Software function point analysis method and device, equipment, storage medium and product

By automatically analyzing the workload to be evaluated by the software project, the invalid functional processes and functional subprocess types are determined, and the problem of inefficient manual analysis in the existing technology is solved, and the comprehensive and accurate analysis of software function point is achieved.

CN119088451APending Publication Date: 2024-12-06CHINA MOBILE GROUP DESIGN INST +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411085553.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the prior art, COSMIC software analysts need to manually check and analyze line by line when performing functional point calculations, resulting in inefficiency, easy to miss and misrated reviews, and it is difficult to obtain comprehensive analysis results.

Method used

By obtaining the workload split file to be evaluated for software projects, determining the types of invalid functional processes and functional subprocesses, and performing automated analysis based on the assignment correspondence relationship, obtaining software functional point analysis results, improving analysis efficiency, and reducing the risk of missed and missed evaluations .

Benefits of technology

It realizes the automation of software function point analysis, improves analysis efficiency, reduces the risk of missed and misrated reviews, and ensures comprehensive software analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088451A_ABST
    Figure CN119088451A_ABST
Patent Text Reader

Abstract

The invention discloses a software function point analysis method and device, equipment, a storage medium and a product, and the method comprises the steps: obtaining a to-be-evaluated workload split file of a software project; the to-be-evaluated workload split file records the function process of the software project, the attribute of the function process and the attribute value of the function process, and the attribute of the function process comprises a function sub-process; then, determining an invalid function process in the to-be-evaluated workload split file according to the reference file, and determining the type of a function sub-process according to the function process corresponding to the function sub-process in the to-be-evaluated workload split file and the attribute and the attribute value of the function sub-process; and finally, based on the set assignment corresponding relation, according to the types of the invalid function processes and the function sub-processes, performing assignment on each function sub-process to obtain a software function point analysis result, so that manual analysis is not needed, the analysis efficiency is improved, the risks of missed evaluation and wrong evaluation are reduced, and comprehensive analysis of the software is realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of software analysis, and in particular to a method, device, equipment, storage medium and product for analyzing software function points. Background Art

[0002] COSMIC (Common Software Measurement International Consortium) is a method and metric for calculating software function points. It aims to provide precise, accurate and comparable evaluation results of the functional scale in the software development process by quantitatively evaluating software functions. Currently, COSMIC software analysts need to check and analyze the COSMIC functional split table line by line when performing analysis work, which is time-consuming and labor-intensive, with low analysis efficiency, and is prone to omissions and errors, making it difficult to obtain comprehensive analysis results. Summary of the invention

[0003] Based on this, the present invention provides a software function point analysis method, device, equipment, storage medium and product to solve the defects of low efficiency and incomplete analysis caused by manual analysis of software functions in the prior art.

[0004] To achieve the above object, an embodiment of the present invention provides a software function point analysis method, comprising:

[0005] Obtain a workload split file for a software project to be evaluated; wherein the workload split file to be evaluated records the functional process of the software project, the attributes of the functional process and the attribute values ​​thereof, wherein the attributes of the functional process include functional sub-processes;

[0006] Determine the invalid functional process in the workload splitting file to be evaluated according to the obtained reference file;

[0007] Determine the type of the functional sub-process according to the functional process to which the functional sub-process belongs, the attributes and attribute values ​​of the functional process in the workload splitting file to be evaluated;

[0008] Based on the set assignment correspondence relationship, the functional sub-processes of the workload split file to be evaluated are assigned values ​​according to the types of the invalid functional process and the functional sub-process, so as to obtain the software function point analysis result.

[0009] To achieve the above object, an embodiment of the present invention further provides a software function point analysis device, comprising:

[0010] The module for obtaining the file to be evaluated is used to obtain the workload split file to be evaluated of the software project; wherein the workload split file to be evaluated records the functional process of the software project, the attributes of the functional process and the attribute values ​​thereof, and the attributes of the functional process include functional sub-processes;

[0011] An invalid function determination module, used to determine the invalid function process in the workload splitting file to be evaluated according to the acquired reference file;

[0012] A functional sub-process type determination module, used to determine the type of the functional sub-process according to the functional process to which the functional sub-process belongs, the attributes and attribute values ​​of the functional process in the workload splitting file to be evaluated;

[0013] The analysis module is used to assign values ​​to the functional sub-processes of the workload split file to be evaluated based on the set assignment correspondence relationship and according to the types of the invalid functional process and the functional sub-process, so as to obtain the software function point analysis result.

[0014] To achieve the above-mentioned purpose, an embodiment of the present invention further provides a software function point analysis device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the software function point analysis method as described in any of the above-mentioned embodiments is implemented.

[0015] To achieve the above-mentioned purpose, an embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the software function point analysis method as described in any of the above-mentioned embodiments.

[0016] To achieve the above objectives, an embodiment of the present invention further provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the software function point analysis method as described in any of the above embodiments.

[0017] Compared with the prior art, the software function point analysis method, apparatus, device, storage medium and product disclosed in the embodiments of the present invention first obtain the workload split file to be evaluated of the software project; wherein the workload split file to be evaluated records the functional process of the software project, the attributes of the functional process and their attribute values, and the attributes of the functional process include functional sub-processes; then, the invalid functional process in the workload split file to be evaluated is determined according to the reference file, and the type of the functional sub-process is determined according to the functional process and its attributes and attribute values ​​corresponding to the functional sub-process in the workload split file to be evaluated; finally, based on the set assignment correspondence relationship, each functional sub-process is assigned a value according to the type of the invalid functional process and the functional sub-process, so as to obtain the software function point analysis result, without the need for manual analysis, thereby improving the analysis efficiency and reducing the risk of missed evaluation and wrong evaluation, thereby realizing a comprehensive analysis of the software. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a flowchart of a software function point analysis method provided by an embodiment of the present invention;

[0020] Figure 2 It is a schematic diagram of part of the contents of a workload splitting table to be evaluated provided by an embodiment of the present invention;

[0021] Figure 3 is a knowledge graph sample diagram provided by an embodiment of the present invention;

[0022] Figure 4 It is a structural schematic diagram of a software function point analysis device provided by an embodiment of the present invention;

[0023] Figure 5 The present invention is a schematic diagram of the structure of a software function point analysis device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] See also Figure 1 , Figure 1 1 is a flow chart of a software function point analysis method provided by an embodiment of the present invention. Specifically, the software function point analysis method comprises steps S11 to S14:

[0026] S11. Obtain a workload split file for a software project; wherein the workload split file for a software project records the functional process of the software project, the attributes of the functional process and their attribute values, and the attributes of the functional process include functional sub-processes.

[0027] For example, the workload splitting file to be evaluated may be in the form of a table, a mind map, or other forms, which are not limited here. Generally, the workload splitting file to be evaluated is generally in the form of a workload splitting table to be evaluated, see Figure 2 A schematic diagram of part of the contents of the workload split table to be evaluated, wherein the workload split table to be evaluated records parameters such as customer requirements, functional users, functional user requirements, trigger events, functional processes, sub-process descriptions corresponding to the functional processes (functional sub-processes), data movement types, data groups, data attributes, etc. The values ​​corresponding to each parameter are filled in according to the actual situation of the software project and are not limited here.

[0028] S12. Determine the invalid functional process in the workload splitting file to be evaluated according to the obtained reference file.

[0029] Specifically, a reference file is obtained in advance, and the functional process in the workload split file to be evaluated is compared with the functional process recorded in the reference file to determine the invalid functional process in the workload split file to be evaluated. The reference file may include the technical specification of the software project and / or the previous workload split file.

[0030] S13. Determine the type of the functional sub-process according to the functional process to which the functional sub-process belongs, the attributes and attribute values ​​of the functional process in the workload splitting file to be evaluated.

[0031] Specifically, see Figure 2 In the workload splitting file to be evaluated, the type of the functional sub-process is analyzed according to the data in the row where each functional sub-process is located. Exemplarily, the type of the functional sub-process is determined by comparing the reuse degree between each functional sub-process in the workload splitting file to be evaluated, and / or the type of the functional sub-process is determined by the data flow direction of the functional sub-process.

[0032] S14. Based on the set assignment correspondence relationship, assign values ​​to the functional sub-processes of the workload split file to be evaluated according to the types of the invalid functional process and the functional sub-process, to obtain software function point analysis results.

[0033] Specifically, the assignment correspondence is preset, and the types of invalid functional processes and different functional sub-processes each have a set score corresponding to them. Based on the assignment correspondence, each functional sub-process in the workload split file to be evaluated that has completed the determination of invalid functional processes and functional sub-processes is assigned a value to obtain the software function point analysis result. The values ​​assigned to each functional sub-process can be added to obtain the total score of the software function point, such as Figure 2 The function point scores (CFP) of each functional sub-process shown are 1. It can be understood that the specific value of CFP should be set according to the actual situation and is not limited to Figure 2 After obtaining the software function point analysis results, human resources and development time are reasonably allocated for the development of the software project based on the results to balance the development progress of each software project.

[0034] Compared with the prior art, the invalid functional process in the workload splitting file to be evaluated is determined according to the reference file, the type of the functional sub-process is determined according to the functional process and its attributes and attribute values ​​corresponding to the functional sub-process in the workload splitting file to be evaluated, and finally, based on the set assignment correspondence, each functional sub-process is assigned a value according to the type of the invalid functional process and the functional sub-process to obtain the software function point analysis result. No manual analysis is required, which improves the analysis efficiency and reduces the risk of missed evaluation and wrong evaluation, thereby realizing a comprehensive analysis of the software.

[0035] In one embodiment, when the reference file includes a technical specification of the software project, the process of determining the invalid functions in the workload split file to be evaluated according to the obtained reference file includes:

[0036] According to the technical requirements specification of the software project, the first invalid functional process in the workload splitting file to be evaluated is determined; wherein the similarity between the first invalid functional process and any functional process in the technical requirements specification is less than a first set similarity threshold.

[0037] Specifically, the reference documents include the technical requirements specification of the software project, which is used to define the functional processes that need to be analyzed. By comparing the functions in the technical requirements specification with those in the COSMIC workload splitting table, the functional processes in the workload file to be evaluated that do not fall in the technical requirements specification are accurately eliminated. The full Chinese name of COSMIC is Common Software Measurement International Consortium, and the full English name is Common Software Measurement International Consortium.

[0038] In one embodiment, determining the first invalid functional process in the workload splitting file to be evaluated based on the technical specification of the software project includes:

[0039] Constructing a knowledge graph with the functional process as an entity according to the workload splitting file to be evaluated;

[0040] Extract the set function process and its characteristics in the technical specification of the software project;

[0041] Call the cosine similarity calculation formula to calculate the similarity between the functional process in the workload splitting file to be evaluated and the set functional process based on the attributes and attribute values ​​of the entities in the knowledge graph, as well as the set functional process and its characteristics, so as to determine the first invalid functional process in the workload splitting file to be evaluated.

[0042] Exemplarily, taking the workload splitting file to be evaluated as the COSMIC workload splitting table as an example, the process of determining the first invalid functional process is as follows:

[0043] Get the text data of the COSMIC workload split table to be evaluated, build the knowledge graph of the COSMIC workload split table, take the functional process as the entity, and the entity has the following attributes: customer demand, functional user, functional user demand, trigger event, sub-process description (functional sub-process), data movement type, data group, data attribute, and read the corresponding data as the attribute value of the attribute. Store the three elements of "entity-attribute-attribute value" in the knowledge graph, see Figure 3 A sample knowledge graph is shown.

[0044] The Bert-BiLSTM model is used to extract entities and related features from the technical specification of the requirements, and the extracted entities and related features are matched and compared with the functional process entities and related attributes in the knowledge graph. If the corresponding data cannot be found for matching, semantic similarity is used for judgment, and cosine similarity is used for similarity judgment. If the match fails and the similarity is less than the first set similarity threshold, it is judged that the functional process is not in the submitted requirements, and the row where it is located in the COSMIC workload split table is marked in yellow, indicating that it is an "invalid" row, and the functional process is the first invalid functional process.

[0045] The formula for cosine similarity is as follows:

[0046]

[0047] Among them, Ai represents the relevant features of the entities extracted from the technical specification book (such as: 5G cell data presentation, sub-process description: 5G cell layer display function, 5G cell drop rate data display, etc.), Bi represents the relevant attributes of the entities in the COSMIC workload splitting table knowledge graph (such as: 5G cell data presentation, sub-process description: 5G cell layer display, sub-process description: 5G cell drop rate data display), and n is the total number of attributes. For example, first use the entities with the same functional process, and then calculate the similarity of the attributes of the two.

[0048] In one embodiment, when the reference file includes a previous workload split file of the software project, the process of determining the invalid functions in the workload split file to be evaluated based on the acquired reference file includes:

[0049] Determine, based on the previous workload split file of the software project, a second invalid functional process in the workload split file to be evaluated; wherein the similarity between the second invalid functional process and at least one functional process in the previous workload split file is greater than a second set similarity threshold;

[0050] and / or,

[0051] Input the designated element group data in the workload splitting file to be evaluated into a pre-trained necessity assessment model to obtain invalid elements in the workload splitting file to be evaluated; wherein, the invalid elements include invalid functional processes and / or invalid functional sub-processes. Specifically, input the designated element group data in the workload splitting file to be evaluated into a pre-trained necessity assessment model, and output the necessity assessment result of the designated element in the workload splitting file to be evaluated; wherein, when the necessity assessment result of the designated element is unnecessary, the designated element is an invalid element; the necessity assessment model uses the previous workload splitting file as the first training data, and the previous workload splitting file records the previous designated element group data, and the designated elements in the first training data are pre-labeled with the necessity type; the designated element group data includes the functional process and its attributes and attribute values, and the designated element includes the functional process; and / or, the designated element group data includes the functional sub-process and its affiliated functional process, the attributes and attribute values ​​of the affiliated functional process, and the designated element includes the functional sub-process.

[0052] Specifically, the reference files include previous workload split files of software projects. Invalid functional processes may include relatively similar functional processes in previous workload split files, and may also include some pre-defined functional sub-processes. Invalid functional processes are less difficult to develop for software projects. Therefore, when analyzing software projects, it is necessary to determine which are invalid functional processes in order to accurately analyze the functional points of the workload split files to be evaluated later.

[0053] Exemplarily, taking the workload splitting file to be evaluated as the COSMIC workload splitting table to be evaluated (i.e., the COSMIC workload splitting table of this period) and the workload splitting file of the previous period as the COSMIC workload splitting table of the previous period as an example, the confirmation process of the invalid function includes:

[0054] 1. Eliminate duplication between the functional processes of this period and those of previous periods.

[0055] 2. Analyze the necessity of the functional processes and / or functional sub-processes of this period, and eliminate unnecessary functional processes and / or functional sub-processes.

[0056] Taking the necessity analysis of functional sub-process as an example, the specific process is as follows:

[0057] Collect the previous COSMIC workload split tables as training texts, annotate the functional sub-processes in each previous split table, and mark which functional sub-processes are necessary, optional or unnecessary. The annotations of the functional sub-processes and all the attributes and attribute values ​​stored in the previous knowledge graph are used as features to input the Bert pre-training model to extract feature vectors for each functional sub-process, and the logistic regression model is used for training to obtain the necessity evaluation model.

[0058] Obtain the unmarked yellow data in the COSMIC workload splitting table to be evaluated, and use the necessity evaluation model to judge the functional sub-process to confirm whether the functional sub-process is necessary, and mark the row of the non-essential functional sub-process in the COSMIC workload splitting table to be evaluated in yellow, indicating "invalid", that is, the non-essential functional sub-process is an invalid functional sub-process. Similarly, identify whether the functional process is necessary, and mark the row of the non-essential functional process in the COSMIC workload splitting table to be evaluated in yellow, indicating "invalid".

[0059] The training process of the necessity assessment model is as follows:

[0060] S1: Extract relevant features from past functional process data. These features include: customer needs, functional users, functional user needs, triggering events, functional processes, sub-process descriptions, data movement types, data groups, data attributes and other information.

[0061] S2: The extracted features are preprocessed as the input of the Bert model to obtain the processed feature vector. Among them, BERT (Bidirectional Encoder Representations from Transformers) is a natural language processing (NLP) model based on deep learning.

[0062] S3: Use the processed feature vector as the input of the logistic regression algorithm for training to obtain the necessity assessment model.

[0063] Among them, the logistic regression model training process is as follows:

[0064] S31: Construct linear regression function: z=ω1x1+ω2x2+…+ω i x i +bwhere x i is the feature vector processed in step S2. i Indicates the corresponding weight.

[0065] S32: Construct model function: introduce activation function to map the linear regression function value to the [0,1] interval. The activation function is as follows:

[0066]

[0067] From the above process, the model function can be obtained as follows, where i represents the number of feature vectors and j represents the number of samples:

[0068]

[0069] S33: Construct probability formula: When the function value z approaches positive infinity, the value of the activation function h approaches 1, and when the function value z approaches negative infinity, the value of the activation function h approaches 0. Since the model in this paper mainly determines whether the functional process / functional sub-process is necessary, it belongs to a two-classification problem. Therefore, when the function value h(z)≥0.5, it is predicted that y=1, indicating that the functional process / functional sub-process is necessary. When the function value h(z)<0.5, it is predicted that y=0, indicating that the functional process / functional sub-process is not necessary.

[0070] Obviously, the probability corresponding to y=1 is h(z), and the probability corresponding to y=0 is 1-h(z), so the probability formula is:

[0071] P(y=1|z)=h(z)

[0072] P(y=0|z)=1-h(z)

[0073] The above formula can be integrated into a probability formula as follows:

[0074] P(y|x)=[h(z)] y ·[1-h(z)] 1-y

[0075] S34: Construct likelihood function: Since the probability obtained from each sample is independent of each other, the maximum likelihood function can be obtained as follows:

[0076]

[0077] In order to facilitate calculation and solve using gradient descent method, take the logarithm of the formula and take the negative value:

[0078]

[0079] S35: Construct loss function: Its loss function is

[0080]

[0081] Where m is the number of samples

[0082] S36: Use gradient descent method to update parameter ω: In order to improve the accuracy of prediction, this patent uses gradient descent method to solve the equation numerically, and its update formula is as follows, where a represents the learning rate.

[0083]

[0084] S37: training model: taking the feature vector processed in step S2 as the input of the logistic regression algorithm for training, and obtaining the necessity assessment model.

[0085] Further, the determining, based on the previous workload splitting file of the software project, the second invalid functional process in the workload splitting file to be evaluated includes:

[0086] Constructing a knowledge graph with the functional process as an entity according to the workload splitting file to be evaluated;

[0087] Extracting past functional processes, attributes of the past functional processes and attribute values ​​thereof from the past workload splitting file of the software project;

[0088] Call the cosine similarity calculation formula to calculate the similarity between the functional process in the workload split file to be evaluated and the past functional process based on the attributes and attribute values ​​of the entities in the knowledge graph, as well as the attributes and attribute values ​​of the past functional processes, so as to determine the second invalid functional process in the workload split file to be evaluated.

[0089] Specifically, the software project includes multiple phases of tasks, and each phase of the task corresponds to a workload split file. When evaluating the workload split file of the current phase, it is necessary to consider whether similar functional processes have appeared in previous phases of tasks (i.e., identifying the second invalid functional process). The development of functional processes that have appeared in the past can use the data of the past phase, so its development difficulty is greatly reduced. The confirmation process of the second invalid functional process is as follows:

[0090] Get the previous functional data text (past workload split file), use the Bert-BiLSTM model to extract entities from the previous functional data text, match and compare the extracted entities with the functional process entities in the COSMIC split table knowledge graph, and if the corresponding data can be found, match them. If the match fails, use semantic similarity judgment and cosine similarity for similarity judgment. If the match is successful or the similarity is greater than the second set similarity threshold, it is judged that the functional process is repeated with the previous function. Mark the row in the COSMIC workload split table in yellow to indicate an "invalid" row. Among them, Bert-BiLSTM is a natural language processing technology based on deep learning. It combines the BERT model with the bidirectional long short-term memory network (BiLSTM) to achieve more accurate text classification.

[0091] The formula for cosine similarity is as follows:

[0092]

[0093] Among them, Ci represents the attribute value of the entity extracted from the previous function (such as: heat map display, single sign-on), and Di represents the attribute value of the entity in the current COSMIC workload splitting table knowledge graph (such as: heat map display, single sign-on).

[0094] In one embodiment, determining the type of the functional sub-process according to the functional process to which the functional sub-process belongs, the attributes and attribute values ​​of the functional process in the workload splitting file to be evaluated includes:

[0095] Input the functional sub-process in the workload splitting file to be evaluated, the functional process to which the functional sub-process belongs, the attributes of the functional process and the attribute values ​​thereof, into a pre-trained cross-layer recognition model, output the cross-layer recognition result of the functional sub-process in the workload splitting file to be evaluated, and set the type of the functional sub-process for data same-layer transmission as invalid when the cross-layer recognition result is the functional sub-process; wherein the cross-layer recognition model uses the previous workload splitting file as the second training data, the previous workload splitting file records the previous functional sub-process and the functional process to which it belongs, the attributes of the functional process and the attribute values ​​thereof, and the previous functional sub-process is pre-labeled with a cross-layer situation label;

[0096] and / or,

[0097] The reuse degree between each functional sub-process in the workload split file to be evaluated is calculated to determine the type of the functional sub-process according to the reuse degree.

[0098] Specifically, in the workload split file to be evaluated, the development difficulty of the functional sub-process for data transmission on the same layer is relatively low. In addition, if the functional sub-process is reused, the overall development difficulty will be reduced accordingly, so it is necessary to accurately identify such data. For example, the process of determining whether the functional sub-process is cross-layer and cross-boundary is as follows:

[0099] Collect the previous COSMI C workload split tables as training texts, annotate the functional sub-processes in each split table, mark which functional sub-processes are cross-layer and cross-boundary, and mark the data inflow and data output of the functional sub-process (that is, the previous functional sub-processes are also pre-annotated with data inflow and outflow labels). The annotation of the functional sub-process and all the corresponding attributes and attribute values ​​stored in the knowledge graph in the past are used as feature inputs. A pre-training model such as the Bert pre-training model is used to extract feature vectors for each functional sub-process. The logistic regression algorithm is used for training to obtain a cross-layer recognition model. The specific training process is similar to the training process of the necessity assessment model.

[0100] Obtain the unmarked yellow data in the COSMI C workload splitting table to be evaluated, and make a judgment through the cross-layer identification model. If the functional sub-process does not cross the layer (that is, the data is transmitted on the same layer), the row where the functional sub-process is located in the COSMI C workload splitting table is marked in yellow to indicate "invalid".

[0101] Further, the attributes of the functional process also include the data group of the functional sub-process and the data attributes of the functional sub-process; the calculating the reuse degree between the functional sub-processes in the workload split file to be evaluated to determine the type of the functional sub-process according to the reuse degree includes:

[0102] Calculate the semantic similarity between each functional sub-process according to each functional sub-process to be analyzed in the workload split file to be evaluated and its semantic related information; wherein the semantic related information of the functional sub-process includes the functional process to which the functional sub-process belongs, the attributes of the functional process to which it belongs except the functional sub-process and their attribute values; wherein the functional sub-process to be analyzed is the functional sub-process except the functional sub-process corresponding to the invalid functional process, that is, the functional sub-process to be analyzed does not include the functional sub-process corresponding to the invalid functional process;

[0103] Calculate the data similarity between each functional sub-process according to the data group and data attribute of each functional sub-process to be analyzed in the workload split file to be evaluated;

[0104] Determine the flow similarity between each functional sub-process according to the data inflow and outflow information of each functional sub-process to be analyzed in the workload split file to be evaluated;

[0105] The reuse degree between each functional sub-process to be analyzed is determined according to the semantic similarity, the data similarity and the flow similarity.

[0106] Specifically, the degree of reuse of functional sub-processes is determined and functional duplication is eliminated. The process combines the three elements of functional sub-processes, data movement type (whether they are on the same layer and in the same boundary), and data group (data attribute) for evaluation. The semantic similarity, flow similarity and data similarity between each functional sub-process are calculated, and then the degree of reuse between each functional sub-process is analyzed.

[0107] Among them, for the workload split table to be evaluated, the data is judged to be on the same level and the same boundary: read the sub-process description column (the column where the functional sub-process is located), and compare the data inflow end identifier and data output end identifier of the current row with those of the remaining rows. If the data inflow end identifier of the current row is the same as the data inflow end identifier of one of the rows, and the data outflow end identifier of the current row is the same as its data outflow end identifier, then the data movement type of the current row is marked as duplicate.

[0108] Determine the sub-process description (functional sub-process): read the sub-process description column, perform semantic judgment on the current row and the remaining rows, calculate the similarity, and mark it as a duplicate when the similarity is greater than the set first reuse similarity threshold.

[0109] Judge the data groups and data attributes: read the columns where the data groups and data attributes are located, and perform semantic judgment on the data groups and data attributes of the current row and the remaining rows, and calculate the similarity. If the similarity is greater than the set second reuse similarity threshold, it is marked as a duplicate.

[0110] Further, determining the reuse degree between each functional sub-process to be analyzed according to the semantic similarity, the data similarity and the flow similarity includes:

[0111] When two of the semantic similarity, data similarity and flow direction similarity between two functional sub-processes to be analyzed are greater than corresponding reuse similarity thresholds, setting the type of one of the functional sub-processes to the first type;

[0112] When the semantic similarity, data similarity and flow direction similarity between two functional sub-processes to be analyzed are all greater than the corresponding reuse similarity thresholds, setting the type of one of the functional sub-processes to the second type;

[0113] When the number of semantic similarity, data similarity and flow similarity between the first functional sub-process and any other functional sub-process to be analyzed that is greater than the corresponding reuse similarity threshold is less than two, the type of the first functional sub-process is set to the third type; wherein the first functional sub-process is any one of the functional sub-processes to be analyzed.

[0114] Specifically, if all three elements are repeated, the functional sub-process is judged as "reuse" (i.e., the second type), and the row is filled with yellow. If two elements are repeated, it is judged as "reuse" (i.e., the first type), and the row is filled with red. In other cases, it is judged as "new" function points (i.e., the third type), and no color is filled.

[0115] After completing the judgment of invalid functional processes, invalid functional sub-processes, reuse degree and necessity of functional processes (functional sub-processes), assign values ​​to each functional sub-process and calculate the total CFP of the current project.

[0116] For example, based on the functional sub-process behavior granularity, the CFP column of all "reused" and "invalid" rows is modified to 0, the CFP column of all "reused" rows is modified to 0.33, and the CFP column of all "newly added" rows is modified to 1. The sum of the values ​​in the CFP column is the CFP value of the software.

[0117] It is worth noting that the CFP values ​​of various types are not limited to the above specific values, but need to meet the following rules: the CFP value of the "reused" and "invalid" rows is smaller than the CFP value of the "reused" row, and the CFP value of the "reused" row is smaller than the CFP value of the "newly added" row.

[0118] Furthermore, the previous functional sub-processes are also pre-labeled with data inflow and outflow labels, and the output of the cross-layer recognition model also includes data inflow and outflow information of the functional sub-processes.

[0119] Furthermore, the workload splitting file to be evaluated is a workload splitting table to be evaluated, and the workload splitting table includes at least one functional process, and the attributes of the functional process include at least one of customer requirements, functional users, functional user requirements, trigger time, functional sub-processes, data movement types, data groups and data attributes; the customer requirements correspond to at least one functional process, and the functional process corresponds to at least one functional sub-process. Figure 2 shown.

[0120] Compared with the prior art, the method provided in the embodiment of the present invention first obtains a workload split file to be evaluated of a software project; wherein the workload split file to be evaluated records the functional process of the software project, the attributes of the functional process and their attribute values, and the attributes of the functional process include functional sub-processes; then, the invalid functional processes in the workload split file to be evaluated are determined according to the reference file, and the types of the functional sub-processes are determined according to the functional processes and their attributes and attribute values ​​corresponding to the functional sub-processes in the workload split file to be evaluated; finally, based on the set assignment correspondence relationship, each functional sub-process is assigned a value according to the types of the invalid functional processes and the functional sub-processes to obtain the software function point analysis results, without the need for manual analysis, thereby improving the analysis efficiency and reducing the risks of missed evaluation and wrong evaluation, thereby achieving a comprehensive analysis of the software.

[0121] See also Figure 4 The embodiment of the present invention further provides a software function point analysis device, comprising:

[0122] The to-be-evaluated file acquisition module 21 is used to acquire the to-be-evaluated workload split file of the software project; wherein the to-be-evaluated workload split file records the functional process of the software project, the attributes of the functional process and their attribute values, and the attributes of the functional process include functional sub-processes;

[0123] An invalid function determination module 22, used to determine the invalid function process in the workload splitting file to be evaluated according to the acquired reference file;

[0124] A function sub-process type determination module 23 is used to determine the type of the function sub-process according to the function sub-process to which it belongs, the attributes and attribute values ​​of the function sub-process in the workload splitting file to be evaluated;

[0125] The analysis module 24 is used to assign values ​​to the functional sub-processes of the workload split file to be evaluated based on the set assignment correspondence relationship and according to the types of the invalid functional process and the functional sub-process, so as to obtain the software function point analysis result.

[0126] In one implementation, the invalid function determination module 22 is specifically configured to:

[0127] When the reference file includes the technical requirements specification of the software project, the first invalid functional process in the workload splitting file to be evaluated is determined based on the technical requirements specification of the software project; wherein the similarity between the first invalid functional process and any functional process in the technical requirements specification is less than a first set similarity threshold.

[0128] In one implementation, the invalid function determination module 22 is specifically configured to:

[0129] Constructing a knowledge graph with the functional process as an entity according to the workload splitting file to be evaluated;

[0130] Extract the set function process and its characteristics in the technical specification of the software project;

[0131] Call the cosine similarity calculation formula to calculate the similarity between the functional process in the workload splitting file to be evaluated and the set functional process based on the attributes and attribute values ​​of the entities in the knowledge graph, as well as the set functional process and its characteristics, so as to determine the first invalid functional process in the workload splitting file to be evaluated.

[0132] In one implementation, the reference file includes a past workload split file of the software project, and the invalid function determination module 22 is further used to:

[0133] Determine, based on the previous workload split file of the software project, a second invalid functional process in the workload split file to be evaluated; wherein the similarity between the second invalid functional process and at least one functional process in the previous workload split file is greater than a second set similarity threshold;

[0134] and / or,

[0135] The designated element group data in the workload split file to be evaluated is input into a pre-trained necessity evaluation model to obtain the invalid elements in the workload split file to be evaluated.

[0136] In one embodiment, determining the second invalid functional process in the workload split file to be evaluated based on the previous workload split file of the software project includes:

[0137] Constructing a knowledge graph with the functional process as an entity according to the workload splitting file to be evaluated;

[0138] Extracting past functional processes, attributes of the past functional processes and attribute values ​​thereof from the past workload splitting file of the software project;

[0139] Call the cosine similarity calculation formula to calculate the similarity between the functional process in the workload split file to be evaluated and the past functional process based on the attributes and attribute values ​​of the entities in the knowledge graph, as well as the attributes and attribute values ​​of the past functional processes, so as to determine the second invalid functional process in the workload split file to be evaluated.

[0140] In one implementation, the functional sub-process type determination module 23 is specifically used to:

[0141] Input the functional sub-process in the workload splitting file to be evaluated, the functional process to which the functional sub-process belongs, the attributes of the functional process and the attribute values ​​thereof, into a pre-trained cross-layer recognition model, output the cross-layer recognition result of the functional sub-process in the workload splitting file to be evaluated, and set the type of the functional sub-process for data same-layer transmission as invalid when the cross-layer recognition result is the functional sub-process; wherein the cross-layer recognition model uses the previous workload splitting file as the second training data, the previous workload splitting file records the previous functional sub-process and the functional process to which it belongs, the attributes of the functional process and the attribute values ​​thereof, and the previous functional sub-process is pre-labeled with a cross-layer situation label;

[0142] and / or,

[0143] The reuse degree between each functional sub-process in the workload split file to be evaluated is calculated to determine the type of the functional sub-process according to the reuse degree.

[0144] In one embodiment, the attributes of the functional process also include the data group of the functional sub-process and the data attributes of the functional sub-process; and the calculating the reuse degree between the functional sub-processes in the workload split file to be evaluated to determine the type of the functional sub-process according to the reuse degree includes:

[0145] Calculate the semantic similarity between each functional sub-process according to each functional sub-process to be analyzed in the workload split file to be evaluated and its semantic related information; wherein the semantic related information of the functional sub-process includes the functional process to which the functional sub-process belongs, the attributes of the functional process to which it belongs except the functional sub-process and their attribute values; wherein the functional sub-process to be analyzed is the functional sub-process except the functional sub-process corresponding to the invalid functional process;

[0146] Calculate the data similarity between each functional sub-process according to the data group and data attribute of each functional sub-process to be analyzed in the workload split file to be evaluated;

[0147] Determine the flow similarity between each functional sub-process according to the data inflow and outflow information of each functional sub-process to be analyzed in the workload split file to be evaluated;

[0148] The reuse degree between each functional sub-process to be analyzed is determined according to the semantic similarity, the data similarity and the flow similarity.

[0149] In one embodiment, determining the reuse degree between each functional sub-process to be analyzed according to the semantic similarity, the data similarity and the flow similarity includes:

[0150] When two of the semantic similarity, data similarity and flow direction similarity between two functional sub-processes to be analyzed are greater than corresponding reuse similarity thresholds, setting the type of one of the functional sub-processes to the first type;

[0151] When the semantic similarity, data similarity and flow direction similarity between two functional sub-processes to be analyzed are all greater than the corresponding reuse similarity thresholds, setting the type of one of the functional sub-processes to the second type;

[0152] When the number of semantic similarity, data similarity and flow similarity between the first functional sub-process and any other functional sub-process to be analyzed that is greater than the corresponding reuse similarity threshold is less than two, the type of the first functional sub-process is set to the third type; wherein the first functional sub-process is any one of the functional sub-processes to be analyzed.

[0153] In one embodiment, the previous functional sub-processes are also pre-labeled with data inflow and outflow labels, and the output of the cross-layer recognition model also includes data inflow and outflow information of the functional sub-processes.

[0154] In one embodiment, the workload splitting file to be evaluated is a workload splitting table to be evaluated, and the workload splitting table includes at least one functional process, and the attributes of the functional process include at least one of customer requirements, functional users, functional user requirements, trigger time, functional sub-processes, data movement types, data groups and data attributes; the customer requirements correspond to at least one functional process, and the functional process corresponds to at least one functional sub-process.

[0155] It is worth noting that the working principle of the software function point analysis device provided in the above embodiment can refer to the working process of the software function point analysis method provided in any of the above embodiments, which will not be described in detail here.

[0156] Compared with the prior art, the software function point analysis device provided by the embodiment of the present invention first obtains the workload splitting file to be evaluated of the software project; wherein the workload splitting file to be evaluated records the functional process of the software project, the attributes of the functional process and their attribute values, and the attributes of the functional process include functional sub-processes; then, the invalid functional process in the workload splitting file to be evaluated is determined according to the reference file, and the type of the functional sub-process is determined according to the functional process and its attributes and attribute values ​​corresponding to the functional sub-process in the workload splitting file to be evaluated; finally, based on the set assignment correspondence relationship, each functional sub-process is assigned a value according to the type of the invalid functional process and the functional sub-process, so as to obtain the software function point analysis result, without the need for manual analysis, thereby improving the analysis efficiency and reducing the risk of missed evaluation and wrong evaluation, thereby realizing a comprehensive analysis of the software.

[0157] See also Figure 5 The embodiment of the present invention further provides a software function point analysis device, comprising a processor 31, a memory 32, and a computer program stored in the memory 32 and configured to be executed by the processor 31. When the processor 31 executes the computer program, the steps in the software function point analysis method embodiment are implemented, for example Figure 1 S11 to S14 in the above description; or, the processor 31 implements the functional process of each module in the above-mentioned device embodiments when executing the computer program.

[0158] Exemplarily, the computer program may be divided into one or more modules, which are stored in the memory 32 and executed by the processor 31 to complete the present invention. The one or more modules may be a series of computer program instruction segments that can complete a specific functional process, and the instruction segments are used to describe the execution process of the computer program in the software function point analysis device. For example, the computer program may be divided into multiple modules, and the specific working process of each module may refer to the working process of the software function point analysis device described in the above embodiment, which will not be repeated here.

[0159] The software function point analysis device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The software function point analysis device may include, but is not limited to, a processor 31 and a memory 32. Those skilled in the art may understand that the software function point analysis device may also include an input and output device, a network access device, a bus, etc.

[0160] The processor 31 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 31 is the control center of the software function point analysis device, and uses various interfaces and lines to connect various parts of the entire software function point analysis device.

[0161] The memory 32 can be used to store the computer program and / or module, and the processor 31 implements various functional processes of the software function point analysis device by running or executing the computer program and / or module stored in the memory 32, and calling the data stored in the memory 32. The memory 32 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a functional process (such as an image playback functional process, etc.); the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory 32 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0162] Wherein, if the module integrated in the software function point analysis device is implemented in the form of a software function process unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor 31. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0163] An embodiment of the present invention further provides a computer program product, including a computer program / instruction, wherein when the computer program / instruction is executed by a processor, the software function point analysis method as described in any of the above embodiments is implemented.

[0164] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A software function point analysis method, characterized in that: include: Obtain a workload split file for a software project to be evaluated; wherein the workload split file to be evaluated records the functional process of the software project, the attributes of the functional process and the attribute values ​​thereof, wherein the attributes of the functional process include functional sub-processes; Determine the invalid functional process in the workload splitting file to be evaluated according to the obtained reference file; Determine the type of the functional sub-process according to the functional process to which the functional sub-process belongs, the attributes and attribute values ​​of the functional process in the workload splitting file to be evaluated; Based on the set assignment correspondence relationship, the functional sub-processes of the workload split file to be evaluated are assigned values ​​according to the types of the invalid functional process and the functional sub-process, so as to obtain the software function point analysis result.

2. The software function point analysis method according to claim 1, characterized in that: When the reference file includes a technical specification of the software project, the process of determining the invalid functions in the workload split file to be evaluated according to the obtained reference file includes: According to the technical requirements specification of the software project, determine the first invalid functional process in the workload split file to be evaluated; wherein the similarity between the first invalid functional process and any functional process in the technical requirements specification is less than a first set similarity threshold.

3. The software function point analysis method according to claim 2, characterized in that: Determining the first invalid functional process in the workload splitting file to be evaluated according to the technical specification of the software project includes: Constructing a knowledge graph with the functional process as an entity according to the workload splitting file to be evaluated; Extract the set function process and its characteristics in the technical specification of the software project; Call the cosine similarity calculation formula to calculate the similarity between the functional process in the workload splitting file to be evaluated and the set functional process based on the attributes and attribute values ​​of the entities in the knowledge graph, as well as the set functional process and its characteristics, so as to determine the first invalid functional process in the workload splitting file to be evaluated.

4. The software function point analysis method according to claim 1, characterized in that: When the reference file includes a previous workload split file of the software project, the process of determining the invalid functions in the workload split file to be evaluated according to the acquired reference file includes: Determine, based on the previous workload split file of the software project, a second invalid functional process in the workload split file to be evaluated; wherein the similarity between the second invalid functional process and at least one functional process in the previous workload split file is greater than a second set similarity threshold; and / or, The designated element group data in the workload split file to be evaluated is input into a pre-trained necessity assessment model to obtain invalid elements in the workload split file to be evaluated; wherein the invalid elements include invalid functional processes and / or invalid functional sub-processes.

5. The software function point analysis method according to claim 4, characterized in that: Determining the second invalid function process in the workload split file to be evaluated based on the previous workload split file of the software project includes: Constructing a knowledge graph with the functional process as an entity according to the workload splitting file to be evaluated; Extracting past functional processes, attributes of the past functional processes and attribute values ​​thereof from the past workload splitting file of the software project; Call the cosine similarity calculation formula to calculate the similarity between the functional process in the workload split file to be evaluated and the past functional process based on the attributes and attribute values ​​of the entities in the knowledge graph, as well as the attributes and attribute values ​​of the past functional processes, so as to determine the second invalid functional process in the workload split file to be evaluated.

6. The software function point analysis method according to claim 1, characterized in that: The determining the type of the functional sub-process according to the functional process to which the functional sub-process belongs, the attributes and attribute values ​​of the functional process in the workload splitting file to be evaluated includes: Input the function sub-process in the workload splitting file to be evaluated, the function process to which the function sub-process belongs, the attributes of the function process to which it belongs and the attribute values ​​thereof into a pre-trained cross-layer recognition model, output the cross-layer recognition result of the function sub-process in the workload splitting file to be evaluated, and set the type of the function sub-process whose cross-layer recognition result is data same-layer transmission to be invalid; and / or, The reuse degree between each functional sub-process in the workload split file to be evaluated is calculated to determine the type of the functional sub-process according to the reuse degree.

7. The software function point analysis method according to claim 6, characterized in that: The attributes of the functional process also include the data group of the functional sub-process and the data attributes of the functional sub-process; the calculating the reuse degree between the functional sub-processes in the workload split file to be evaluated to determine the type of the functional sub-process according to the reuse degree includes: Calculate the semantic similarity between each functional sub-process according to each functional sub-process to be analyzed in the workload split file to be evaluated and its semantic related information; wherein the semantic related information of the functional sub-process includes the functional process to which the functional sub-process belongs, the attributes of the functional process to which it belongs except the functional sub-process and their attribute values; wherein the functional sub-process to be analyzed is the functional sub-process except the functional sub-process corresponding to the invalid functional process; Calculate the data similarity between each functional sub-process according to the data group and data attribute of each functional sub-process to be analyzed in the workload split file to be evaluated; Determine the flow similarity between each functional sub-process according to the data inflow and outflow information of each functional sub-process to be analyzed in the workload split file to be evaluated; The reuse degree between each functional sub-process to be analyzed is determined according to the semantic similarity, the data similarity and the flow similarity.

8. The software function point analysis method according to claim 7, characterized in that: Determining the reuse degree between each functional sub-process to be analyzed according to the semantic similarity, the data similarity and the flow similarity includes: When two of the semantic similarity, data similarity and flow direction similarity between two functional sub-processes to be analyzed are greater than corresponding reuse similarity thresholds, setting the type of one of the functional sub-processes to the first type; When the semantic similarity, data similarity and flow direction similarity between two functional sub-processes to be analyzed are all greater than the corresponding reuse similarity thresholds, setting the type of one of the functional sub-processes to the second type; When the number of semantic similarity, data similarity and flow similarity between the first functional sub-process and any other functional sub-process to be analyzed that is greater than the corresponding reuse similarity threshold is less than two, the type of the first functional sub-process is set to the third type; wherein the first functional sub-process is any one of the functional sub-processes to be analyzed.

9. The software function point analysis method according to any one of claims 1 to 8, characterized in that: The workload splitting file to be evaluated is a workload splitting table to be evaluated, and the workload splitting table includes at least one functional process, and the attributes of the functional process include at least one of customer requirements, functional users, functional user requirements, trigger time, functional sub-processes, data movement types, data groups and data attributes; the customer requirements correspond to at least one functional process, and the functional process corresponds to at least one functional sub-process.

10. A software function point analysis device, characterized in that: include: The module for obtaining the file to be evaluated is used to obtain the workload split file to be evaluated of the software project; wherein the workload split file to be evaluated records the functional process of the software project, the attributes of the functional process and the attribute values ​​thereof, and the attributes of the functional process include functional sub-processes; An invalid function determination module, used to determine the invalid function process in the workload splitting file to be evaluated according to the acquired reference file; A functional sub-process type determination module, used to determine the type of the functional sub-process according to the functional process to which the functional sub-process belongs, the attributes and attribute values ​​of the functional process in the workload splitting file to be evaluated; The analysis module is used to assign values ​​to the functional sub-processes of the workload split file to be evaluated based on the set assignment correspondence relationship and according to the types of the invalid functional process and the functional sub-process, so as to obtain the software function point analysis result.

11. A software function point analysis device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the software function point analysis method as claimed in any one of claims 1 to 9 when executing the computer program.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the software function point analysis method according to any one of claims 1 to 9.

13. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the software function point analysis method as described in any one of claims 1 to 9 is implemented.