Design method for multi-user collaborative modeling
By using recurrent neural network models to predict operational intent and eliminate conflicts in multi-user collaborative modeling, the common conflict problems in multi-user collaborative operation are resolved, and collaboration efficiency and model accuracy are improved.
Patent Information
- Application Number
- CN202311550285.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-11-20
- Publication Date
- 2025-05-20
AI Technical Summary
In multi-user collaborative modeling, operational conflicts are prone to occur during user coordination across geographical locations and model verification, and the existing technology has not yet provided an effective solution.
By identifying, collecting and converting user editing historical data, establishing and training recurrent neural network RNN models, predicting the intent tags of each operation, and eliminating the conflicts according to the predicted intent during operation conflicts, and determining the reserved operations.
It realizes collaborative editing of multiple users, solves the problem of conflict-prone problems of multi-user collaborative operation, improves collaboration efficiency, and ensures the accuracy of the model and real-time editing and coordination.
Smart Images
Figure CN120020810A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of collaborative modeling and simulation, and particularly to a design method for multi-user collaborative modeling. Background Art
[0002] In the fields of scientific research, engineering design, and innovation, multiple users often need to collaboratively participate in the process of model design and verification. However, multi-user collaborative modeling involves problems such as user coordination across geographical locations, model verification, and operation conflicts during simulation testing, but the existing technologies do not have methods for solving operation conflicts. Summary of the Invention
[0003] The present invention provides a design method for multi-user collaborative modeling, which can solve the technical problems in the existing technologies.
[0004] The present invention provides a design method for multi-user collaborative modeling, wherein the method includes:
[0005] Performing user identification according to user information, where the user information includes a user unique identity identifier;
[0006] Collecting user editing historical data, where the user editing historical data includes an operation sequence and an intention label for each operation;
[0007] Converting the user editing historical data;
[0008] Establishing a recurrent neural network (RNN) model;
[0009] Training the recurrent neural network model using the converted user editing historical data to minimize the prediction error of the intention label;
[0010] Receiving operations performed by multiple users;
[0011] Predicting the intention label for each operation using the trained recurrent neural network model;
[0012] In the case of operation conflicts, eliminating the conflicts according to the predicted intention labels to determine the operations to be retained.
[0013] Preferably, the method further includes:
[0014] After eliminating the conflicts and completing the multi-user collaborative modeling of the target model, publishing the established target model, assigning a model unique identity identifier to the published target model, and providing a locking function.
[0015] Preferably, the method further includes:
[0016] Receiving a user request for requesting to open the published target model;
[0017] Open the published target model according to the user request, and create a copy of the published target model at the same time.
[0018] Preferably, the recurrent neural network model includes an input layer, an RNN layer, and an output layer. The input layer receives the embedded representation of the operation sequence. The RNN layer uses long short-term memory network recurrent units to capture the sequence information of the operation sequence. The output layer outputs the predicted intent label corresponding to each operation.
[0019] Preferably, it is determined whether there is a conflict according to the conflict detection function.
[0020] The present invention also provides a design system for multi-user collaborative modeling, wherein the system includes:
[0021] An identification unit for performing user identification according to user information, where the user information includes a user unique identity identifier;
[0022] A collection unit for collecting user editing history data, where the user editing history data includes an operation sequence and the intent label of each operation;
[0023] A conversion unit for converting the user editing history data;
[0024] A model establishment unit for establishing a recurrent neural network RNN model;
[0025] A training unit for training the recurrent neural network model using the converted user editing history data to minimize the prediction error of the intent label;
[0026] A first receiving unit for receiving the operations performed by multiple users. For each operation, use the trained recurrent neural network model to predict the intent label of each operation;
[0027] A conflict elimination unit for eliminating conflicts according to the predicted intent label in the case of operation conflicts to determine the retained operations.
[0028] Preferably, the system further includes:
[0029] A publishing unit for publishing the established target model after eliminating conflicts and completing multi-user collaborative modeling of the target model, assigning a model unique identity identifier to the published target model, and providing a locking function.
[0030] Preferably, the system further includes:
[0031] A second receiving unit for receiving a user request, where the user request is used to request to open the published target model;
[0032] An opening unit for opening the published target model according to the user request, and creating a copy of the published target model at the same time.
[0033] Preferably, the recurrent neural network model includes an input layer, an RNN layer, and an output layer. The input layer receives the embedded representation of the operation sequence. The RNN layer uses long short-term memory network recurrent units to capture the sequential information of the operation sequence. The output layer outputs the predicted intent label corresponding to each operation.
[0034] Preferably, it is determined whether there is a conflict according to the conflict detection function.
[0035] Through the above technical solution, the recurrent neural network model can be trained according to the user's editing historical data, and the trained recurrent neural network model can be used to predict the intent label of each operation. In the case of conflicts in operations, the conflicts can be eliminated according to the predicted intent label to determine the operations to be retained. Thus, multi-user collaborative editing can be achieved, the problem of easy conflicts in multi-user collaborative operations is solved, and the collaboration efficiency is provided. Description of the Drawings
[0036] The included drawings are used to provide a further understanding of the embodiments of the present invention. They form a part of the specification, are used to illustrate the embodiments of the present invention, and are used to explain the principles of the present invention together with the written description. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0037] Figure 1 A flowchart showing a design method for multi-user collaborative modeling according to an embodiment of the present invention is shown. Detailed Embodiments
[0038] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and in no way limits the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.
[0039] It should be noted that the terms used herein are for the purpose of describing specific embodiments only and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular forms are also intended to include the plural forms. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they specify the presence of features, steps, operations, devices, components, and / or combinations thereof.
[0040] Unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions, and values set forth in these embodiments do not limit the scope of the present invention. At the same time, it should be understood that, for the sake of convenience of description, the dimensions of the various parts shown in the drawings are not drawn in actual proportional relationships. Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the said technologies, methods, and devices should be regarded as part of the authorized specification. In all the examples shown and discussed herein, any specific value should be construed as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values. It should be noted that like reference numerals and letters denote like items in the following drawings, and thus, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.
[0041] Figure 1 The flowchart of a design method for multi-user collaborative modeling according to an embodiment of the present invention is shown.
[0042] As Figure 1 shown, an embodiment of the present invention provides a design method for multi-user collaborative modeling, wherein the method includes:
[0043] S100, perform user identification according to user information, where the user information includes a user unique identity identifier;
[0044] For example, a unique identity identifier can be assigned to a user through an identity authentication service. By identifying the unique identity identifier of each user, a user storage space is allocated to it (this space is used to store models, data, and other information uploaded by the user, as well as to complete the model synchronization function after collaborative editing with other users), which ensures that the data and model information of each user are properly managed and avoids data conflicts and confusion.
[0045] S102, collect user editing history data, where the user editing history data includes an operation sequence and an intention tag for each operation;
[0046] Among them, each operation includes a position, content, and operation type.
[0047] S104, perform conversion on the user editing history data;
[0048] That is, convert the operation sequence and intent label into a format suitable for the input of the RNN. For positions and characters, an embedding layer can be used to convert them into vector representations.
[0049] S106, establish a Recurrent Neural Network (RNN) model;
[0050] S108, use the converted user editing history data to train the Recurrent Neural Network model to minimize the prediction error of the intent label;
[0051] S110, receive operations executed by multiple users;
[0052] S112, use the trained Recurrent Neural Network model to predict the intent label of each operation;
[0053] Among them, the accuracy of the prediction can be measured by selecting an appropriate loss function (such as cross-entropy).
[0054] S114, in the case of operation conflicts, eliminate the conflicts according to the predicted intent label to determine the operations to be retained.
[0055] That is, infer the user intent according to the trained Recurrent Neural Network model, predict operation conflicts, and perform operation conversion according to the prediction results.
[0056] For example, for conflicting operations, eliminate the conflicts according to the predicted intent label, and the operations can be selected based on the relative importance of the user intent. Alternatively, if the model predicts that the intent of one operation is stronger than that of another operation, the operation with the stronger predicted intent can be selected.
[0057] Through the above technical solution, the Recurrent Neural Network model can be trained according to the user editing history data, and the trained Recurrent Neural Network model can be used to predict the intent label of each operation. In the case of operation conflicts, eliminate the conflicts according to the predicted intent label to determine the operations to be retained. Thus, multi-user collaborative editing can be achieved, the problem of easy conflicts in multi-user collaborative operations is solved, and the collaboration efficiency is provided.
[0058] That is, real-time editing and coordination can be achieved, enabling multiple users to edit and adjust the model simultaneously. Even in the case of operation conflicts, they can be eliminated to improve the accuracy of the model. For operations without conflicts, the operations can be directly executed.
[0059] According to an embodiment of the present invention, the method further includes:
[0060] After eliminating conflicts and completing the multi-user collaborative modeling of the target model, publish the established target model, assign a unique model identity to the published target model, and provide a locking function.
[0061] Thus, the target model can be published for other users to view. By assigning a unique identity to the model and providing a locking function, unauthorized editing can be prevented.
[0062] According to an embodiment of the present invention, the method further includes:
[0063] Receiving a user request for requesting to open the published target model;
[0064] Opening the published target model according to the user request, and simultaneously creating a copy of the published target model.
[0065] Thus, the user can open the model and perform simulation verification (simulation test). Moreover, multiple users can view the simulation process and results simultaneously, accelerating the process of model verification. By creating a copy of the target model, conflicts between multiple users can be avoided: multiple users can view the simulation effect of the model at the same time (the simulation interface will display the dynamic effect of the model to show the running situation of the model).
[0066] In addition, real-time collaboration functions can be achieved between users through real-time communication protocols such as WebSocket, allowing multiple users to have real-time discussions and exchanges in the simulation interface, enabling users to instantly share ideas, ask questions, and conduct real-time communication. Users can post comments, chat, and share insights and feedback.
[0067] For example, a web-based model editor can be used, allowing users to collaboratively edit the model through the Internet or local area network and complete the simulation to verify the functions and performance of the model.
[0068] According to an embodiment of the present invention, the recurrent neural network model includes an input layer, an RNN layer, and an output layer. The input layer receives the embedded representation of the operation sequence. The RNN layer uses long short-term memory (LSTM) recurrent units to capture the sequence information of the operation sequence. The output layer outputs the predicted intent label corresponding to each operation.
[0069] According to an embodiment of the present invention, it is determined whether there is a conflict according to the conflict detection function.
[0070] The conflict elimination process in the present invention is described below with reference to an example.
[0071] In this example, each operation is represented by a position, a character, and an operation type. An RNN is used to predict the intent of the operation, and then the operation to be retained is determined according to the prediction result. For the sake of simplicity of the example, it is assumed that there are only two intent labels: INSERT (insert) and DELETE (delete). Specifically, the conflict elimination process is as follows:
[0072] 1. Symbol Definition:
[0073] A = (pos A , content A ): Operation A, where pos ) represents the position and content A represents the content.
[0074] B = (pos B , content B ): Operation B, where pos B represents the position and content B represents the content.
[0075] 2. Algorithm Flow:
[0076] a. Initial State: The document is empty.
[0077] b. User Editing: The user performs Operation A (insert / delete): A = (pos(A), content(A)). The user performs Operation B (insert / delete): B = (pos(B), content(B)).
[0078] The RNN model predicts the intention of Operation A, where represents the probability that Operation A is an insert operation, represents the probability that Operation A is a delete operation.
[0079] The RNN model predicts the intention of Operation B, where represents the probability that Operation B is an insert operation, represents the probability that Operation B is a delete operation.
[0080] c. Conflict Detection: If Operations A and B conflict at the same position: pos(A) = pos(B).
[0081] C(A, B): Conflict Detection Function, where C(A, B) = 1 indicates that Operations A and B conflict, and C(A, B) = 0 indicates no conflict.
[0082] T(A, B): Operation Transformation Function, which processes the results of Operations A and B according to the operation transformation rules.
[0083] d. Operation Transformation:
[0084] If C(A, B) = 1, that is, Operations A and B conflict:
[0085] If and that is, both Operations A and B are insert operations:
[0086] If time_A < time_B, the merge operation is T(A, B) = (pso A , content A + content B )
[0087] Otherwise, the merge operation is T(A, B) = (pos B , content B + content A ).
[0088] If and That is, operation A is an insertion and operation B is a deletion:
[0089] The merge operation is T(A, B) = A, retaining operation A.
[0090] If and That is, operation A is a deletion and operation B is an insertion:
[0091] The merge operation is T(A, B) = B, retaining operation B.
[0092] If and That is, both operation A and operation B are deletion operations:
[0093] The merge operation is T(A, B), selecting the operation with the longer content,
[0094] That is, T(A, B) = argmax(len(content A ), len(content B ))
[0095] If C(A, B) = 0, that is, operation A and operation B have no conflict:
[0096] The merge operation is T(A, B) = A, retaining operation A.
[0097] Finally, the merged operation is merged_operation = T(A, B)
[0098] The embodiment of the present invention also provides a design system for multi - user collaborative modeling, wherein the system includes:
[0099] An identification unit for identifying users according to user information, where the user information includes user unique identity identifiers;
[0100] A collection unit for collecting user editing historical data, where the user editing historical data includes an operation sequence and an intent label for each operation;
[0101] A conversion unit for converting the user editing historical data;
[0102] A model building unit for building a recurrent neural network model RNN;
[0103] A training unit for training the recurrent neural network model using the converted user editing historical data to minimize the prediction error of the intent label;
[0104] A first receiving unit for receiving operations performed by multiple users, and for each operation, predicting the intent label of each operation using the trained recurrent neural network model;
[0105] A conflict elimination unit for eliminating conflicts based on the predicted intent labels in the case of conflicting operations to determine the operations to be retained.
[0106] Through the above technical solution, a recurrent neural network model can be trained based on user editing historical data, and the trained recurrent neural network model can be used to predict the intent label of each operation. In the case of conflicting operations, conflicts are eliminated based on the predicted intent labels to determine the operations to be retained. Thus, multi-user collaborative editing can be achieved, the problem of easy conflicts in multi-user collaborative operations is solved, and the collaboration efficiency is provided.
[0107] According to an embodiment of the present invention, the system further includes:
[0108] A publishing unit for publishing the established target model after conflict elimination and completion of multi-user collaborative modeling of the target model, assigning a unique model identity to the published target model, and providing a locking function.
[0109] According to an embodiment of the present invention, the system further includes:
[0110] A second receiving unit for receiving a user request for requesting to open the published target model;
[0111] An opening unit for opening the published target model according to the user request and simultaneously creating a copy of the published target model.
[0112] According to an embodiment of the present invention, the recurrent neural network model includes an input layer, an RNN layer, and an output layer. The input layer receives the embedded representation of the operation sequence, the RNN layer uses long short-term memory network recurrent units to capture the sequence information of the operation sequence, and the output layer outputs the predicted intent label corresponding to each operation.
[0113] According to an embodiment of the present invention, it is determined whether there is a conflict according to a conflict detection function.
[0114] The above system and Figure 1 the above method correspond to each other. For specific examples, reference may be made to the description of the above method for Figure 1 , which will not be elaborated here.
[0115] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by orientation words such as "front, back, up, down, left, right", "lateral, vertical, perpendicular, horizontal" and "top, bottom", etc. is usually based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description. Without contrary instructions, these orientation words do not indicate and imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation on the protection scope of the present invention; the orientation words "inside, outside" refer to the inside and outside relative to the contour of each component itself.
[0116] For the convenience of description, spatial relative terms such as "above...", "over...", "on the upper surface of...", "above" can be used here to describe the spatial positional relationship between a device or feature shown in the figure and other devices or features. It should be understood that the spatial relative terms are intended to include different orientations in use or operation in addition to the orientation described in the figure for the device. For example, if the device in the drawing is inverted, the device described as "above other devices or structures" or "over other devices or structures" will then be positioned "below other devices or structures" or "under other devices or structures". Thus, the exemplary term "above..." can include both the orientations of "above..." and "below...". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and corresponding interpretations should be made for the spatial relative descriptions used here.
[0117] In addition, it should be noted that the use of words such as "first", "second" to limit components is only for the convenience of distinguishing the corresponding components. Without additional statements, the above words have no special meanings. Therefore, it should not be construed as a limitation on the protection scope of the present invention.
[0118] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A design method for multi-user collaborative modeling, characterized in that: The method includes: Perform user identification based on user information, wherein the user information includes a unique user identity; Collect user editing history data, which includes operation sequences and the intention label of each operation; Convert user edit history data; Establish a recurrent neural network RNN model; Use the converted user editing history data to train a recurrent neural network model to minimize the prediction error of intent labels; Receive operations performed by multiple users; Use the trained recurrent neural network model to predict the intent label of each operation; In the case of conflicting actions, the conflicts are resolved based on the predicted intent labels to determine the actions to retain.
2. The method according to claim 1, characterized in that: The method further includes: After eliminating conflicts and completing multi-user collaborative modeling of the target model, the established target model is published, a unique model identity is assigned to the published target model, and a locking function is provided.
3. The method according to claim 2, characterized in that The method further includes: receiving a user request, where the user request is used to request to open a published target model; Opens a published target model upon user request and creates a copy of the published target model.
4. The method according to claim 3, characterized in that: The recurrent neural network model includes an input layer, an RNN layer, and an output layer. The input layer receives the embedded representation of the operation sequence, the RNN layer uses a long short-term memory network recurrent unit to capture the sequence information of the operation sequence, and the output layer outputs the predicted intent label corresponding to each operation.
5. The method according to claim 4, characterized in that Determine whether there is a conflict based on the conflict detection function.
6. A multi-user collaborative modeling design system, characterized in that: The system includes: An identification unit, used to identify a user based on user information, wherein the user information includes a unique identification of the user; A collection unit, used to collect user editing history data, where the user editing history data includes an operation sequence and an intention label of each operation; A conversion unit, used for converting the user editing history data; A model building unit, used to build a recurrent neural network RNN model; A training unit, for training a recurrent neural network model using the converted user editing history data to minimize the prediction error of the intent label; A first receiving unit is used to receive operations performed by multiple users, and for each operation, use the trained recurrent neural network model to predict the intention label of each operation; The conflict elimination unit is used to eliminate the conflict according to the predicted intention label when there is a conflict in the operation, so as to determine the retained operation.
7. The system according to claim 6, characterized in that The system also includes: The publishing unit is used to publish the established target model after eliminating conflicts and completing multi-user collaborative modeling of the target model, assign a unique model identity to the published target model, and provide a locking function.
8. The system according to claim 7, characterized in that The system also includes: A second receiving unit is used to receive a user request, where the user request is used to request to open the published target model; The opening unit is used to open the published target model according to the user's request and create a copy of the published target model.
9. The system according to claim 8, characterized in that The recurrent neural network model includes an input layer, an RNN layer, and an output layer. The input layer receives the embedded representation of the operation sequence, the RNN layer uses a long short-term memory network recurrent unit to capture the sequence information of the operation sequence, and the output layer outputs the predicted intent label corresponding to each operation.
10. The system according to claim 9, characterized in that Determine whether there is a conflict based on the conflict detection function.