Software modeling interaction platform

Abstract

The invention relates to the technical field of software modeling, and discloses a software modeling interaction platform which comprises a behavior data first processing module used for matrix processing of modeling behavior data; the second behavior data processing module is used for processing the matrix generated by the first behavior data processing module to obtain a new matrix; the matching engine is used for extracting historical user behavior data of which the type is the same as that of the software platform model associated with the current user behavior data from the behavior data storage module; the strategy module is used for extracting a first quantity of historical user behavior data at the front position from the historical user type data sequence, extracting a corresponding software platform model and recommending the software platform model to the current user; according to the method, the modeling result of the software model established by the user can be predicted in real time according to the modeling progress of the user, the failed or successful software platform model is recommended, and support is provided for modeling of the user.

Description

Technical Field

[0001] The present invention relates to the technical field of software modeling, and more specifically, to a software modeling interactive platform. Background Art

[0002] Software analysis modeling embodies the idea of ​​software design and builds a bridge between system requirements and system implementation. Software engineers develop software systems that meet design goals according to the models established by designers, and software maintenance and improvement are also based on software analysis models. Software modeling systems in the prior art generally only provide a basic modeling environment and cannot guide users in modeling during the modeling process. Summary of the invention

[0003] The present invention provides a software modeling interactive platform, which solves the technical problem in the related art that the software modeling system cannot guide the user to perform modeling during the user's modeling process.

[0004] According to one aspect of the present invention, there is provided a software modeling interaction platform, comprising:

[0005] A modeling behavior data collection module is used to extract the user's operations when modeling the software platform model, and record and generate modeling behavior data;

[0006] A behavior data storage module, which is used to store the user's modeling behavior data;

[0007] The first behavior data processing module is used to perform matrix processing on the modeling behavior data. The matrix processing method includes:

[0008] Establish an n*m matrix for a modeling behavior data, the number of columns m of the matrix is ​​the same as the maximum value of the operation steps of the modeling behavior data;

[0009] Calculate the number of sub-bureaus. The number of sub-bureaus is a positive integer greater than or equal to 1. The calculation formula for the maximum value of the number of sub-bureaus j is as follows Where n is the maximum value of the operation steps of the modeling behavior data;

[0010] First, the modeling behavior data is segmented according to the operation steps and the number of divisions. The number of segments is 2 j ;

[0011] Calculate the weight coefficient. The calculation formula of the weight coefficient is as follows:

[0012]

[0013] Define is the i-th weight coefficient corresponding to the j-th branch number;

[0014] It represents the average value of the assigned values ​​of the modeling behavior of the 2i-1th segment of the j+1th branch. For example, the 2i-1th segment of the j+1th branch contains three operation steps. The assigned values ​​of the three modeling behaviors corresponding to these three operation steps are 1, 1.5, and 3 respectively. Then the average value of the assigned values ​​of these three modeling behaviors is 2.25.

[0015] represents the average value of the assigned values ​​of the modeling behavior of the 2i-th segment of the j+1-th segment number;

[0016] According to the above formula, all the branch numbers are traversed to obtain the weight coefficient corresponding to each branch number. The number of weight coefficients corresponding to each branch number is 2 j-1 ;

[0017] Label the weight coefficients to each row of the matrix and generate the elements of the matrix;

[0018] The method to generate the elements of the matrix is:

[0019] For a row of the matrix, extract the weight coefficient of the row, assign 1 to the element that intersects the column corresponding to the operation step corresponding to the 2i-1th segment of the j+1th branch number and the row, assign -1 to the element that intersects the column corresponding to the operation step corresponding to the 2ith segment of the j+1th branch number and the row, and the weight coefficient of the row corresponds to the i-th weight coefficient of the j-th branch number;

[0020] The second behavior data processing module is used to process the matrix generated by the first behavior data processing module to obtain a new matrix, including: traversing all rows of the matrix and deleting rows whose corresponding weight coefficients are less than a set first weight threshold;

[0021] The recommendation start engine is used to extract the user's current user behavior data and determine whether to make recommendations to the user based on the number of operation steps in the user behavior data;

[0022] A matching engine, which is used to extract historical user behavior data of the same type as the software platform model associated with the current user behavior data from the behavior data storage module;

[0023] A sequence generation unit, which processes the current user behavior data through the first behavior data processing module and the second behavior data processing module to generate a new matrix;

[0024] The distance between historical user behavior data and current user behavior data is calculated using the following formula:

[0025]

[0026] Where A represents the new matrix generated by the current user behavior data, B represents the new matrix generated by the historical user behavior data, e is a natural constant, Pr is the maximum value of the operation step of the current user behavior data, and Hi is the maximum value of the operation step of the current user behavior data;

[0027] Sorting the extracted historical user behavior data in ascending order according to the distance from the current user behavior data to generate a historical user behavior data sequence;

[0028] A strategy module extracts a first number of historical user behavior data at the front position from the historical user type data sequence, and then determines whether the proportion of software platform model operation failures corresponding to the historical user behavior data exceeds a first proportion threshold. If so, the successfully operated software platform model corresponding to the extracted historical user behavior data is recommended to the current user. If not, the failed software platform model corresponding to the extracted historical user behavior data is recommended to the current user.

[0029] Furthermore, the user's modeling behavior data includes operation steps and modeling behaviors corresponding to the operation steps;

[0030] The operation steps are continuous values ​​starting from 0.

[0031] Furthermore, the modeling behavior data only includes the modeling behavior for modeling one software platform model.

[0032] Furthermore, when calculating the number of divisions, if the maximum value of the operation step is 2 x +1, then delete the last operation step of the modeling behavior data and the modeling behavior corresponding to the operation step.

[0033] Furthermore, one strategy of the matching engine is that the operation step is greater than a preset first operation step threshold.

[0034] Furthermore, the amount of historical user behavior data extracted by the matching engine is limited, and a set amount of historical user behavior data is randomly extracted.

[0035] Further, when the successfully running software platform model corresponding to the extracted historical user behavior data is recommended to the current user, the historical user behavior data corresponding to the successfully running software platform model extracted by the policy module is selected from the historical user behavior data sequence from front to back.

[0036] Further, when the failed software platform model corresponding to the extracted historical user behavior data is recommended to the current user, the historical user behavior data corresponding to the failed software platform model extracted by the policy module is selected from the historical user behavior data sequence from front to back.

[0037] Furthermore, the first behavior data processing module and the second behavior data processing module simultaneously process the current user behavior data and the historical user behavior data to obtain a new matrix.

[0038] Furthermore, the first behavior data processing module and the second behavior data processing module pre-process the historical user behavior data to obtain a new matrix and store it. The storage may be performed by the behavior data storage module or by a dedicated historical data storage module.

[0039] The beneficial effects of the present invention are:

[0040] The present invention can follow the modeling progress of the user to predict the modeling results of the software model established by the user in real time, and give recommendations on failed or successful software platform models to provide support for user modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a correspondence table between the value assignment and the modeling behavior of the present invention;

[0042] Figure 2 is Formula 1 of the present invention;

[0043] Figure 3 is Formula 2 of the present invention;

[0044] Figure 4 is Formula 3 of the present invention;

[0045] Figure 5 It is a module schematic diagram of a software modeling interactive platform of the present invention.

[0046] In the figure: modeling behavior data collection module 101, behavior data storage module 102, behavior data first processing module 103, behavior data second processing module 104, recommendation start engine 105, matching engine 106, sequence generation unit 107, strategy module 108. DETAILED DESCRIPTION

[0047] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and is not a limitation of the scope of protection, applicability or examples set forth in the claims. The functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Various examples may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0048] Embodiment 1

[0049] like Figure 5 As shown, a software modeling interactive platform includes:

[0050] The modeling behavior data collection module 101 is used to extract the user's operations when modeling the software platform model, and record and generate modeling behavior data;

[0051] In one embodiment of the present invention, the software platform model is a platform-independent model, and may be an independent model based on the MDA (Model Driven Architecture) architecture.

[0052] A behavior data storage module 102, which is used to store the user's modeling behavior data;

[0053] The user's modeling behavior data includes operation steps and modeling behaviors corresponding to the operation steps;

[0054] The operation steps are continuous values ​​starting from 0.

[0055] For example, if the value of the operation step is 10, the corresponding modeling behavior is to create a new node;

[0056] In the data preprocessing stage, we assign values ​​to modeling behaviors. In one embodiment of the present invention, a corresponding relationship between the assignment and the modeling behavior is provided as follows: Figure 1 The corresponding relationship between the assignment and modeling behavior is shown in the table;

[0057] Here, the above nodes, node values, connections, and connection values ​​are explained. For example, in the object model of the independent model based on the MDA architecture, the node represents the object, the node value represents the attribute and operation of the object, the connection represents the relationship between the objects, and the connection value represents the attribute of the relationship between the objects;

[0058] For another example, in the package model of an independent model based on the MDA architecture, nodes represent packages, node values ​​represent the attributes and operations of packages, and connections represent the dependencies between packages;

[0059] In one embodiment of the present invention, the modeling operation of continuous deletion is used as only one operation step, and the modeling behavior corresponding to the operation step is continuous deletion, and the corresponding assignment is 7; here, continuous deletion can be continuous deletion of nodes, relationships, etc.

[0060] In one embodiment of the present invention, the modeling behavior data only includes the modeling behavior for modeling one software platform model.

[0061] The first behavior data processing module 103 is used to perform matrix processing on the modeling behavior data. The matrix processing method includes:

[0062] Establish an n*m matrix for a modeling behavior data, the number of columns m of the matrix is ​​the same as the maximum value of the operation steps of the modeling behavior data;

[0063] Calculate the number of sub-bureaus. The number of sub-bureaus is a positive integer greater than or equal to 1. The calculation formula for the maximum value of the number of sub-bureaus j is as follows: Figure 2 As shown in Formula 1, where n is the maximum value of the operation steps of the modeling behavior data;

[0064] First, the modeling behavior data is segmented according to the operation steps and the number of divisions. The number of segments is 2 j ;

[0065] For example, when segmenting according to the number of divisions being 1, the number of segments is 1, and the segment contains the modeling behavior data of all operation steps;

[0066] For example, when the number of divisions is 2, the number of divisions is 2, and each division contains modeling behavior data of half the number of operation steps;

[0067] Calculate the weight coefficient. The calculation formula of the weight coefficient is as follows Figure 3 As shown in Formula 2, the definition is the i-th weight coefficient corresponding to the j-th branch number;

[0068] It represents the average value of the assigned values ​​of the modeling behavior of the 2i-1th segment of the j+1th branch. For example, the 2i-1th segment of the j+1th branch contains three operation steps. The assigned values ​​of the three modeling behaviors corresponding to these three operation steps are 1, 1.5, and 3 respectively. Then the average value of the assigned values ​​of these three modeling behaviors is 2.25.

[0069] represents the average value of the assigned values ​​of the modeling behavior of the 2i-th segment of the j+1-th segment number;

[0070] According to the above formula, all the branch numbers are traversed to obtain the weight coefficient corresponding to each branch number. The number of weight coefficients corresponding to each branch number is 2 j-1 ;

[0071] Label the weight coefficients to each row of the matrix and generate the elements of the matrix;

[0072] The method to generate the elements of the matrix is:

[0073] For a row of the matrix, extract the weight coefficient of the row, assign 1 to the element that intersects the column corresponding to the operation step corresponding to the 2i-1th segment of the j+1th branch number and the row, assign -1 to the element that intersects the column corresponding to the operation step corresponding to the 2ith segment of the j+1th branch number and the row, and the weight coefficient of the row corresponds to the i-th weight coefficient of the j-th branch number;

[0074] In one embodiment of the present invention, if the maximum value of the operation step is 2 x +1, then delete the last operation step of the modeling behavior data and the modeling behavior corresponding to the operation step.

[0075] In one embodiment of the present invention, n is an even number.

[0076] The second behavior data processing module 104 is used to process the matrix generated by the first behavior data processing module 103 to obtain a new matrix, including: traversing all rows of the matrix and deleting rows whose corresponding weight coefficients are less than a set first weight threshold.

[0077] The recommendation start engine 105 is used to extract the current user behavior data of the user and determine whether it is necessary to make a recommendation to the user based on the number of operation steps in the user behavior data;

[0078] One strategy for judging is: the operation step is greater than a preset first operation step threshold. The first operation step threshold may be an average value of the operation steps of the historical user behavior data associated with the software platform model associated with the current user behavior data of the user.

[0079] The matching engine 106 is used to extract historical user behavior data of the same type as the software platform model associated with the current user behavior data from the behavior data storage module 102;

[0080] The types of software platform models here refer to classifications such as package models and object models. An example of the same type is that two software platform models are both object models.

[0081] In one embodiment of the present invention, the amount of historical user behavior data extracted by the matching engine 106 is limited, and the selected strategy may be random extraction.

[0082] A sequence generation unit 107, which processes the current user behavior data through the first behavior data processing module and the second behavior data processing module to generate a new matrix;

[0083] The distance between historical user behavior data and current user behavior data is calculated using the following formula: Figure 4 As shown in Formula 3;

[0084] Where A represents the new matrix generated by the current user behavior data, B represents the new matrix generated by the historical user behavior data, e is a natural constant, Pr is the maximum value of the operation step of the current user behavior data, and Hi is the maximum value of the operation step of the current user behavior data;

[0085] Sorting the extracted historical user behavior data in ascending order according to the distance from the current user behavior data to generate a historical user behavior data sequence;

[0086] The strategy module 108 extracts a first number of historical user behavior data at the front position from the historical user type data sequence, and then determines whether the proportion of software platform model operation failures corresponding to the historical user behavior data exceeds a first proportion threshold. If it exceeds the first proportion threshold, the corresponding successfully operated software platform model in the extracted historical user behavior data is recommended to the current user. If it does not exceed the first proportion threshold, the corresponding failed software platform model in the extracted historical user behavior data is recommended to the current user.

[0087] In one embodiment of the present invention, when recommending the successfully running software platform model corresponding to the extracted historical user behavior data to the current user, the historical user behavior data corresponding to the successfully running software platform model extracted by the policy module 108 is selected from the historical user behavior data sequence from the front to the back. The number of selections is the second number, and if it is less than the second number, all are selected.

[0088] In one embodiment of the present invention, when recommending the failed software platform model corresponding to the extracted historical user behavior data to the current user, the historical user behavior data corresponding to the failed software platform model extracted by the policy module 108 is selected from the historical user behavior data sequence from the front to the back. The number of selections is the third number, and if it is less than the third number, all are selected.

[0089] In one embodiment of the present invention, the first behavior data processing module and the second behavior data processing module simultaneously process the current user behavior data and the historical user behavior data to obtain a new matrix.

[0090] In one embodiment of the present invention, the first behavior data processing module and the second behavior data processing module pre-process the historical user behavior data to obtain a new matrix and store it. The storage can be performed by the behavior data storage module 102 or by a dedicated historical data storage module. This method is convenient for reuse and reduces the amount of data processing each time the matching engine 106 runs.

[0091] The embodiment of the present embodiment is described above in conjunction with the accompanying drawings, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present embodiment, ordinary technicians in this field can also make many forms without departing from the purpose of the present embodiment and the scope of protection of the claims, all of which are within the protection of the present embodiment.

Claims

1. A software modeling interactive platform, characterized in that: include: A modeling behavior data collection module is used to extract the user's operations when modeling the software platform model, and record and generate modeling behavior data; A behavior data storage module, which is used to store the user's modeling behavior data; The first behavior data processing module is used to perform matrix processing on the modeling behavior data. The matrix processing method includes: Establish an n*m matrix for a modeling behavior data, the number of columns m of the matrix is ​​the same as the maximum value of the operation steps of the modeling behavior data; Calculate the number of sub-bureaus. The number of sub-bureaus is a positive integer greater than or equal to 1. The calculation formula for the maximum value of the number of sub-bureaus j is as follows Where n is the maximum value of the operation steps of the modeling behavior data; First, the modeling behavior data is segmented according to the operation steps and the number of divisions. The number of segments is 2 j ; Calculate the weight coefficient. The calculation formula of the weight coefficient is as follows: Define is the i-th weight coefficient corresponding to the j-th branch number; It represents the average value of the assigned values ​​of the modeling behavior of the 2i-1th segment of the j+1th branch. For example, the 2i-1th segment of the j+1th branch contains three operation steps. The assigned values ​​of the three modeling behaviors corresponding to these three operation steps are 1, 1.5, and 3 respectively. Then the average value of the assigned values ​​of these three modeling behaviors is 2.

25. represents the average value of the assigned values ​​of the modeling behavior of the 2i-th segment of the j+1-th segment number; According to the above formula, all the branch numbers are traversed to obtain the weight coefficient corresponding to each branch number. The number of weight coefficients corresponding to each branch number is 2 j-1 ; Label the weight coefficients to each row of the matrix and generate the elements of the matrix; The method to generate the elements of the matrix is: For a row of the matrix, extract the weight coefficient of the row, assign 1 to the element that intersects the column corresponding to the operation step corresponding to the 2i-1th segment of the j+1th branch number and the row, assign -1 to the element that intersects the column corresponding to the operation step corresponding to the 2ith segment of the j+1th branch number and the row, and the weight coefficient of the row corresponds to the i-th weight coefficient of the j-th branch number; The second behavior data processing module is used to process the matrix generated by the first behavior data processing module to obtain a new matrix, including: traversing all rows of the matrix and deleting rows whose corresponding weight coefficients are less than a set first weight threshold; The recommendation start engine is used for the user's current user behavior data and determines whether a recommendation is needed for the user based on the number of operation steps in the user behavior data; A matching engine, which is used to extract historical user behavior data of the same type as the software platform model associated with the current user behavior data from the behavior data storage module; A sequence generation unit, which processes the current user behavior data through the first behavior data processing module and the second behavior data processing module to generate a new matrix; The distance between historical user behavior data and current user behavior data is calculated using the following formula: Where A represents the new matrix generated by the current user behavior data, B represents the new matrix generated by the historical user behavior data, e is a natural constant, Pr is the maximum value of the operation step of the current user behavior data, and Hi is the maximum value of the operation step of the current user behavior data; Sorting the extracted historical user behavior data in ascending order according to the distance from the current user behavior data to generate a historical user behavior data sequence; A strategy module extracts a first number of historical user behavior data at the front position from the historical user type data sequence, and then determines whether the proportion of software platform model operation failures corresponding to the historical user behavior data exceeds a first proportion threshold. If so, the successfully operated software platform model corresponding to the extracted historical user behavior data is recommended to the current user. If not, the failed software platform model corresponding to the extracted historical user behavior data is recommended to the current user.

2. A software modeling interactive platform according to claim 1, characterized in that: The user's modeling behavior data includes operation steps and modeling behaviors corresponding to the operation steps; The operation steps are continuous values ​​starting from 0.

3. A software modeling interactive platform according to claim 1, characterized in that: The modeling behavior data only includes the modeling behavior for modeling one software platform model.

4. A software modeling interactive platform according to claim 1, characterized in that: When calculating the number of divisions, if the maximum value of the operation step is 2 x +1, then delete the last operation step of the modeling behavior data and the modeling behavior corresponding to the operation step.

5. A software modeling interactive platform according to claim 1, characterized in that: One strategy of the matching engine is that the operation step is greater than a preset first operation step threshold.

6. A software modeling interactive platform according to claim 1, characterized in that: The amount of historical user behavior data extracted by the matching engine is limited, and a set amount of historical user behavior data is randomly extracted.

7. A software modeling interactive platform according to claim 1, characterized in that: When recommending the successfully running software platform model corresponding to the extracted historical user behavior data to the current user, the historical user behavior data corresponding to the successfully running software platform model extracted by the policy module is selected from the historical user behavior data sequence from front to back.

8. A software modeling interactive platform according to claim 1, characterized in that: When recommending the failed software platform model corresponding to the extracted historical user behavior data to the current user, the historical user behavior data corresponding to the failed software platform model extracted by the policy module is selected from the historical user behavior data sequence from front to back.

9. A software modeling interactive platform according to claim 1, characterized in that: The first behavior data processing module and the second behavior data processing module simultaneously process the current user behavior data and the historical user behavior data to obtain a new matrix.

10. A software modeling interactive platform according to claim 1, characterized in that: The first behavior data processing module and the second behavior data processing module pre-process the historical user behavior data to obtain a new matrix and store it. The storage can be performed by the behavior data storage module or by a dedicated historical data storage module.

Images