Multi-device synchronized mobile office collaboration methods, systems, and computer devices

Abstract

This application provides a multi-device synchronized mobile office collaboration method, system, and computer device. The method includes: acquiring and analyzing the path hierarchy name, character count before and after text modification, and editing operation type of each field to generate a structure weight mapping sequence; analyzing the change magnitude of the target document before and after editing based on the structure weight mapping sequence, determining the quantified value of the change magnitude, and comparing the quantified value of the change magnitude with a change magnitude threshold to determine whether to initiate synchronization filtering; if synchronization filtering is initiated, filtering out the synchronization target fields of the target document, performing synchronization trigger analysis on the input events of each synchronization target field of the target document within the current preset time period, analyzing the input events that meet the synchronization trigger criteria, determining the synchronization control magnitude value of the current synchronization target field to be controlled, and determining the synchronization mode configuration to adjust the current synchronization execution mode of the synchronization target field to be controlled. This improves the accuracy of data synchronization.

Description

Technical Field

[0001] This application relates to the field of data synchronization technology, and in particular to mobile office collaboration methods, systems and computer devices that enable multi-device synchronization. Background Technology

[0002] The field of data synchronization technology encompasses methods and systems for maintaining data consistency across multiple terminal devices. It involves key technologies such as data change capture, transmission, merging, and conflict resolution. Focusing on inconsistencies arising from data transmission across multiple nodes in a distributed environment, it ensures data consistency across different devices in time and space through mechanisms such as incremental transmission, state tracking, and version management. This includes client-side local storage management, data synchronization protocol design, data change capture mechanisms, conflict identification and handling methods, synchronization triggering mechanisms, and cross-network environment synchronization adaptation methods. Applied to scenarios such as mobile office, remote collaboration, and smart terminal interconnection, it features complex data state management and terminal behavior scheduling characteristics. Among these, multi-terminal synchronization in mobile... An office collaboration system refers to a system architecture built on network communication mechanisms that supports real-time synchronization of office content such as documents, tasks, and messages among multiple terminal devices. Addressing the data consistency requirements in cross-platform office environments, it tackles issues such as data change synchronization, state conflict handling, offline access, and network recovery synchronization arising from multiple device accesses. It employs a version vector mechanism for change tracking, utilizes a content-based incremental synchronization algorithm to transmit differing data, combines preset conflict judgment rules for data merging, provides offline availability through a local data persistence structure, and uses a push-triggered mechanism to notify users of state changes. An adaptation layer is used to handle platform differences for different terminal systems, ensuring stable data collaboration capabilities across various mobile devices under varying network conditions.

[0003] Traditional multi-terminal synchronized mobile office collaboration technologies focus on version number management and change status tracking during data synchronization. Under conditions of dense terminal input events and frequent cross-operations, they lack a dynamic scheduling mechanism for characterizing the density of editing behaviors and the triggering rhythm, which leads to the risk of false triggering or delay in synchronization response, affecting the accuracy of data synchronization. Summary of the Invention

[0004] The purpose of this application is to address the shortcomings of existing technologies by proposing a multi-terminal synchronized mobile office collaboration method, system, and computer device.

[0005] To achieve the above objectives, this application adopts the following technical solution: a multi-terminal synchronized mobile office collaboration method, wherein the multi-terminal synchronized mobile office collaboration method includes:

[0006] Analyze the parameter set of field editing behavior of the target document to obtain the path level name, the number of characters before and after text modification, and the type of editing operation for each field;

[0007] Based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation, a structural weight mapping sequence is generated.

[0008] The changes in the target document before and after editing are analyzed based on the structural weight mapping sequence. The quantitative value of the changes is determined, and the quantitative value of the changes is compared with the change threshold to determine whether to start synchronous filtering.

[0009] When synchronous filtering is initiated, the synchronous target fields of the target document are filtered out to form a synchronous target field set, and a field-level change labeling sequence is generated;

[0010] Based on the field-level change calibration sequence, the input events of each synchronous target field of the target document within the current preset time period are analyzed for synchronization triggering. Each synchronous target field is determined to meet the input events for synchronization triggering, and a corresponding synchronization triggering sequence is generated independently for each synchronous target field.

[0011] Analyze the continuous triggering time in the synchronization trigger sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled;

[0012] Based on the synchronization control amplitude value of the target field to be controlled, determine the synchronization mode configuration, and adjust the current synchronization execution mode of the target field to be controlled according to the synchronization mode configuration.

[0013] In one embodiment, the multi-terminal synchronized mobile office collaboration method further includes:

[0014] Obtain the operation user identifier and field number of the current input event to be synchronized, and determine the correspondence between them and the initial operation user identifier of the field number in the belonging chain sequence value to determine whether the operation user of the current input event to be synchronized and the initial operation user are the same user.

[0015] If the user operating the current input event to be synchronized for the field number is not the same user as the initial user, the field number will be added to the set of inconsistent field numbers.

[0016] Based on the set of inconsistent attribution field numbers, analyze the inconsistent attribution fields in the density analysis window of the target document to determine the distribution density in each density analysis window;

[0017] Fields within the density analysis window whose distribution density is greater than the density threshold are classified as fields belonging to the risk zone, forming a set of field numbers belonging to the risk zone.

[0018] Analyze the record table of input events to be synchronized on the fields of the risk zone in the set of field numbers of the risk zone to determine the operation type and character modification range of each input event to be synchronized in each risk zone field.

[0019] Grouping each input event to be synchronized in a risk zone field into a group, sorting the input events to be synchronized in each group according to the timestamp of arrival at the server, and obtaining the grouping sequence of each field;

[0020] Based on the operation type and character modification range of each pair of adjacent input events to be synchronized in the grouping sequence of each field, analyze the type difference score and overlap, and determine the semantic difference of adjacent input events to be synchronized in the grouping sequence of each field.

[0021] Based on the semantic differences between adjacent input events to be synchronized in each field grouping sequence, conflicting intent fields are identified, and the synchronization of input events for the conflicting intent fields is paused.

[0022] In one embodiment, the multi-terminal synchronized mobile office collaboration method further includes:

[0023] Adjacent input events to be synchronized that have a semantic difference exceeding the conflict confirmation threshold in the intent conflict field are recorded in the intent conflict matching table of the intent conflict field. The intent conflict matching table records the operation user identifiers, operation details, and semantic difference between the two conflicting input events.

[0024] The conflict details of adjacent operations whose semantic differences exceed the conflict confirmation threshold are pushed to the interface of the relevant user corresponding to the input event that caused the conflict, requesting the relevant user to confirm the version selection or manually merge the content.

[0025] In one embodiment, the step of analyzing the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation to generate a structural weight mapping sequence includes:

[0026] Based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation, the path hierarchy name is split layer by layer and the path hierarchy value is analyzed. The number of characters before and after text modification is analyzed to determine the text modification length. The types of editing operations are grouped and classified according to the editing intent to determine the operation intensity level.

[0027] Normalization matching is performed on the path level value, text modification length, and operation intensity level of each field to obtain the normalized path level value, normalized text modification length value, and normalized operation intensity level value of each field. The normalized path level value, normalized text modification length value, and normalized operation intensity level value of each field are then combined into a vector to obtain the field normalization parameter set.

[0028] Based on the path level normalization value, text modification length normalization value, and operation intensity level normalization value of each field in the field normalization parameter set, the structural weight value of each field is analyzed using the structural weight value analysis formula to generate a structural weight mapping sequence.

[0029] In one embodiment, the step of performing synchronization trigger analysis on the input events of each synchronization target field of the target document within the current preset time period based on the field-level change calibration sequence, determining that each synchronization target field meets the input events for synchronization triggering, and independently generating a corresponding synchronization trigger sequence for each synchronization target field includes:

[0030] Based on the field-level change calibration sequence, obtain the timestamp and character modification count of the input events of each synchronized target field of the target document within the current preset time period, sort the input events of each synchronized target field in chronological order, and obtain the time sequence value of each synchronized target field.

[0031] Based on the time sequence value of each synchronization target field, the time interval and number of character changes between each pair of adjacent input events are calculated, and the ratio of the number of character changes to the time interval for each pair of adjacent input events is calculated to obtain the editing frequency of each pair of adjacent input events for each synchronization target field.

[0032] By using a trend analysis sliding window, the editing frequency of each pair of adjacent input events in the N most recent input events of each synchronization target field is obtained and analyzed to determine the input distribution trend coefficient of each synchronization target field within the current preset time period;

[0033] A synchronous analysis sliding window is used to obtain the time interval of the nearest Q-second adjacent input events for each synchronous target field and analyze it to determine the average time interval of each synchronous target field within the current preset time period.

[0034] A synchronous analysis sliding window is used to obtain the number of character changes in each synchronous target field within the most recent Q seconds for analysis, and to determine the total number of character changes in each synchronous target field within the current preset time period;

[0035] The average time interval, total number of character changes, and input distribution trend coefficient of each synchronization target field within the current preset time period are compared with the synchronization triggering standard to determine whether the input events of each synchronization target field within the current preset time period meet the synchronization triggering standard.

[0036] If there are input events that meet the synchronization triggering criteria within the current preset time period, record the triggering time of the input events that meet the synchronization triggering criteria in the synchronization triggering sequence of the corresponding synchronization target field.

[0037] In one embodiment, analyzing the continuous trigger times in the synchronization trigger sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled includes:

[0038] Based on the analysis of the synchronization trigger sequence of each synchronization target field, the time interval between adjacent trigger times in the synchronization trigger sequence of each synchronization target field is determined;

[0039] The synchronization triggering change trend of each synchronization target field is determined by analyzing the time interval between adjacent triggering times in the synchronization triggering sequence of each synchronization target field.

[0040] Analyze the changes in trigger frequency for each synchronization target field to determine the amount of change in trigger frequency for each synchronization target field;

[0041] The cumulative offset direction of the frequency change of each synchronization target field is determined by analyzing the change in the trigger frequency of each synchronization target field.

[0042] The synchronous triggering change trend is the cumulative offset direction of the triggering rhythm first speeding up and then slowing down or the frequency change is positive. The synchronous target field is the synchronous target field to be adjusted.

[0043] The synchronization control amplitude of the target field to be controlled is analyzed using the synchronization control amplitude calculation formula to determine the current synchronization control amplitude value of the target field to be controlled.

[0044] In one embodiment, the step of analyzing the semantic differences between adjacent input events to be synchronized in each field grouping sequence, determining the intention conflict field, and pausing the synchronization of input events for the intention conflict field includes:

[0045] Adjacent input events to be synchronized that have a semantic difference exceeding the difference threshold are placed in the conflict processing queue. The influence range of adjacent input events to be synchronized in the conflict processing queue is judged by the logical order of the order of events to be synchronized to determine the dependency and coverage relationship between operations. The processing order is determined according to the dependency and coverage relationship, and the result of the priority judgment is output to the queue.

[0046] Based on the queue priority judgment result, adjacent input events to be synchronized in the conflict processing queue that have a semantic difference exceeding the conflict confirmation threshold are judged according to priority order, and the field to which the adjacent input event to be synchronized is determined as the intention conflict field.

[0047] The conflicting intent field is added to the set of field numbers to be separated, and the synchronization of fields in the set of field numbers to be separated is paused.

[0048] A multi-terminal synchronized mobile office collaboration system, comprising: a behavior parameter analysis module, a weight analysis module, a change amplitude analysis module, a synchronization filtering module, a synchronization trigger analysis module, a synchronization control amplitude analysis module, and a synchronization control module;

[0049] The behavior parameter analysis module is used to analyze the field editing behavior parameter set of the target document to obtain the path level name, the number of characters before and after text modification, and the type of editing operation for each field.

[0050] The weight analysis module is used to analyze each field based on its path hierarchy name, the number of characters before and after text modification, and the type of editing operation, and generate a structural weight mapping sequence.

[0051] The change magnitude analysis module is used to analyze the change magnitude of the target document before and after editing based on the structural weight mapping sequence, determine the quantitative value of the change magnitude, and compare the quantitative value of the change magnitude with the change magnitude threshold to determine whether to start synchronous filtering.

[0052] The synchronous filtering module is used to filter out the synchronous target fields of the target document to form a synchronous target field set when synchronous filtering is started, and generate a field-level change labeling sequence.

[0053] The synchronization trigger analysis module is used to perform synchronization trigger analysis on the input events of each synchronization target field of the target document within the current preset time period according to the field-level change calibration sequence, determine the input events that meet the synchronization trigger for each synchronization target field, and generate a corresponding synchronization trigger sequence for each synchronization target field independently.

[0054] The synchronization control amplitude analysis module is used to analyze the continuous triggering time in the synchronization triggering sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled.

[0055] The synchronization control module is used to determine the synchronization mode configuration based on the synchronization control amplitude value of the synchronization target field to be controlled, and adjust the current synchronization execution mode of the synchronization target field to be controlled according to the synchronization mode configuration.

[0056] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described multi-terminal synchronized mobile office collaboration method.

[0057] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described multi-terminal synchronized mobile office collaboration method.

[0058] Compared with the prior art, the advantages and positive effects of this application are as follows:

[0059] In this application, by analyzing the field editing behavior parameter set of the target document, the path hierarchy name, the number of characters before and after text modification, and the editing operation type of each field are obtained. Based on the analysis of the path hierarchy name, the number of characters before and after text modification, and the editing operation type of each field, a structural weight mapping sequence is generated. The change range of the target document before and after editing is analyzed based on the structural weight mapping sequence to determine the quantification value of the change range. The quantification value of the change range is compared with a change range threshold to determine whether to initiate synchronous filtering. If synchronous filtering is initiated, the synchronous target fields of the target document are selected to form a synchronous target field set, and a field-level... The system employs a change calibration sequence. Based on this sequence, it performs synchronization trigger analysis on the input events of each synchronized target field in the target document within the current preset time period. It determines the input events that meet the synchronization trigger criteria for each synchronized target field and generates a corresponding synchronization trigger sequence for each. The system analyzes the continuous trigger times within the synchronization trigger sequence for each synchronized target field to determine the synchronization control amplitude value for the target field to be controlled. Based on the synchronization control amplitude value of the target field to be controlled, it determines the synchronization mode configuration and adjusts the current synchronization execution mode of the target field to be controlled according to the synchronization mode configuration. Therefore, under conditions of dense terminal input events and frequent cross-operations, the dynamic scheduling mechanism for characterizing the density of editing behaviors and the trigger rhythm avoids the risk of false triggering or delay in synchronization response, improving the accuracy of data synchronization. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating a multi-device synchronized mobile office collaboration method in one embodiment of this application.

[0061] Figure 2 This is a schematic diagram of the structure of a multi-device synchronized mobile office collaboration system in one embodiment. Detailed Implementation

[0062] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0063] In the description of this application, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, in the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0064] In one embodiment, such as Figure 1 As shown, a multi-device synchronized mobile office collaboration method is provided, including the following steps:

[0065] Step S220: Analyze the field editing behavior parameter set of the target document to obtain the path level name, the number of characters before and after text modification, and the editing operation type for each field.

[0066] The field editing behavior parameter set is the raw editing data stream captured by the system in real time, which specifically includes user ID, timestamp, operation instructions, target field path, and modified content, etc.

[0067] Step S240: Analyze the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation to generate a structure weight mapping sequence.

[0068] In one embodiment, an analysis is performed based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation to generate a structural weight mapping sequence, including:

[0069] Based on the path hierarchy name, the number of characters before and after text modification, and the type of editing operation for each field, the path hierarchy name is split layer by layer and the path hierarchy value is analyzed. The number of characters before and after text modification is analyzed to determine the text modification length. The type of editing operation is grouped and categorized according to the editing intent to determine the operation intensity level. The path hierarchy value, text modification length, and operation intensity level of each field are normalized and matched to obtain the normalized values ​​of the path hierarchy, text modification length, and operation intensity level for each field. These values ​​are then combined into a vector to obtain the field normalization parameter set. Based on the normalized values ​​of the path hierarchy, text modification length, and operation intensity level for each field in the field normalization parameter set, the structural weight value of each field is analyzed using the structural weight value analysis formula to generate a structural weight mapping sequence.

[0070] In one example, taking a collaboratively edited "Project Market Analysis Report" document as an example, when user A performs a modification operation on the document title "2025 Q3 Market Strategy", the first field of the field editing behavior parameter set record is {UserID:“UserA”,Timestamp:1695692821000,Action:“Update”,Path:“doc / title / text”,Content:“2025 Third Quarter Market Core Strategy”}. First, the path layer in the first field is... The level name "doc / title / text" is split layer by layer to obtain a level array ["doc", "title", "text"]. This array is then quantified according to a preset structural importance rule, where "doc" is set to 1, "title" to 2, and "text" to 3. A weighted summation is used to calculate the path level value: 1 × 0.2 + 2 × 0.5 + 3 × 0.3 = 2.1. Next, the length of the modified text in the first field is counted. The original text length was 11 characters, and the modified text length is 15 characters. The absolute value of the difference in character count before and after text modification is calculated, resulting in a text modification length of 4. Then, the editing operation type "Update" in the first field is grouped and categorized according to editing intent. The editing operation types are pre-defined as five levels: {Create: 1, Insert: 2, Delete: 3, Modify: 4, Format: 5}. Therefore, the current editing operation type "Modify" is categorized as operation intensity level 4. Finally, these three items—path level value 2.1, text modification length 4, and operation intensity level 4—are subjected to a normalization matching process. This normalization matching process maps the values ​​of each dimension... Within the interval [0,1], the processing standard is based on statistical analysis of editing behavior of 100,000 documents in the document library. The maximum observed value of the path level is 10, the maximum value of the text modification length is 500 characters, and the maximum value of the operation intensity level is 5. Therefore, the normalized value of the path level is 2.1 / 10=0.21, the normalized value of the text modification length is 4 / 500=0.008, and the normalized value of the operation intensity level is 4 / 5=0.8. These three normalized values ​​are combined into a vector [0.21,0.008,0.8] in the field normalization parameter set to obtain the field normalization parameter set.

[0071] In one embodiment, the formula for analyzing structural weight values ​​is:

[0072] ;

[0073] in, For the first The structural weight values ​​of each field, For the first The normalized matching adjustment coefficient is obtained by adjusting the first... The path hierarchy normalized value, text modification length normalized value, and operation intensity level normalized value of each field are obtained by setting proportional weights. For the first The normalized parameter at the th The mapping results in each field are mapped by mapping the path level normalized value, text modification length normalized value, or operation intensity level normalized value to the first field. The fields are obtained. For the first The normalized text modification length value for each field is obtained by dividing the text modification length by the maximum modification length in the document. For the first The normalized value of the operation intensity level for each field is obtained by dividing the operation intensity level by the preset maximum operation intensity level. The mean of the normalized values ​​of the operation intensity levels for all fields is obtained by applying the normalized values ​​of all fields. We obtain the arithmetic mean. This is a normalized parameter index, indicating the position of the normalized parameter in the field's normalized parameter set. For field indexing, it represents the first [number]th [item] in the task document. One field, This represents the total number of normalized parameters.

[0074] In one example, taking the collaboratively edited "Project Market Analysis Report" document mentioned above, based on the vector [0.21, 0.008, 0.8] corresponding to the first field in the field normalization parameter set, according to the structural weight value analysis formula... Calculate the structural weight value of the first field; in this structural weight value analysis formula, taking the title field as an example, its index... , The total number of normalized parameters, here These correspond to the normalized values ​​for path level, text modification length, and operation intensity level, respectively. The normalized matching adjustment factor is determined through regression analysis of typical editing sessions of over five thousand users. This adjustment aims to balance the impact of different editing behaviors on structural importance. The analysis results show that path hierarchy has the largest impact on structural importance, followed by editing operation intensity, while text modification length has the smallest impact. Based on this, the weights of the normalized values ​​for path hierarchy, text modification length, and operation intensity level are set as follows: , and , For the first The normalized parameter at the th The mapping result in each field, that is, the value in the field normalization parameter set, therefore , , , , , This is the normalized mean of the operation intensity level values ​​for all modified fields in the current document. Assume the current document has a paragraph field in addition to the title. The operation intensity level has been modified, and its normalized value is [value missing]. ,but The calculation logic of the structural weight value analysis formula lies in the fact that the numerator part is calculated by weighted summation. It integrates the combined effects of field position, operation intent, and modified length, and uses the square root item. The weights of the intermediate-length modification were non-linearly amplified, and the denominator was... As a moderating factor, when the operational intensity of a certain field deviates significantly from the average operational intensity of the current document, the denominator increases, thereby suppressing the structural weight value of that field. The advantage of the structural weight value analysis formula is that by quantifying and integrating multi-dimensional and non-linear editing behavior characteristics and introducing mean dispersion as a dynamic adjustment, the calculation of the structural weight value not only reflects the static attributes of a single edit but also incorporates its dynamic context in the current editing session. Now, let's input the numerical values ​​for calculation:

[0075] ;

[0076] ;

[0077] ;

[0078] ;

[0079] In this context, the structural weight value refers to the dimensionless numerical value assigned to each field in a multi-device synchronized task document during synchronization calculation. It measures the relative importance of that field in the current editing state. Its meaning lies in comprehensively considering the field's hierarchical position, the extent of content modification, and the intensity of operational actions. Through normalization and non-linear combination under a unified scale, this structural weight value allows for comparison between different fields. Its effect is to provide a quantitative basis for subsequent synchronization scheduling and conflict handling. Fields with larger structural weight values ​​are preferentially included in the target set during synchronization, while fields with smaller structural weight values ​​are given secondary importance, thus establishing a quantifiable field priority order. This calculation result... This refers to the structural weight value of the title field. This structural weight value is a dimensionless number, and its magnitude (between 0 and 1) directly reflects the relative importance of the field among all current changes. Subsequent sorting and filtering will be based on this structural weight value. A structural weight value between 0.3 and 0.6 is defined as a medium-importance change. Finally, the structural weight values ​​of all fields are collected to generate a structural weight mapping sequence.

[0080] Step S260: Analyze the changes in the target document before and after editing based on the structural weight mapping sequence, determine the quantification value of the changes, and compare the quantification value of the changes with the change threshold to determine whether to start synchronous filtering.

[0081] In one example, taking the collaboratively edited "Project Market Analysis Report" document mentioned above, assume that the structure weight mapping sequence includes the structure weight values ​​of the header field (i.e., the first field). Weight of the paragraph field (i.e., the second field) The sequence [0.3802, 0.1534] is used to calculate the correspondence between fields based on their relative positions in the document structure. This calculation first abstracts the document structure into a tree model, where each field is a node in the tree. The relative position of a node is determined by its order in the depth-first traversal. The position of the title field is 1, and the position of the body paragraph is 2. Next, the relationship between the structural difference features of multiple fields (i.e., the differences calculated by Euclidean distance) and the change magnitude of the structural weight mapping sequence is determined. This process is achieved by calculating the Euclidean distance between the structural weight mapping sequences of the two versions of the document before and after editing. Assuming that the structural weight value of all fields is 0 before editing, the quantified value of the change magnitude is the norm of the current structural weight mapping sequence. The system sets a change magnitude threshold. This threshold is based on a simulation experiment of the collaborative editing process of 1000 different types of documents (such as reports, meeting minutes, and contracts). The distribution of the structural weight mapping sequence norm when effective synchronization is triggered is statistically analyzed. The lower quartile value of 0.15 is used as the judgment standard for the change magnitude threshold. When the quantized value of the change magnitude is 0.4099, which is greater than 0.15, the synchronization filtering process is initiated to filter out the synchronization target fields of the target documents to form a synchronization target field set. When the quantized value of the change magnitude is less than or equal to 0.15, it is judged as a low-amplitude change, the synchronization filtering process is not initiated, and regular synchronization is directly performed on the changed fields.

[0082] Step S280: When synchronous filtering is started, the synchronous target fields of the target document are filtered out to form a synchronous target field set, and a field-level change labeling sequence is generated.

[0083] The field-level change labeling sequence includes the selected field ID, the path level normalized value of the field, the structural weight value of the field, the field correspondence label of the field, and so on.

[0084] In one example, taking the collaboratively edited "Project Market Analysis Report" document mentioned above, the synchronous screening criterion is to set a structural weight value threshold, which is defined as 1.2 times the average structural weight value of all currently changed fields, i.e. Each structural weight value in the structural weight mapping sequence is compared with its threshold. The structural weight value of the title field (0.3802) is greater than 0.32016, so it is selected as the synchronization target field. The structural weight value of the body paragraph field (0.1534) is less than 0.32016, so it is excluded. Finally, the selected field ID, the path level indicator corresponding to the field (i.e., the path level normalized value of 0.21), the structural weight value (0.3802), and the field corresponding relationship label are packaged into a data structure to generate a field-level change labeling sequence.

[0085] Step S300: Based on the field-level change calibration sequence, perform synchronous trigger analysis on the input events of each synchronous target field of the target document within the current preset time period, determine the input events that meet the synchronous trigger for each synchronous target field, and generate a corresponding synchronous trigger sequence for each synchronous target field independently.

[0086] The synchronous trigger sequence includes character change frequency records, editing operation type classification identifiers, and density feature labels.

[0087] In one embodiment, based on the field-level change calibration sequence, the input events of each synchronized target field of the target document within the current preset time period are analyzed for synchronization triggering. This determines that each synchronized target field meets the input events for synchronization triggering, and generates a corresponding synchronization triggering sequence independently for each synchronized target field, including:

[0088] Based on the field-level change calibration sequence, the timestamps and character modification counts of input events for each synchronized target field in the target document within the current preset time period are obtained. The input events for each synchronized target field are sorted chronologically to obtain the chronological sequence value for each synchronized target field. Based on the chronological sequence value of each synchronized target field, the time interval and character change count between each pair of adjacent input events are calculated, and the ratio of the character change count to the time interval for each pair of adjacent input events is calculated to obtain the editing frequency of each pair of adjacent input events for each synchronized target field. A trend analysis sliding window is used to analyze the editing frequency of each pair of adjacent input events in the N most recent input events for each synchronized target field to determine the input distribution trend coefficient for each synchronized target field within the current preset time period. A synchronization analysis sliding window is then used. The system uses a window to analyze the time intervals of adjacent input events within the most recent Q seconds for each synchronization target field, determining the average time interval for each synchronization target field within the current preset time period. It also uses a synchronization analysis sliding window to analyze the number of character changes within the most recent Q seconds for each synchronization target field, determining the total number of character changes for each synchronization target field within the current preset time period. The system compares the average time interval, total number of character changes, and input distribution trend coefficient for each synchronization target field within the current preset time period with the synchronization triggering criteria to determine whether the input events for each synchronization target field within the current preset time period meet the synchronization triggering criteria. For input events within the current preset time period that meet the synchronization triggering criteria, the trigger times of these events are recorded in the synchronization triggering sequence of the corresponding synchronization target field.

[0089] Specifically, the process involves obtaining a field-level change calibration sequence, collecting the timestamps and character modification counts of input events for each synchronized target field of the target document within the current preset time period of the terminal, sorting the input events of each synchronized target field in chronological order, and obtaining the time sequence value of each synchronized target field.

[0090] In one example, a field-level change marker sequence is obtained, which indicates that the field to be synchronized is the title field (ID 1). Then, input events on the title field within the current preset time period are collected. The input event records include timestamps (in milliseconds) and the number of character modifications (positive for additions, negative for deletions). For example, if a user performs a series of operations on the title field within 10 seconds, the records are as follows: {Timestamp:1695692822105,Chars:2},{Timestamp:16956928 {22950,Chars:-1},{Timestamp:1695692823500,Chars:3},{Timestamp:1695692824800,Chars:1}; First, these input events are sorted in ascending order according to their timestamps. Since the collected input events are already basically ordered, the sorting process is mainly for verification and organization to ensure the strict temporal order of the event sequence. The sorted input event sequence remains logically unchanged, but each input event is assigned a sequential index starting from 1 to obtain the time sequence value.

[0091] Specifically, based on the time sequence value, the input distribution trend coefficient of character changes between consecutive input events is calculated. That is, the editing frequency of consecutive input events is detected, and the number of character changes and the time interval between adjacent input events are calculated to obtain the input distribution trend coefficient.

[0092] In one example, based on the time sequence values, i.e., the sorted event sequence [{Index:1,Timestamp:1695692822105,Chars:2},{Index:2,Timestamp:1695692822950,Chars:-1},{Index:3,Timestamp:1695692823500,Chars:3},{Index:4,Timestamp:1695692824800,Chars:1}], the input distribution trend coefficient of character changes between consecutive input events is calculated. This calculation first calculates the time interval between adjacent input events pairwise ( ) and number of character changes ( ),in, m Number each pair of adjacent input events, between input event 1 and input event 2. millisecond, Character, between input event 2 and input event 3, millisecond, Character, between input event 3 and input event 4, millisecond, The character is then analyzed, and the editing frequency of consecutive input events is detected. The editing frequency is calculated by measuring the rate of change of the character over time intervals. The measurement is based on the number of characters per second (characters / second), where the character change rate is the rate calculated using the time interval. For example, the editing frequency in the first time interval is... The input distribution trend coefficient is calculated by multiplying the number of character changes per second by the time interval between adjacent input events. This coefficient is defined as the ratio of the standard deviation of the character change rate to the mean within a trend analysis sliding window consisting of the most recent N input events (N is usually 5). Assuming the current trend analysis sliding window has only 4 input events (i.e., 3 pairs of adjacent input events), and the mean is... If the sample standard deviation is 2.59, then the input distribution trend coefficient is 2.59 / 2.47 ≈ 1.048.

[0093] Specifically, the input distribution trend coefficient is used to characterize the continuity of operations, classify the behavior types composed of character change range and operation continuity, construct the density feature label of the current editing behavior (the density feature label includes the average time interval, total number of character changes and input distribution trend coefficient), compare the density feature label with the synchronization triggering standard, and establish the synchronization triggering sequence.

[0094] Specifically, in this embodiment, the calculated input distribution trend coefficient is 1.048 (used to quantify operation continuity). The behavior type composed of character change range and operation continuity is classified. This process compares the three indicators of total character change, average time interval and input distribution trend coefficient in the most recent synchronous analysis sliding window (e.g., the currently captured 5-second window) with the synchronous triggering standard (i.e., a set of preset thresholds). Table 1, the collaborative editing behavior parameter statistics table, lists the original user behavior statistics used to determine each threshold.

[0095] Table 1. Statistics of Collaborative Editing Behavior Parameters:

[0096] ;

[0097] As shown in Table 1, by analyzing 500 typical collaborative editing session raw records, the thresholds for various aspects of the synchronization triggering standard were set: The process for setting the time interval density threshold is to calculate the distribution of the number of input events per unit time (1 second) in all sessions, assuming that the upper quartile value is 5 events / second, that is, the time interval density threshold is 200 milliseconds. If the average time interval between the current adjacent input events is... The threshold for character change is set by calculating the mean (assuming a result of 25) and standard deviation (assuming a result of 10) of the character modification quantity distribution within the synchronous analysis sliding window (5 seconds). The sum of these two values ​​is 35 characters, which is used as the character change quantity threshold. If the total number of character changes by the user within the current synchronous analysis sliding window is... The number of characters is less than the character change threshold. The process of setting the input distribution trend coefficient threshold is to statistically analyze the coefficient of variation distribution in stable editing states across all sessions, and take its critical value (e.g., 1.5) as the input distribution trend coefficient threshold. This input distribution trend coefficient reflects the rhythm stability of the input; the smaller the input distribution trend coefficient, the more stable the rhythm (continuous input), and the larger the input distribution trend coefficient, the more discrete the rhythm (intermittent input). The density feature label is compared with the synchronization trigger standard. Specifically, the indicator values ​​of the current behavior {average time interval: 898.3ms, character change amount: 7, input distribution trend coefficient: 1.048} are compared with the standard {time interval density threshold: 200ms, character change amount threshold: 35, input distribution trend coefficient threshold: 1.5} for item-by-item matching. In this example, the average time interval 898.3ms is greater than 200ms (low frequency), the character change amount 7 is less than 35, and the input distribution trend coefficient 1.048 is less than 1.5 (indicating a relatively stable rhythm, but considering the time interval, it belongs to slow and stable input). The system logic is set to trigger synchronization when: the average time interval is less than a threshold (high frequency of character changes), the number of character changes is greater than a threshold (significant changes), or the average time interval is greater than a threshold and the input distribution trend coefficient is greater than a threshold. The average time interval being greater than the threshold and the input distribution trend coefficient being greater than the threshold indicates that the user is in a discontinuous editing state (i.e., there are obvious gaps in thinking or pausing). Synchronization is performed using these highly discrete pauses to ensure data updates without interfering with the user's continuous thinking process. In this example, since none of the indicators meet the triggering conditions for high-frequency, large-scale modifications, or slow and unstable input, the system determines that the current editing is low-density and does not trigger synchronization. If the user performs a paste operation in the next second, and the number of character changes becomes 50, the comparison result becomes {greater than, greater than, less than}. At this point, the triggering condition is met, and the trigger time of the input event that meets the synchronization triggering criteria is recorded in the synchronization trigger sequence of the corresponding synchronization target field.

[0098] Step S320: Analyze the continuous triggering time in the synchronization trigger sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled.

[0099] In one embodiment, the continuous triggering time in the synchronization trigger sequence of each synchronization target field is analyzed to determine the synchronization control amplitude value of the current synchronization target field to be controlled, including:

[0100] Analyze the synchronization trigger sequence of each synchronization target field to determine the time interval between adjacent trigger times in the synchronization trigger sequence of each synchronization target field; analyze the time interval between adjacent trigger times in the synchronization trigger sequence of each synchronization target field to determine the synchronization trigger change trend of each synchronization target field; analyze the trigger frequency change of each synchronization target field to determine the amount of trigger frequency change of each synchronization target field; analyze the amount of trigger frequency change of each synchronization target field to determine the cumulative offset direction of the frequency change of each synchronization target field; synchronization target fields whose synchronization trigger change trend is that the trigger rhythm first speeds up and then slows down, or whose cumulative offset direction of frequency change is positive, are the synchronization target fields to be controlled; use the synchronization control amplitude calculation formula to analyze the synchronization control amplitude of the synchronization target fields to be controlled, and determine the synchronization control amplitude value of the current synchronization target field to be controlled.

[0101] If the synchronous triggering trend does not meet the requirement that the triggering rhythm first accelerates and then slows down or the offset direction is negative, then the current synchronization mode will be maintained and no amplitude adjustment calculation will be performed.

[0102] Specifically, the continuous trigger times in the synchronization trigger sequence of each synchronization target field are compared, the adjacent trigger time intervals are detected and the changing trend is calculated, the synchronization operation response frequency is recorded, and the change in trigger frequency of each synchronization target field is obtained.

[0103] In one example, a synchronization trigger sequence is based on a certain synchronization target field. This synchronization trigger sequence records the time points when the synchronization conditions are met each time. For example, a synchronization trigger sequence might be a set of time nodes containing detailed records: [{Time:1695692830000,Freq:1.2,Type:"Insert",Density:High},{Time:1695692832500,Freq:0.8,Type:"Update",Density:Medium},...]. Here, Freq represents the character change frequency record, Type represents the edit operation type classification identifier, and Density represents the density feature label. To simplify the description, the trend analysis below will focus on the timestamp sequence [1695692830000, 1695692832500, ...] in this sequence. First, the consecutive trigger times in the synchronous trigger sequence are compared, the time interval between adjacent trigger times is detected, and the synchronous trigger change trend is calculated. The time interval between the first adjacent trigger times is 2500 milliseconds, the time interval between the second adjacent trigger times is 1500 milliseconds, and the time interval between the third adjacent trigger times is 3000 milliseconds. The synchronous trigger change trend is measured by calculating the slope or first difference of the time interval sequence between adjacent trigger times. The first difference sequence is [-1000, 1500], indicating that the triggering rhythm first speeds up and then slows down.

[0104] It should be understood that the system only considers it necessary to calculate the "adjustment range" and intervene to adjust the synchronization mode configuration when the synchronization trigger change trend shows "fast at first and slow later" (i.e., unstable burst input). If the input is stable, synchronization can be performed directly.

[0105] In one example, the frequency of synchronization operation triggers is recorded. This frequency is defined as the number of triggers occurring within a statistical sliding window (e.g., 10 seconds). For instance, within 10 seconds from timestamp 1695692830000 to 1695692840000, there are 4 triggers, resulting in a trigger frequency of 0.4 triggers per second. By comparing the changes in trigger frequency within consecutive statistical sliding windows, the change in trigger frequency is obtained. Based on the change in trigger frequency, the cumulative offset direction of the frequency change is determined. For example, if the trigger frequency in the previous statistical sliding window was 0.2 triggers per second, and the current trigger frequency is 0.4 triggers per second, the change in trigger frequency is +0.2 triggers per second. If the cumulative offset direction of the frequency change is positive, indicating a tendency towards more frequent triggers, then the synchronization mode configuration needs to be adjusted (if the direction is reversed, maintain or synchronize immediately).

[0106] It should be understood that if the cumulative offset direction is positive, it means that the synchronization operation is becoming more and more frequent and needs to be adjusted. Usually, merging is used to reduce jitter. If the cumulative offset direction is negative, it means that the synchronization operation is becoming more and more sparse. Usually, hold or instant synchronization is used.

[0107] Specifically, the synchronization control amplitude of the target field to be controlled is analyzed using the synchronization control amplitude calculation formula to determine the current synchronization control amplitude value of the target field to be controlled, and the synchronization mode configuration is determined. The synchronization control amplitude value is calculated using the synchronization control amplitude calculation formula, and the synchronization execution order is adjusted according to the synchronization control amplitude value to obtain the synchronization mode configuration. The synchronization control amplitude value calculation formula is as follows:

[0108] ;

[0109] in, To synchronize the amplitude value; An index for the time interval between adjacent trigger times; The upper limit of the index represents the total number of time intervals between adjacent triggering events within the analysis period (if there are N triggering events, then m = N-1). For the first The normalized value of the time interval between adjacent trigger times, by using the trigger time difference Divide by base time get; For the first The normalized value of the response frequency of the starting trigger event corresponding to the time interval between each adjacent trigger time is used to determine the trigger frequency at the start of this time interval. Divide by the reference frequency get, To determine the normalized cumulative value of the difference mapping sequence in the field-level change calibration sequence, the cumulative difference value is obtained by... Divide by the maximum character length get.

[0110] The difference mapping sequence consists of the structural weight values ​​of all synchronization target fields in the field-level change calibration sequence.

[0111] Among them, the cumulative difference value This is the cumulative value of all structural weights in the difference mapping sequence.

[0112] Specifically, based on the change in trigger frequency, for example, if the frequency was 0.2 times / second in the previous unit of time (10 seconds) and is currently 0.4 times / second, the change is +0.2 times / second. If the cumulative offset direction of the frequency change is positive, indicating a tendency towards more frequent changes, then the synchronization response configuration needs to be adjusted (if it's negative, maintain or immediately respond to the synchronization mode configuration). The synchronization control amplitude value is calculated using the following formula. Calculate the synchronous control amplitude value, where, To synchronize the adjustment of the amplitude value, For the index that triggered the event, To analyze the number of triggered events within the period, here (This indicates there are 3 intervals, i.e., 4 trigger events). For the first The normalized value of each adjacent trigger time interval. For the first The normalized value of the response frequency corresponding to the start time of each time interval is obtained by using the trigger frequency. Divide by the reference frequency get, The maximum synchronization frequency that the system can handle is set to 2 times / second. Assuming the instantaneous frequencies corresponding to the first three trigger events are 0.3 times / second, 0.5 times / second, and 0.35 times / second respectively, then... , , , The normalized cumulative value of the difference mapping sequence in the field-level change calibration sequence is the sum of the structure weight values ​​of all fields to be synchronized, divided by the maximum character length. (Set to 10% of the total document length, for example, if the document is 10,000 characters long, then it is 1,000). Assume the sum of the structural weight values ​​of the current synchronization target field. ,but The logic behind this formula for calculating the synchronization amplitude lies in the fact that the numerator quantifies the "potential energy" of a synchronization event by summing the products of time and frequency. That is, the shorter the interval and the higher the frequency of the event combination, the greater the contribution. The square root term in the denominator introduces the influence of the content change amplitude. When the content differences are significant, the denominator is increased to reduce the control amplitude, avoiding overly frequent synchronization during major structural changes. The advantage of this formula is that it combines the triggering rhythm of the time dimension with the change amplitude of the content dimension to generate a comprehensive scheduling index. Now, let's perform numerical calculations:

[0113] ;

[0114] ;

[0115] .

[0116] The synchronization control amplitude value refers to a dimensionless numerical quantification of the control index during multi-terminal synchronization. It represents the combined influence of the time interval trend of triggering events, the frequency distribution of responses, and the magnitude of field differences. Specifically, this synchronization control amplitude value is calculated using a unified normalization process, allowing different parameters to be compared on the same scale, thus deriving an intensity metric that can be directly used for synchronization scheduling. Its effect is to serve as the core basis for adjusting the synchronization execution order. The value directly corresponds to the scheduling priority. A larger synchronization control amplitude value indicates dense triggering with limited differences, and the system tends to execute synchronization quickly. A smaller value indicates scattered triggering or significant differences, and the system slows down the synchronization pace, thereby providing a quantitative operational reference for maintaining cross-terminal data consistency.

[0117] Step S340: Determine the synchronization mode configuration based on the synchronization control amplitude value of the target field to be controlled, and adjust the current synchronization execution mode of the target field to be controlled according to the synchronization mode configuration.

[0118] The synchronization mode configuration includes synchronization granularity scheduling parameters, execution order adjustment basis, and synchronization response configuration items.

[0119] The execution order adjustment is based on the weight value of the data packet to be synchronized in the sending queue (e.g., priority, normal, delayed) used to determine the priority of the input event to be synchronized. The execution order adjustment is not only based on the order, but also on the "priority weight". For example, a high-amplitude task is moved to the front of a low-amplitude task.

[0120] The synchronization response configuration items refer to control parameters related to network transmission, including request timeout, retry limit, and whether to block user interface (UI) interaction feedback settings. For example: whether to require immediate server ACK (acknowledgment), what the timeout period should be, and whether to display a "synchronizing" spinning icon on the UI.

[0121] Specifically, the synchronization granularity scheduling parameters (i.e., the number of fields merged in a single synchronization packet and the time span) in the synchronization mode configuration are determined by combining the set of fields constituting the synchronization target. The synchronization granularity scheduling parameters refer to "the amount of data included in a single synchronization" or "the degree of merging". Larger granularity means more merging and lower frequency.

[0122] The synchronization target field set includes the structural weight values ​​of the fields. The logic for determining the synchronization granularity scheduling parameter should be: if the synchronization target field set contains high-weight (important) fields, the synchronization granularity scheduling parameter is small (no merging, fast transmission); if the synchronization target field set contains low-weight fields, the synchronization granularity scheduling parameter is large (multiple merging points before transmission).

[0123] Specifically, determining the synchronization granularity scheduling parameters (i.e., the number of fields merged in a single synchronization packet and the time span) in the synchronization mode configuration, based on the set of fields constituting the synchronization target, includes: weighted evaluation of the structural weight values ​​of all fields in the synchronization target field set; if there are core fields in the synchronization target field set with structural weight values ​​greater than the importance threshold, a smaller synchronization granularity scheduling parameter is determined (i.e., reducing the merging wait time, such as 500 milliseconds) to ensure rapid synchronization of core content; if the synchronization target field set consists entirely of low-weight fields, a larger synchronization granularity scheduling parameter is determined (i.e., increasing the merging quantity, such as accumulating to 50 characters or 2 seconds) to reduce network overhead. The synchronization granularity scheduling parameter refers to the 'size of data included in a single synchronization' or the 'degree of merging'. Larger granularity means more merging and a lower frequency.

[0124] In one example, the synchronization control amplitude value of 0.248 is a dimensionless control amplitude value, ranging from 0 to 1. The system sets the range as follows: [0, 0.3) is low amplitude, tending to merge synchronization; [0.3, 0.7) is medium amplitude, performing regular synchronization; [0.7, 1.0] is high amplitude, performing priority, immediate synchronization. The current synchronization control amplitude value of 0.248 is in the low amplitude range, so the system will adjust the synchronization execution order, choosing to merge the current synchronization with the next synchronization request to obtain the synchronization mode configuration. For example, the generated synchronization mode configuration is: {Mode: "Merge Synchronization", Synchronization granularity scheduling parameters: "Cache wait time = 3000ms, maximum number of characters to merge = 100", Execution order adjustment basis: "Priority = Low", Synchronization response configuration item: "Background silent transmission, no UI blocking"}.

[0125] The aforementioned multi-terminal synchronized mobile office collaboration method analyzes the field editing behavior parameter set of the target document to obtain the path hierarchy name, the number of characters before and after text modification, and the editing operation type of each field. Based on this analysis, a structural weight mapping sequence is generated. The structural weight mapping sequence is used to analyze the change range of the target document before and after editing, determine the quantified value of the change range, and compare this quantified value with a change range threshold to determine whether to initiate synchronization filtering. If synchronization filtering is initiated, the synchronized target fields of the target document are selected to form a synchronized target field set, and a field-level change calibration sequence is generated. Based on the field-level change calibration sequence, the input events of each synchronized target field of the target document within the current preset time period are analyzed for synchronization triggering to determine the input events that meet the synchronization triggering criteria and generate a synchronization trigger sequence for each synchronized target field. The continuous trigger times in the synchronization trigger sequence of each synchronized target field are analyzed to determine the synchronization control amplitude value of the currently controlled synchronized target field. Based on the synchronization control amplitude value of the controlled synchronized target field, the synchronization mode configuration is determined, and the current synchronization execution mode of the controlled synchronized target field is adjusted according to the synchronization mode configuration. Therefore, under conditions of dense terminal input events and frequent cross-operations, the dynamic scheduling mechanism of characterizing the density of editing behavior and the triggering rhythm avoids the risk of false triggering or delay in synchronous response and improves the accuracy of data synchronization.

[0126] In one embodiment, the multi-device synchronized mobile office collaboration method further includes:

[0127] Obtain the user identifier and field number of the current input event to be synchronized, and compare them with the initial user identifier of the field number in the attribution chain sequence value to determine whether the user of the current input event to be synchronized and the initial user are the same user. If the user of the current input event to be synchronized and the initial user are not the same user, add the field number to the attribution inconsistency field number set. Analyze the attribution inconsistency fields in the density analysis window of the target document based on the attribution inconsistency field number set to determine the distribution density in each density analysis window. Classify the fields in the density analysis window with a distribution density greater than the density threshold into attribution risk zone fields, forming the field number set of the attribution risk zone. Then, classify the fields in the attribution risk zone... The analysis of the pending input event records in the risk zone fields of the set is used to determine the operation type and character modification range of each pending input event in each risk zone field. The pending input events in each risk zone field are grouped together, and the pending input events within each group are sorted according to their arrival timestamps on the server to obtain the grouping sequence for each field. The type difference score and overlap are analyzed based on the operation type and character modification range of each pair of adjacent pending input events in each field grouping sequence to determine the semantic difference between adjacent pending input events in each field grouping sequence. Based on the semantic difference between adjacent pending input events in each field grouping sequence, the intention conflict fields are identified, and the synchronization of input events in the intention conflict fields is paused.

[0128] Among them, the input events to be synchronized are the input events that need to be synchronized at present.

[0129] This involves calling the initial editing behavior records of fields associated with the task document, analyzing the first and second user identifier, operation time, and field number of each field in the task document, establishing a user affiliation chain according to the field number, and obtaining the affiliation chain sequence value.

[0130] In one example, before synchronization, to ensure data consistency, the system calls the initial field editing behavior record associated with the task document. This record is a triplet of {field number, first user ID, creation time} automatically archived by the system at the beginning of document creation and when subsequent content is added. For example, for the "Project Market Analysis Report" document, its initial record is:

[0131] [{FieldID:1,User:“UserA”,Time:1695692000000},{FieldID:2,User:“UserA”,Time:1695692050000},{FieldID:3,User:“UserB”,Time:1695692100000}], where Field 1 is the title, Field 2 is the first paragraph, both created by UserA, and Field 3 is the second paragraph, created by UserB. These records are sorted and organized according to their field numbers to create a mapping table or chain structure with the field number as the key and the user ID and time as the value. This structure is the user affiliation chain, and the affiliation chain sequence value is obtained.

[0132] Specifically, based on the attribution chain sequence value, the correspondence between the current user identifier and the first user identifier in the attribution chain is determined, the set of fields with inconsistent user attribution is screened, and the set of attribution inconsistency numbers is obtained.

[0133] In one example, based on the mapping relationship of the attribution chain sequence values, i.e., {1:“UserA”, 2:“UserA”, 3:“UserB”}, the correspondence between the user identifier of the currently occurring editing operation and the first user identifier recorded in the attribution chain sequence value is determined. Assuming that the current operation is initiated by user C and its target is field 2, the user identifier of the current operation “UserC” is extracted, and the first user identifier of field 2 is retrieved from the attribution chain sequence value to obtain “UserA”. Then, “UserC” and “UserA” are compared for a complete string match. Since “UserC” is not equal to “UserA”, it is determined that the attribution is inconsistent, and the number 2 of the field is recorded. If user A modifies field 2, it is determined that the attribution is consistent, and no record is made. By traversing all currently pending input events, all fields with inconsistent user attribution are screened out, and the set of inconsistent field numbers is obtained.

[0134] Specifically, based on the set of inconsistent field numbers, the inconsistent fields in the density analysis window of the target document are analyzed to determine the distribution density in each density analysis window; fields in the density analysis window with a distribution density greater than the density threshold are classified as fields in the risk zone, forming a set of field numbers for the risk zone, and a collaborative conflict judgment result is established.

[0135] In one example, based on the set of inconsistent field numbers, such as {2,5,6,8,12}, the distribution density of inconsistent fields in the entire document's field sequence is calculated. This calculation first determines a density analysis window, which can be a fixed number of consecutive fields (e.g., 10 consecutive fields) or a logical part of the document (e.g., a chapter). Taking a density analysis window with 10 fields as an example, when the density analysis window slides from field 1 to field 10, the inconsistent field numbers within the density analysis window are {2,5,6,8}, with a distribution density of 4 / 10 = 0.4. When the density analysis window slides to fields 3 to 12, the inconsistent field numbers within the density analysis window are {2,5,6,8}. The inconsistent field numbers are {5, 6, 8, 12}, with a distribution density of 0.4. The system sets a density threshold based on the analysis of historical collaboration conflict data. Statistics show that when the density of non-creator editing behavior in a local area exceeds 0.3, the probability of version conflict increases by 70%. Therefore, the density threshold is set to 0.3. The calculated distribution density of 0.4 is compared with the density threshold of 0.3. Since 0.4 > 0.3, the current density analysis window area is determined to constitute an attribution risk zone. All field numbers {1, 2, ..., 10} within this density analysis window are filtered and recorded as the set of field numbers for the attribution risk zone, establishing the collaboration conflict judgment result. The collaboration conflict judgment result includes the set of field numbers for the attribution risk zone, the field numbers with inconsistent user attribution, and the attribution change distribution index (i.e., the distribution density of the inconsistent attribution fields calculated above).

[0136] Among them, based on the collaborative conflict judgment results, the record table of input events to be synchronized on the same risk zone field of multiple terminals is analyzed, the operation type and character modification range of each input event to be synchronized in the record table of input events to be synchronized are calculated, and the input events to be synchronized are grouped and sorted according to the field number to obtain the field grouping sequence.

[0137] In one example, based on the collaborative conflict judgment result, which specifies that the set of field numbers {1,2,…,10} belonging to the risk zone is the risk zone, and the fields within the risk zone are the risk zone fields, the analysis then focuses on the input events (i.e., input events to be synchronized) from multiple different terminals on these fields. Assume that on risk zone field 6 (a key data point), operations from users B and C are received simultaneously, and user B's input event record is {User:“B”,Type:“Delete”,Range:[5,10]}.} means to delete the 5th to 10th characters. User C's operation record is {User:“C”,Type:“Update”,Range:[7,8],Content:“9”}, which means to modify the 7th to 8th characters to “9”. First, calculate the operation type and character modification range of each input event to be synchronized. Then, group and sort these concurrent input events according to the field number. All input events to be synchronized for field 6 are grouped together. Within the group, they are sorted according to the timestamp of the input event to be synchronized arriving at the server, thus obtaining the field grouping sequence of field 6.

[0138] Specifically, based on the field grouping sequence, the semantic differences of input events to be synchronized in the same group are compared, a semantic priority queue is constructed based on the semantic differences, the influence range of adjacent input events to be synchronized in the semantic priority queue is judged in order, and the queue priority judgment result is generated.

[0139] In one example, based on the field grouping sequence, specifically the field grouping sequence for field 6 [input event to be synchronized B, For input event C to be synchronized, the semantic differences between adjacent input events to be synchronized within the same group are compared. This process does not rely on complex natural language processing models, but is carried out through a set of quantification rules based on input event attributes. The rules are as follows: First, calculate the type difference score of adjacent input events to be synchronized by calling the system's preset type distance matrix. In this type distance matrix, the distance between "delete" and "update" is defined as 0.8. Second, calculate the overlap of the influence range of adjacent input events to be synchronized. The range of input event B to be synchronized is [5,10], and the range of input event C to be synchronized is [7,8]. The overlap interval is [7,8], and the overlap is calculated as overlap length / union length = 2 / (10-5+1) = 0.33. Finally, the type difference score and the overlap are weighted and summed. The weights are set to 0.7 and 0.3 according to experimental data. The semantic difference between input events B and C to be synchronized is 0.8×0.7+0.33×0.3=0.56+0.099=0.659.

[0140] In one embodiment, based on the semantic differences between adjacent input events to be synchronized in each field grouping sequence, conflicting intent fields are identified, and the synchronization of input events for conflicting intent fields is paused, including:

[0141] Adjacent input events awaiting synchronization whose semantic difference exceeds the difference threshold are placed in a conflict processing queue. The influence range of adjacent input events awaiting synchronization in the conflict processing queue is judged according to the logical order of their order to determine the priority judgment result of the queue. Based on the priority judgment result of the queue, adjacent input events awaiting synchronization in the conflict processing queue whose semantic difference exceeds the conflict confirmation threshold are judged according to the priority order, and the field to which the adjacent input event awaiting synchronization belongs is determined as the intention conflict field. The intention conflict field is added to the set of fields to be separated and processed, and the synchronization of fields in the set of fields to be separated and processed is suspended.

[0142] In one example, a semantic priority queue is constructed based on this semantic difference degree. A low difference threshold of 0.2 is set. If the semantic difference degree is less than or equal to this threshold, it is considered a collaborative edit that can be automatically merged; if it is greater than this threshold, it is considered a potential conflict. If the semantic difference degree is less than or equal to this threshold, it is considered a non-conflicting operation, and the system executes an automatic merging strategy to synchronize data. The semantic difference degree of the input events B and C to be synchronized is 0.659, which is much greater than 0.2. Therefore, input events B and C to be synchronized are placed in the conflict processing queue. The influence range of adjacent input events to be synchronized in the conflict processing queue is judged sequentially. Since the range of input event B completely includes... The scope of the input event C to be synchronized was determined, and an execution dependency relationship existed. A queue priority judgment result was generated. Based on the queue priority judgment result, conflict positions were filtered according to the semantic difference degree of 0.659 between the input events B and C to be synchronized in field 6. The system set a high conflict confirmation threshold of 0.6. This conflict confirmation threshold was determined by reviewing and analyzing the conflicts that were manually resolved in the past, and the semantic difference quantiles that can cover 95% of the real intention conflicts were identified. Since 0.659 is greater than 0.6, the system confirmed that a high probability of intention conflict had occurred in field 6. The system then marked it as a field to be separated and added field number 6 to the set of fields to be separated and processed. This means that the synchronization of this field will be suspended.

[0143] In one embodiment, the multi-device synchronized mobile office collaboration method further includes:

[0144] Adjacent input events to be synchronized that have a semantic difference exceeding the conflict confirmation threshold in the intent conflict field are recorded in the intent conflict matching table of the intent conflict field. The intent conflict matching table records the operation user identifiers, operation details, and semantic difference of the two conflicting input events. The conflict details of adjacent operations with semantic differences exceeding the conflict confirmation threshold are pushed to the interface of the relevant user, requesting the relevant user to perform relevant processing.

[0145] Specifically, based on the queue priority judgment result, conflict positions are filtered according to semantic differences, and an intent conflict matching table for the corresponding fields is constructed.

[0146] The conflict details include the user identifiers of both parties involved in the conflict, the time of the operation, the specific text modifications made, and the calculated semantic difference.

[0147] In one example, an intent conflict matching table corresponding to field 6 is constructed. This intent conflict matching table is a structured data that records the user identifiers, operation details, and semantic differences of the two parties involved in the input event conflict. The format is {FieldID:6,Conflicts:[{User:“B”,Action:{…}},{User:“C”,Action:{…}}],SemanticDifference:0.659}. The conflict details will be pushed to the interface of the relevant users, requiring manual confirmation or merging, resulting in a field sequence identifier set. The field sequence identifier set includes semantic priority labels, field numbers of the intent conflict fields, and a conflict position correspondence table.

[0148] Among them, the semantic priority label is a semantic priority label marked according to the queue sorting result.

[0149] Among them, the conflict position correspondence table is the mapping of overlapping intervals (such as the mapping between UserB[5,10] and UserC[7,8]).

[0150] The aforementioned multi-terminal synchronized mobile office collaboration method structurally encodes the path hierarchy, modification length, and editing type of fields. It constructs behavioral density feature labels by combining the time interval of input events and the range of character changes. The method dynamically adjusts the synchronization rhythm by comparing these density labels with triggering standards. It establishes a correspondence between operating users and initial field editing records and determines the location of ownership conflicts. Based on the degree of semantic differences and the relationship between operations, it constructs a conflict sorting sequence, achieving collaborative decoupling of conflicting fields and accurate filtering of target fields. This enhances the ability to determine field granularity and the response to behavioral triggering rhythm in multi-terminal synchronized mobile office scenarios, improving the level of data consistency management across multiple terminals. However, it lacks an ownership determination process based on the order of editing records in user operation identification, leading to errors in conflict field coverage. Furthermore, it lacks structural grouping and logical sorting mechanisms in scenarios with differentiated semantic operations across terminals, resulting in a lack of basis for data merging. Consequently, when multiple users edit document fields simultaneously, the system processes data solely based on timestamp order, leading to the loss of critical operation results and confusion in merging logic.

[0151] The aforementioned multi-terminal synchronized mobile office collaboration method structurally encodes the path hierarchy, modification length, and editing type of fields. It constructs behavioral density feature labels by combining the time interval of input events and the range of character changes. The method dynamically adjusts the synchronization rhythm by comparing these density labels with triggering standards. It establishes a correspondence between operating users and initial field editing records and determines the location of ownership conflicts. Based on semantic differences and the relationship between operations, it constructs a conflict sorting sequence, achieving collaborative decoupling of conflicting fields and accurate filtering of target fields. This enhances the ability to determine field granularity and respond to behavioral triggering rhythms in multi-terminal synchronized mobile office scenarios, improving the level of data consistency management across multiple terminals. However, it lacks an ownership determination process based on the order of editing records in user operation identification, leading to errors in conflict field coverage. Furthermore, it lacks structural grouping and logical sorting mechanisms in scenarios with differentiated semantic operations across terminals, resulting in a lack of basis for data merging. Consequently, when multiple users edit document fields simultaneously, the system processes data solely based on timestamp order, leading to the loss of critical operation results and confusion in merging logic.

[0152] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0153] In one embodiment, such as Figure 2 As shown, a multi-terminal synchronized mobile office collaboration system includes: a behavior parameter analysis module 110, a weight analysis module 120, a change amplitude analysis module 130, a synchronization filtering module 140, a synchronization trigger analysis module 150, a synchronization control amplitude analysis module 160, and a synchronization control module 170.

[0154] The behavior parameter analysis module 110 is used to analyze the field editing behavior parameter set of the target document and obtain the path level name, the number of characters before and after text modification, and the type of editing operation for each field.

[0155] The weight analysis module 120 is used to analyze each field based on its path hierarchy name, the number of characters before and after text modification, and the type of editing operation, and generate a structural weight mapping sequence.

[0156] The change magnitude analysis module 130 is used to analyze the change magnitude of the target document before and after editing based on the structural weight mapping sequence, determine the quantitative value of the change magnitude, and compare the quantitative value of the change magnitude with the change magnitude threshold to determine whether to start synchronous filtering.

[0157] The synchronous filtering module 140 is used to filter out the synchronous target fields of the target document to form a synchronous target field set when synchronous filtering is started, and to generate a field-level change label sequence.

[0158] The synchronization trigger analysis module 150 is used to perform synchronization trigger analysis on the input events of each synchronization target field of the target document within the current preset time period based on the field-level change calibration sequence, and to determine the input events that meet the synchronization trigger to generate the synchronization trigger sequence for each synchronization target field.

[0159] The synchronization control amplitude analysis module 160 is used to analyze the continuous triggering time in the synchronization trigger sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled.

[0160] The synchronization control module 170 is used to determine the synchronization mode configuration based on the synchronization control amplitude value of the synchronization target field to be controlled, and adjust the current synchronization execution mode of the synchronization target field to be controlled according to the synchronization mode configuration.

[0161] In one embodiment, the multi-device synchronized mobile office collaboration system further includes: an intent conflict analysis module, used to obtain the operation user identifier and field number of the current input event to be synchronized, and to determine the correspondence between the field number and the initial operation user identifier in the attribution chain sequence value to determine whether the operation user of the current input event to be synchronized and the initial operation user are the same user; if the operation user of the current input event to be synchronized and the initial operation user are not the same user, the field number is added to the attribution inconsistency field number set; based on the attribution inconsistency field number set, the attribution inconsistency fields in the density analysis window of the target document are analyzed to determine the distribution density in each density analysis window; fields in the density analysis window with a distribution density greater than the density threshold are classified as attribution risk area fields, constituting the attribution risk area. The system identifies a set of field numbers belonging to a risk zone; it analyzes the record table of input events to be synchronized on the risk zone fields within the set of field numbers belonging to the risk zone to determine the operation type and character modification range of each input event to be synchronized in each risk zone field; it groups the input events to be synchronized in each risk zone field and sorts them according to the timestamp of arrival at the server to obtain a field grouping sequence; it analyzes the type difference score and overlap based on the operation type and character modification range of each pair of adjacent input events to be synchronized in each field grouping sequence to determine the semantic difference degree of adjacent input events to be synchronized in each field grouping sequence; it analyzes the semantic difference degree of adjacent input events to be synchronized in each field grouping sequence to identify fields with conflicting intents and suspends the synchronization of input events in fields with conflicting intents.

[0162] In one embodiment, the multi-terminal synchronized mobile office collaboration system further includes: a conflict details push module, used to record adjacent input events to be synchronized that have a semantic difference exceeding the conflict confirmation threshold in the intent conflict field in the intent conflict field's intent conflict matching table, the intent conflict matching table recording the operation user identifiers, operation details, and semantic difference of the conflicting input events; and to push the conflict details of the adjacent operations with semantic differences exceeding the conflict confirmation threshold to the interface of the relevant user corresponding to the input event that caused the conflict, requesting the relevant user to confirm the version selection or manually merge the content.

[0163] For specific limitations regarding multi-device synchronized mobile office collaboration systems, please refer to the limitations of multi-device synchronized mobile office collaboration methods mentioned above, which will not be repeated here. Each module in the aforementioned multi-device synchronized mobile office collaboration system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0164] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the above-described multi-terminal synchronized mobile office collaboration method.

[0165] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described multi-terminal synchronized mobile office collaboration method.

[0166] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0167] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0168] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A multi-terminal synchronous mobile office collaboration method, characterized in that, The multi-terminal synchronized mobile office collaboration method includes: Analyze the parameter set of field editing behavior of the target document to obtain the path level name, the number of characters before and after text modification, and the type of editing operation for each field; Based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation, a structural weight mapping sequence is generated. The change magnitude of the target document before and after editing is analyzed based on the structural weight mapping sequence. The quantified value of the change magnitude is determined, and the quantified value of the change magnitude is compared with the change magnitude threshold to determine whether to start synchronous filtering. The quantified value of the change magnitude is the Euclidean distance of the structural weight mapping sequence of the target document before and after editing. When synchronous filtering is initiated, the synchronous target fields of the target document are filtered out to form a synchronous target field set, and a field-level change labeling sequence is generated; Based on the field-level change calibration sequence, the input events of each synchronous target field of the target document within the current preset time period are analyzed for synchronization triggering. Each synchronous target field is determined to meet the input events for synchronization triggering, and a corresponding synchronization triggering sequence is generated independently for each synchronous target field. Analyze the continuous triggering time in the synchronization trigger sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled; Based on the synchronization control amplitude value of the target field to be controlled, determine the synchronization mode configuration, and adjust the current synchronization execution mode of the target field to be controlled according to the synchronization mode configuration; The analysis, based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation, generates a structural weight mapping sequence, including: Based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation, the path hierarchy name is split layer by layer and the path hierarchy value is analyzed. The number of characters before and after text modification is analyzed to determine the text modification length. The types of editing operations are grouped and classified according to the editing intent to determine the operation intensity level. Normalization matching is performed on the path level value, text modification length, and operation intensity level of each field to obtain the normalized path level value, normalized text modification length value, and normalized operation intensity level value of each field. The normalized path level value, normalized text modification length value, and normalized operation intensity level value of each field are then combined into a vector to obtain the field normalization parameter set. Based on the path level normalization value, text modification length normalization value, and operation intensity level normalization value of each field in the field normalization parameter set, the structural weight value of each field is analyzed using the structural weight value analysis formula to generate a structural weight mapping sequence. The formula for analyzing the structural weight values ​​is as follows: ； in, For the first The structural weight values ​​of each field, For the first The normalized matching adjustment coefficient is obtained by adjusting the first... The path hierarchy normalized value, text modification length normalized value, and operation intensity level normalized value of each field are obtained by setting proportional weights. For the first The normalized parameter at the th The mapping results in each field are mapped by mapping the path level normalized value, text modification length normalized value, or operation intensity level normalized value to the first field. The fields are obtained. For the first The normalized text modification length value for each field is obtained by dividing the text modification length by the maximum modification length in the document. For the first The normalized value of the operation intensity level for each field is obtained by dividing the operation intensity level by the preset maximum operation intensity level. The mean of the normalized values ​​of the operation intensity levels for all fields is obtained by applying the normalized values ​​of all fields. We obtain the arithmetic mean. This is a normalized parameter index, indicating the position of the normalized parameter in the field's normalized parameter set. For field indexing, it represents the first [number]th [item] in the task document. One field, This represents the total number of normalized parameters.

2. The multi-terminal synchronized mobile office collaboration method according to claim 1, characterized in that, The multi-terminal synchronized mobile office collaboration method also includes: Obtain the operation user identifier and field number of the current input event to be synchronized, and determine the correspondence between them and the initial operation user identifier of the field number in the belonging chain sequence value to determine whether the operation user of the current input event to be synchronized and the initial operation user are the same user. If the user operating the current input event to be synchronized for the field number is not the same user as the initial user, the field number will be added to the set of inconsistent field numbers. Based on the set of inconsistent attribution field numbers, analyze the inconsistent attribution fields in the density analysis window of the target document to determine the distribution density in each density analysis window; Fields within the density analysis window whose distribution density is greater than the density threshold are classified as fields belonging to the risk zone, forming a set of field numbers belonging to the risk zone. Analyze the record table of input events to be synchronized on the fields of the risk zone in the set of field numbers of the risk zone to determine the operation type and character modification range of each input event to be synchronized in each risk zone field. Grouping each input event to be synchronized in a risk zone field into a group, sorting the input events to be synchronized in each group according to the timestamp of arrival at the server, and obtaining the grouping sequence of each field; Based on the operation type and character modification range of each pair of adjacent input events to be synchronized in the grouping sequence of each field, analyze the type difference score and overlap, and determine the semantic difference of adjacent input events to be synchronized in the grouping sequence of each field. Based on the semantic differences between adjacent input events to be synchronized in each field grouping sequence, conflicting intent fields are identified, and the synchronization of input events for the conflicting intent fields is paused.

3. The multi-terminal synchronized mobile office collaboration method according to claim 2, characterized in that, The multi-terminal synchronized mobile office collaboration method also includes: Adjacent input events to be synchronized that have a semantic difference exceeding the conflict confirmation threshold in the intent conflict field are recorded in the intent conflict matching table of the intent conflict field. The intent conflict matching table records the operation user identifiers, operation details, and semantic difference between the two conflicting input events. The conflict details of adjacent operations whose semantic differences exceed the conflict confirmation threshold are pushed to the interface of the relevant user corresponding to the input event that caused the conflict, requesting the relevant user to confirm the version selection or manually merge the content.

4. The multi-terminal synchronized mobile office collaboration method according to claim 1, characterized in that, The step involves analyzing the input events of each synchronized target field of the target document within the current preset time period based on the field-level change calibration sequence, determining that each synchronized target field meets the input events for synchronization triggering, and generating a corresponding synchronization trigger sequence independently for each synchronized target field, including: Based on the field-level change calibration sequence, obtain the timestamp and character modification count of the input events of each synchronized target field of the target document within the current preset time period, sort the input events of each synchronized target field in chronological order, and obtain the time sequence value of each synchronized target field. Based on the time sequence value of each synchronization target field, the time interval and number of character changes between each pair of adjacent input events are calculated, and the ratio of the number of character changes to the time interval for each pair of adjacent input events is calculated to obtain the editing frequency of each pair of adjacent input events for each synchronization target field. By using a trend analysis sliding window, the editing frequency of each pair of adjacent input events in the N most recent input events of each synchronization target field is obtained and analyzed to determine the input distribution trend coefficient of each synchronization target field within the current preset time period; A synchronous analysis sliding window is used to obtain the time interval of the nearest Q-second adjacent input events for each synchronous target field and analyze it to determine the average time interval of each synchronous target field within the current preset time period. A synchronous analysis sliding window is used to obtain the number of character changes in each synchronous target field within the most recent Q seconds for analysis, and to determine the total number of character changes in each synchronous target field within the current preset time period; The average time interval, total number of character changes, and input distribution trend coefficient of each synchronization target field within the current preset time period are compared with the synchronization triggering standard to determine whether the input events of each synchronization target field within the current preset time period meet the synchronization triggering standard. If there are input events that meet the synchronization triggering criteria within the current preset time period, record the triggering time of the input events that meet the synchronization triggering criteria in the synchronization triggering sequence of the corresponding synchronization target field.

5. The multi-terminal synchronized mobile office collaboration method according to claim 1, characterized in that, The step of analyzing the continuous triggering time in the synchronization trigger sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled includes: Based on the analysis of the synchronization trigger sequence of each synchronization target field, the time interval between adjacent trigger times in the synchronization trigger sequence of each synchronization target field is determined; The synchronization triggering change trend of each synchronization target field is determined by analyzing the time interval between adjacent triggering times in the synchronization triggering sequence of each synchronization target field. Analyze the changes in trigger frequency for each synchronization target field to determine the amount of change in trigger frequency for each synchronization target field; The cumulative offset direction of the frequency change of each synchronization target field is determined by analyzing the change in the trigger frequency of each synchronization target field. The synchronous triggering change trend is the cumulative offset direction of the triggering rhythm first speeding up and then slowing down or the frequency change is positive. The synchronous target field is the synchronous target field to be adjusted. The synchronization control amplitude of the target field to be controlled is analyzed using the synchronization control amplitude calculation formula to determine the current synchronization control amplitude value of the target field to be controlled.

6. The multi-terminal synchronized mobile office collaboration method according to claim 2, characterized in that, The step of analyzing the semantic differences between adjacent input events to be synchronized in each field grouping sequence to determine conflicting intent fields and pausing the synchronization of input events for those conflicting intent fields includes: Adjacent input events to be synchronized that have a semantic difference exceeding the difference threshold are placed in the conflict processing queue. The influence range of adjacent input events to be synchronized in the conflict processing queue is judged by the logical order of the order of events to be synchronized to determine the dependency and coverage relationship between operations. The processing order is determined according to the dependency and coverage relationship, and the result of the priority judgment is output to the queue. Based on the queue priority judgment result, adjacent input events to be synchronized in the conflict processing queue that have a semantic difference exceeding the conflict confirmation threshold are judged according to priority order, and the field to which the adjacent input event to be synchronized is determined as the intention conflict field. The conflicting intent field is added to the set of field numbers to be separated, and the synchronization of fields in the set of field numbers to be separated is paused.

7. A multi-terminal synchronized mobile office collaboration system, characterized in that, The multi-terminal synchronized mobile office collaboration system includes: a behavior parameter analysis module, a weight analysis module, a change amplitude analysis module, a synchronization filtering module, a synchronization trigger analysis module, a synchronization control amplitude analysis module, and a synchronization control module; The behavior parameter analysis module is used to analyze the field editing behavior parameter set of the target document to obtain the path level name, the number of characters before and after text modification, and the type of editing operation for each field. The weight analysis module is used to analyze each field based on its path hierarchy name, the number of characters before and after text modification, and the type of editing operation, and generate a structural weight mapping sequence. The change magnitude analysis module is used to analyze the change magnitude of the target document before and after editing based on the structure weight mapping sequence, determine the quantified value of the change magnitude, and compare the quantified value of the change magnitude with the change magnitude threshold to determine whether to start synchronous filtering. The quantified value of the change magnitude is the Euclidean distance of the structure weight mapping sequence of the target document before and after editing. The synchronous filtering module is used to filter out the synchronous target fields of the target document to form a synchronous target field set when synchronous filtering is started, and generate a field-level change labeling sequence. The synchronization trigger analysis module is used to perform synchronization trigger analysis on the input events of each synchronization target field of the target document within the current preset time period according to the field-level change calibration sequence, determine the input events that meet the synchronization trigger for each synchronization target field, and generate a corresponding synchronization trigger sequence for each synchronization target field independently. The synchronization control amplitude analysis module is used to analyze the continuous triggering time in the synchronization triggering sequence of each synchronization target field to determine the synchronization control amplitude value of the current synchronization target field to be controlled. The synchronization control module is used to determine the synchronization mode configuration based on the synchronization control amplitude value of the synchronization target field to be controlled, and adjust the current synchronization execution mode of the synchronization target field to be controlled according to the synchronization mode configuration. The analysis, based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation, generates a structural weight mapping sequence, including: Based on the path hierarchy name of each field, the number of characters before and after text modification, and the type of editing operation, the path hierarchy name is split layer by layer and the path hierarchy value is analyzed. The number of characters before and after text modification is analyzed to determine the text modification length. The types of editing operations are grouped and classified according to the editing intent to determine the operation intensity level. Normalization matching is performed on the path level value, text modification length, and operation intensity level of each field to obtain the normalized path level value, normalized text modification length value, and normalized operation intensity level value of each field. The normalized path level value, normalized text modification length value, and normalized operation intensity level value of each field are then combined into a vector to obtain the field normalization parameter set. Based on the path level normalization value, text modification length normalization value, and operation intensity level normalization value of each field in the field normalization parameter set, the structural weight value of each field is analyzed using the structural weight value analysis formula to generate a structural weight mapping sequence. The formula for analyzing the structural weight values ​​is as follows: ； in, For the first The structural weight values ​​of each field, For the first The normalized matching adjustment coefficient is obtained by adjusting the first... The path hierarchy normalized value, text modification length normalized value, and operation intensity level normalized value of each field are obtained by setting proportional weights. For the first The normalized parameter at the th The mapping results in each field are mapped by mapping the path level normalized value, text modification length normalized value, or operation intensity level normalized value to the first field. The fields are obtained. For the first The normalized text modification length value for each field is obtained by dividing the text modification length by the maximum modification length in the document. For the first The normalized value of the operation intensity level for each field is obtained by dividing the operation intensity level by the preset maximum operation intensity level. The mean of the normalized values ​​of the operation intensity levels for all fields is obtained by applying the normalized values ​​of all fields. We obtain the arithmetic mean. This is a normalized parameter index, indicating the position of the normalized parameter in the field's normalized parameter set. For field indexing, it represents the first [number]th [item] in the task document. One field, This represents the total number of normalized parameters.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the multi-terminal synchronized mobile office collaboration method as described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-terminal synchronized mobile office collaboration method as described in any one of claims 1 to 6.

Images