processing model generation method and device and data processing method and device |
| CN111860868A (en) * | 2020-07-27 | 2020-10-30 | 深圳前海微众银行股份有限公司 | Training sample construction method, device, equipment and computer readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113781082A (en) | 2021-12-10 |
| WO2022105554A1 (en) | 2022-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110245510B (en) | Method and apparatus for predicting information | |
| WO2022089256A1 (en) | Method, apparatus and device for training federated neural network model, and computer program product and computer-readable storage medium | |
| CN110383791B (en) | Map application crowdsourcing based on blockchain | |
| CN112085159B (en) | User tag data prediction system, method and device and electronic equipment | |
| US11693634B2 (en) | Building segment-specific executable program code for modeling outputs | |
| CN101467170A (en) | Apparatuses, methods and systems for electronic real estate transactions | |
| CN113505520A (en) | Method, device and system for supporting heterogeneous federated learning | |
| CN114547658B (en) | Data processing method, device, equipment and computer readable storage medium | |
| CN111932414A (en) | Training management system and method, computer storage medium and electronic equipment | |
| CN110909235A (en) | Enterprise innovation expert intelligence library platform based on big data | |
| CN112905187A (en) | Compiling method, compiling device, electronic equipment and storage medium | |
| CN115392718A (en) | Processing method, device, equipment and medium of process model | |
| CN110837657B (en) | Data processing method, client, server and storage medium | |
| CN113781082B (en) | Method and device for correcting regional portrait, electronic equipment and readable storage medium | |
| Dhakshan et al. | Role of Emerging Technologies in Education 4.0: Challenges and Future Research Directions | |
| US20220318887A1 (en) | Machine learning model generation platform | |
| US20210248515A1 (en) | Assisted learning with module privacy | |
| US11563576B2 (en) | Distributed anonymous scoring technique | |
| Grigorescu et al. | Strategic Focus For Changing Path to Digital Transformation In Romania–Case Study Gis Open E-Governance System Applications | |
| US8826166B2 (en) | Evaluating and comparing the requirements of a task with the capabilities of an entity | |
| Araujo et al. | Business and technical requirements of Software-as-a-Service: Implications in portuguese enterprise business context | |
| US11928572B2 (en) | Machine learning model generator | |
| US11869007B2 (en) | Geocoding geocode datasets in know your customer blockchain data blocks for spatial analytical risk modeling | |
| US20220318683A1 (en) | Machine learning model aggregator | |
| Khouibiri et al. | Analyzing the Influence of Cloud Business Intelligence on Small and Medium Enterprises A Case Study of Morocco |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |