Agile coach training method, system and device based on intervention analysis and medium

Abstract

The invention provides an agile coach training method and device based on intervention analysis and a storage medium, and relates to the technical field of computers. Comprising the following steps: S1, receiving agile development data; s2, quantitative basic item calculation and abnormal point extraction are carried out, content related to a quality process improvement target in the numerical data is obtained, processing analysis is carried out, and change abnormal values between adjacent iterations are extracted by taking the iterations as dimensions; s3, performing intervention content semantic analysis, performing incremental content semantic recognition, and extracting initial and end iterations, participant roles and process improvement intervention behavior measures; and S4, training the intelligent coach model, perfecting the intelligent coach model, performing measurement item change trend estimation when target and initial quantitative characteristics are input into the model, and providing process improvement intervention behavior recommendation. According to the method, experience sharing in an enterprise is switched through in an automatic mode, and a new automatic enabling method is provided for improvement and agile transformation of an enterprise software process.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an agile coaching training method, system, device and medium based on intervention analysis. Background Art

[0002] Agile development is an iterative, flexible software development methodology that emphasizes teamwork and rapid response to changing requirements. It delivers working software in a timely manner through short iterations, guided by user feedback and continuously adjusted and optimized. Agile development emphasizes communication and transparency, encourages self-organizing teams, improves development efficiency, and meets customer needs.

[0003] Agile's characteristics include being based on project needs, focusing on practical application, and requiring the dissemination and implementation of agile coaches. Therefore, agile coaches are an inevitable role in agile implementation.

[0004] Agile operations in existing organizations require significant effort and time to train and select agile coaches. New teams undergoing agile transformations can either hire agile coaches externally or develop their own through a long process of trial and error. Agile coaches are dedicated to helping organizations and teams adopt agile development methodologies. With extensive agile knowledge and experience, they understand organizational needs and develop implementation plans. Through training, guidance, and support, agile coaches help teams master agile practices and foster a culture of collaboration, self-organization, and continuous learning. Agile coaches can also help resolve team challenges, promote the growth and development of team members, and drive organizational agile transformation to achieve higher efficiency, quality, and customer satisfaction.

[0005] Existing technical solutions do not have a method that can use the company's existing organizational process asset data and experience to replace manual labor to train intelligent agile coaching models. Instead, they simply apply the company's existing software development data to perform quality inspections to improve software development quality. Summary of the Invention

[0006] The present application provides an agile coaching training method, device and storage medium based on intervention analysis to at least solve the above technical problems existing in the prior art.

[0007] According to a first aspect of the present application, there is provided an agile coach training method based on intervention analysis, comprising the following steps: S1, agile development data reception; S2, quantitative basic item calculation and outlier extraction, obtains the content related to the quality process improvement target in the numerical data, performs processing analysis, and extracts the outliers of changes between adjacent iterations with iteration as the dimension; S3, semantic analysis of intervention content, obtains descriptions of team improvement behaviors, mainly from agile retrospectives and team regulations, from the textual data. After incremental analysis, semantic recognition of incremental content is performed to extract the initial and final iterations, the roles of participants, and process improvement intervention behaviors. S4, train the intelligent coaching model, receive the results of the semantic analysis of the intervention content as the intervention behavior, receive the results of the calculation of the quantitative basic items as the change trend, perform causal reasoning analysis, improve the intelligent coaching model, input the goals and initial quantitative features into the model, estimate the change trend of the measurement items, and provide process improvement intervention behavior recommendations.

[0008] In certain embodiments of the first aspect of the present application, in step S1, newly added agile development data is pre-classified and processed at a fixed period. The pre-classification is based on the data type and is divided into text data and numerical data. For different types of data, the processing methods are as follows: The processing method for text data is as follows: rich text type fields are archived according to timestamp, agile development stage, rich text content, and filler to form a rich text dataset R; The processing method for numerical data is: retain the original relational data structure, aggregate various types of data belonging to the agile development link by iteration, and form a numerical data set D.

[0009] In certain embodiments of the first aspect of the present application, the specific method of step S2 is as follows: Configuring quality process improvement target data sets The target scope is limited to the attributes of the agile development link To summarize and thus evaluate the goal; Quality process improvement target data set It consists of: goals, measurement items, and supporting measurement item formulas, including a series of goals and supporting multiple measurement items; Improve target datasets according to quality process Covered in Classification, obtain the required part of the numerical data set , that is, obtaining data related to quality process improvement goals; Extracting numerical datasets Basic item dataset in ,The basic item values are the basic information contained in each R&D project; Improve target data sets based on quality process Use the matching metric formula in the metric calculation to calculate the metric value for each iteration. Analyze the changes in the values of the measurement items between adjacent iterations, calculate the growth ratio, and improve the target data set for the quality process Each target in Calculate and get the growth ratio data set ; Hypothetical target have metric items, then iterate the growth ratio dataset for each metric item Detect outliers separately and calculate the median , to represent the evaluation level of the data set; calculate the first quartile , the third quartile , find the interquartile range , determine the upper edge as , the lower edge is ,in, is the experience value; The initial and final iterations associated with growth rates that are higher than the upper edge and have a positive growth rate, or lower than the lower edge and have a negative growth rate, are the associated iterations with abnormal changes in the current metric item. Therefore, a quantitative outlier dataset is formed , including attributes: project unique identifier, initial and end iterations, measurement item information, measurement item value, measurement item growth rate, single outlier expression:

[0010] in, A unique identifier for the project. For the initial To end the iteration, is the measurement item information, is the measurement item value; is the growth rate of the measurement item.

[0011] In certain embodiments of the first aspect of the present application, the basic item data includes the following attributes: project unique identifier, number of people in each role of the team, project creation time, project end time, iteration cycle, and project software amount.

[0012] In certain embodiments of the first aspect of the present application, the specific method of step S3 is as follows: First, in the rich text dataset Extract keywords from massive amounts of text and identify keywords and phrases related to process improvement plans in the text; retain the complete rich text of the identified keywords or phrases; For the selected rich texts, we compare the differences in text content at the end of each iteration according to the iteration cycle. We use the word embedding model to compare the semantic similarity between the two texts to determine the newly added content as the intervention content text object for each iteration. Get the association relationship between iteration and intervention content text objects. For the same project with the same initial and final iterations, the intervention content texts are merged into intervention content text objects. Each set of association relationships contains the following attributes: project unique identifier, initial and final iterations, and intervention content text objects. Perform information extraction tasks on each set of data in the association relationship in turn; the specific steps are: perform dependency syntax analysis on the intervention content text object, extract the participant roles and measures related to process improvement by analyzing the sentence structure; incorporate the newly extracted attributes into the association relationship to form the intervention content data set ; Single intervention content expression:

[0013] in, A unique identifier for the project. For the initial To end the iteration, Information about the roles of participants in process improvement, Provide information for process improvement initiatives.

[0014] In certain embodiments of the first aspect of the present application, in step S3, for image data in the rich text dataset, the text in the image is recognized by OCR.

[0015] In certain embodiments of the first aspect of the present application, the specific method of step S4 is as follows: Regularly complete intervention content datasets , quantitative outlier dataset , basic item dataset The three datasets are generated by processing, corresponding to the text description of process improvement intervention behavior, the quantitative expression of the more prominent changes in measurement items, and the general factors that may affect the changes in measurement items; After the data foundation is in place, the model training module conducts intelligent coaching model training, analyzes valuable process improvement intervention behaviors, and scores the impact of process improvement intervention behaviors; According to the same project unique identifier, initial and final iteration, the three data sets are connected and combined, among which the basic item data set It is based on the project to which the iteration belongs, that is, based on the unique identifier of the project, to connect and combine with the other two types of data sets; Projects are divided into three categories based on their size: small, medium, and large. The concatenated data are grouped according to project size and metric. Measures that affect the same metric for projects of the same size level are grouped together, regardless of whether the growth rate is positive or negative. The measures in the intervention content are treated as text fields. The texts of the measures in the intervention content grouped by the measurement items are compared pairwise. The texts are converted into vector representations using the word embedding method. The cosine similarity between the vectors is calculated to form the pairwise similarity angle. The measure with the highest sum of similarities within each group is calculated. Record as the first measure, record the first measure The corresponding current measurement item growth ratio is , get the first measure with Other measures of cosine similarity within the set threshold range , corresponding to the growth ratio set of the current measurement item , corresponding to the first measure Cosine similarity set , calculate the first measure The influence scoring formula for the current measurement item is:

[0016] Next, obtain the second measure with the highest similarity sum among the remaining measures in the current group , repeat the above steps to calculate the second measure Score the influence of the current measurement item , until the remaining measures are 0 or the growth ratios of the remaining measures are all negative; Through the above method, the impact score is calculated for each group, and the project scale, measures, measurement items, and scores are input into the model training module to continuously improve the construction of the intelligent coaching data model.

[0017] According to a second aspect of the present application, an agile coach training system based on intervention analysis is provided, comprising: Data receiving module, agile development of data receiving; The anomaly extraction module quantifies basic item calculations and extracts anomalies, obtains content related to quality process improvement goals from the numerical data, performs processing analysis, and extracts abnormal values of changes between adjacent iterations based on iteration dimensions; The intervention analysis module obtains descriptions of team improvement behaviors, mainly from agile retrospectives and team regulations, from the textual data. After performing incremental analysis, it performs semantic recognition of incremental content to extract the initial and final iterations, the roles of participants, and the process improvement intervention behaviors. The model training module receives the results of the semantic analysis of the intervention content as the intervention behavior, receives the results of the calculation of the quantitative basic items as the change trend, performs causal inference analysis, improves the intelligent coaching model, inputs the goals and initial quantitative features into the model, estimates the change trend of the measurement items, and provides process improvement intervention behavior recommendations.

[0018] According to a third aspect of the present application, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the method described in this application.

[0019] According to a fourth aspect of the present application, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the method described in the present application.

[0020] Compared with the prior art, this application has the following beneficial effects: Compared with the traditional application of R&D organization process asset data in a unified development platform, which only optimizes a certain development link, such as developing based on the test improvement direction to improve development efficiency, and detecting problems early through quality inspection of each iterative matter, it lacks the defect of in-depth mining of the data in the unified development platform. This application mines the data in the unified development platform in depth, and mines the influence of intervention behavior through causal analysis between data. The software R&D organization process asset data is divided into two categories: text and numerical, and processed separately. From the text data, the measures in the process improvement are extracted according to the iteration as intervention behavior. From the numerical data, the measurement items with significant numerical growth ratio changes are extracted according to the iteration as anomalies. The causal relationship mapping is established through semantic analysis and calculation of influence scores to train the intelligent coach model. This application uses a certain training to form an intelligent agile coach from the existing software R&D organization process asset data to replace manual labor, and uses an automated model to share experience within the enterprise, providing a new method of automation empowerment for enterprise software process improvement and agile transformation.

[0021] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The above and other objects, features and advantages of the exemplary embodiments of the present application will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present application are shown in an illustrative and non-limiting manner, in which: In the drawings, the same or corresponding reference numerals denote the same or corresponding parts.

[0023] Figure 1 The figure shows the overall method flow chart of the first embodiment of the present application.

[0024] Figure 2 A schematic diagram of the structure of an electronic device according to the third embodiment of the present application is shown. DETAILED DESCRIPTION

[0025] In order to make the purpose, features, and advantages of this application more obvious and easy to understand, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.

[0026] Example 1: This embodiment provides an agile coach training method based on intervention analysis, including the following steps: S1, agile development data reception; Specifically, the data pushed by the unified development platform used by the team to be analyzed is received through methods including standard real-time data interface, regular data synchronization, etc., for unified data storage.

[0027] Furthermore, in a fixed cycle, the newly added agile development data is pre-classified and processed for use in subsequent steps.

[0028] Pre-classification is specifically divided into text data and numerical data according to the type of data.

[0029] The processing methods for different types of data are as follows: The way to process text data is to archive rich text type fields according to timestamp, agile development stage, rich text content, and filler to form a rich text dataset R.

[0030] The processing method for numerical data is: retain the original relational data structure, aggregate various types of data belonging to the agile development link by iteration, and form a numerical data set D.

[0031] Among them, the attributes of agile development links are preset by the system on demand, including: personnel, requirements, code, testing, continuous integration, security, document assistance, user activity, and working hours.

[0032] S2, quantitative basic item calculation and outlier extraction, obtains the content related to the quality process improvement target in the numerical data, performs processing analysis, and extracts the outliers of changes between adjacent iterations with iteration as the dimension; The specific method is as follows: Configuring quality process improvement target data sets The target scope is limited to the attributes of the agile development link Summarize and evaluate the goals.

[0033] Quality process improvement target data set It consists of: goals, measurement items, and supporting measurement item formulas, usually including a series of goals and supporting multiple measurement items.

[0034] The following are examples of quality process improvement objectives: Goal: Improve market responsiveness. This goal can be evaluated comprehensively across requirements and continuous integration. Specifically, the shorter the requirement fulfillment cycle, the shorter the successful release cycle, the higher the probability of a successful release, and the stronger the market responsiveness. Therefore, the metrics are: requirement fulfillment cycle, version release cycle, and version success rate.

[0035] Improve target datasets according to quality process Covered in Classification, obtain the required part of the numerical data set , that is, obtaining data related to quality process improvement goals.

[0036] Extracting numerical datasets Basic item dataset in The basic item values are the basic information contained in each R&D project and are universal. Before the subsequent agile coach recommends improvement suggestions, the model data is normalized using the basic item data to eliminate recommendation deviations caused by large differences in R&D projects.

[0037] Basic item data includes the following attributes: project unique identifier, number of people in each role of the team, project creation time, project end time, iteration cycle, project software amount, etc.

[0038] Improve target data sets based on quality process Use the matching metric formula in the metric calculation to calculate the metric, and calculate the metric value of each iteration in an iteration cycle.

[0039] Analyze the changes in the values of the measurement items between adjacent iterations and calculate the growth ratio (the increase in the latter iteration compared to the previous iteration is positive, and the opposite is negative). Each target in Calculate and get the growth ratio data set .

[0040] It should be noted that due to different definitions of metric items, the change in the direction of better performance of the metric item may have a positive or negative growth rate. Therefore, data normalization is performed. If the change in the direction of better performance of the metric item is a negative growth rate, the calculated growth rate is multiplied by negative one and included in the growth ratio data set. .

[0041] Hypothetical target have metric items, then iterate the growth ratio dataset for each metric item Detect outliers separately and calculate the median (50%), to represent the evaluation level of the data set; calculate the first quartile (25%), the third quartile (75%), find the interquartile range , determine the upper edge as , the lower edge is , This is an empirical value. In agile development scenarios, 0.8 is used as a rule of thumb and can be adjusted as needed.

[0042] The initial and final iterations associated with a growth ratio that is higher than the upper edge and has a positive growth ratio, or lower than the lower edge and has a negative growth ratio, are the iterations associated with abnormal changes in the current metric.

[0043] Therefore, a quantitative outlier dataset is formed , including attributes: project unique identifier, initial and end iterations, measurement item information, measurement item value, and measurement item growth rate. Single outlier expression:

[0044] in, A unique identifier for the project. For the initial To end the iteration, is the measurement item information, is the measurement item value; is the growth rate of the measurement item.

[0045] S3, semantic analysis of intervention content, obtains descriptions of team improvement behaviors, mainly from agile retrospectives and team regulations, from the textual data. After incremental analysis, semantic recognition of incremental content is performed to extract the initial and final iterations, the roles of participants, and process improvement intervention behaviors. The specific method is as follows: In agile development, the five agile meetings are one of the core elements of agile development, and the intervention content is defined as the improvement plan formed by the team self-organization through the agile retrospective meeting.

[0046] First, in the rich text dataset Keyword extraction is performed in the massive amount of text in the dataset to identify keywords and phrases related to process improvement plans. For example, review meeting, team regulations, schedule, process, optimization, improvement, efficiency, etc. The complete rich text of the identified keywords or phrases is retained. At the same time, for the rich text dataset For image data, OCR is used to identify the text in the image. The reason why deep learning models are not used is that the current scene images are generally the required content. In order to be able to retrieve the plans that may be embedded in the image with a small probability, OCR is more efficient than other methods. Using deep learning models to recognize handwritten text in different styles takes up a lot of resources and captures few effective results.

[0047] For the rich text filtered in the previous step, we compare the differences in text content at the end of each iteration. Using a word embedding model, we compare the semantic similarity between the two texts to identify new content, which serves as the intervention content for each iteration. The word embedding algorithm we selected can eliminate minor text changes and, based on semantic understanding, ensure that newly identified intervention content is not just minor adjustments to previous intervention content.

[0048] Therefore, we obtain the association relationship between iterations and intervention content text objects. For the same project with the same initial and final iterations, the intervention content text is merged into the intervention content text object. Therefore, each set of association relationships contains the following attributes: project unique identifier, initial and final iterations, and intervention content text object.

[0049] The information extraction task is performed on each set of data in the association relationship. The specific steps are: Dependency syntactic analysis is performed on the intervention content text object. By analyzing the sentence structure, the participant roles (multiple) and measures (multiple) related to the process improvement are extracted. The newly extracted attributes are incorporated into the association relationship to form the intervention content data set. Individual intervention content expression:

[0050] in, A unique identifier for the project. For the initial To end the iteration, Information about the roles of participants in process improvement, Provide information for process improvement initiatives. S4, train the intelligent coaching model, receive the results of the semantic analysis of the intervention content as the intervention behavior, receive the results of the calculation of the quantitative basic items as the change trend, perform causal reasoning analysis, improve the intelligent coaching model, input the goals and initial quantitative features into the model, estimate the change trend of the measurement items, and provide process improvement intervention behavior recommendations.

[0051] The specific method is as follows: The improvement plan formed by the self-organization of the retrospective team is the "cause", and the growth rate trend of the measurement items between iterations is the "effect", and a cause-effect mapping is established.

[0052] Regularly complete intervention content datasets , quantitative outlier dataset , basic item dataset The three data sets correspond to the text description of process improvement intervention behavior, the quantitative expression of the more prominent changes in measurement items, and the general factors that may affect the changes in measurement items.

[0053] After having the data foundation, the model training module conducts intelligent coaching model training, that is, to achieve changes in each measurement item towards better performance, analyze potentially valuable process improvement intervention behaviors, and score the impact of process improvement intervention behaviors.

[0054] According to the same project unique identifier, initial and final iteration, the three data sets are connected and combined, among which the basic item data set It is based on the project to which the iteration belongs, that is, based on the unique identifier of the project, to connect and combine with the other two types of data sets.

[0055] Projects can be divided into three categories according to their size: small, medium, and large. The connected and combined data can be grouped according to project size and measurement items. Measures that affect the same measurement item of projects of the same size level are put into one group regardless of whether the growth rate is positive or negative.

[0056] The measures in the intervention content are text fields. Since they are entered by the agile team, there are cases where the measures are the same but the expressions are different. Therefore, the texts of the intervention measures in the intervention content grouped by the measurement items are compared pairwise. The texts are converted into vector representations using the word embedding method. The cosine similarity between the vectors is calculated to form the pairwise similarity angle, which is in the range of Calculate the measure with the highest sum of similarities within each group (Process improvement intervention behavior) is recorded as the first measure and the first measure is recorded The corresponding current measurement item growth ratio is , get the first measure with The cosine similarity is within the set threshold range, preferably Other measures within the collection , which correspond to the growth ratio set of the current measurement items , they correspond to the first measure Cosine similarity set , calculate the first measure The influence scoring formula for the current measurement item is:

[0057] Next, obtain the second measure with the highest similarity sum among the remaining measures in the current group (Process improvement intervention behavior), repeat the above steps to calculate the second measure Score the influence of the current measurement item , until the remaining measures are 0 or the growth ratios of the remaining measures are all negative.

[0058] Through the above method, the impact score is calculated for each group, and the project scale, measures, measurement items, and scores are input into the model training module to continuously improve the construction of the intelligent coaching data model.

[0059] After the intelligent coaching data model is improved, intelligent coaching guidance can be carried out. The following two scenarios are provided for further explanation.

[0060] Scenario 1: When a new team in an organization transitions to agile R&D, the system proposed in this invention receives the team's daily work and team organization data on the unified development platform in step S1, and requires the team to enter quality process improvement goals. The feature extraction module of this invention analyzes the metrics of interest in the quality process improvement goals and automatically calculates the iteration status according to the required metrics based on the team's daily work and team organization data. It then assists the new team in its agile transformation by recommending high-impact process improvement initiatives from teams of similar size and providing high-impact improvement initiatives from teams of other sizes for reference. In this way, agile coaches are freed from complex manual intervention and the experience of an organization's agile transformation is automatically analyzed and transformed into intelligent coaching for subsequent generations.

[0061] Scenario 2: When an agile team develops team regulations or improvement measures, the system proposed in this invention can directly summarize the historical agile team's experience measures into quantitative measurement items to estimate performance. This helps the team effectively estimate the improvement effect when formulating improvement measures, enhances the team's improvement motivation, and accurately and reasonably predicts the beneficial and harmful effects of a process improvement intervention behavior, helping the team to better avoid risks.

[0062] This embodiment also provides the following expanded technical solutions.

[0063] Expanded Technical Solution 1: A process improvement intervention affects not only the first iteration after the improvement but also subsequent iterations. Therefore, step S2 analyzes the changes in the metric values between adjacent iterations. This can be further optimized to analyze the changes in the metric values between consecutive iterations and extract quantitative outliers. This can help eliminate growth rate deviations caused by inadequate initial implementation of the intervention and the lack of habit formation.

[0064] Expanded Technical Solution 2: Agile coaching not only assists teams in continuous improvement but also helps them avoid misbehavior. Therefore, when training the intelligent coaching model in step S4, the original goal was to analyze potentially valuable process improvement interventions to achieve positive changes in each metric. This can now be expanded to record interventions that could potentially lead to negative changes in metric performance. This can be used to alert other teams to similar behaviors as soon as possible to mitigate risk.

[0065] Specifically: for each process improvement initiative with a negative numerical growth ratio for each metric item, the score for the negative impact on the metric item is calculated according to the method of step S4.

[0066] Expanded Technical Solution 3: Interventions are often not one-time events, but rather continuous improvements. Therefore, interventions and scores recorded in the intelligent coaching model need to be continuously refined as agile operations deepen. Establishing an evolutionary relationship diagram for interventions can help continuously improve the intelligent coaching model.

[0067] Specifically: Targeted tracking uses team data corresponding to intervention behaviors, analyzes the reasons for the continuous changes in the values of the same measurement items of these teams, and incorporates the intervention behaviors, growth ratios, and impact scores that cause subsequent changes into the original evolutionary relationship of the intervention behaviors, providing a reference basis for other teams.

[0068] Example 2: This second embodiment provides an agile coaching training system based on intervention analysis, including: Data receiving module, agile development of data receiving; The anomaly extraction module quantifies basic item calculations and extracts anomalies, obtains content related to quality process improvement goals from the numerical data, performs processing analysis, and extracts abnormal values of changes between adjacent iterations based on iteration dimensions; The intervention analysis module obtains descriptions of team improvement behaviors, mainly from agile retrospectives and team regulations, from the textual data. After performing incremental analysis, it performs semantic recognition of incremental content to extract the initial and final iterations, the roles of participants, and the process improvement intervention behaviors. The model training module receives the results of the semantic analysis of the intervention content as the intervention behavior, receives the results of the calculation of the quantitative basic items as the change trend, performs causal inference analysis, improves the intelligent coaching model, inputs the goals and initial quantitative features into the model, estimates the change trend of the measurement items, and provides process improvement intervention behavior recommendations.

[0069] The specific methods of each module are the same as those of each step in Example 1 and will not be repeated here.

[0070] Example 3: This third embodiment also provides an electronic device and a readable storage medium.

[0071] Figure 2 A schematic block diagram of an example electronic device that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0072] like Figure 2 As shown, the device includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). The RAM can also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. An input / output (I / O) interface is also connected to the bus.

[0073] Many components in a device are connected to the I / O interface, including: input units, such as a keyboard and mouse; output units, such as various types of displays and speakers; storage units, such as magnetic disks and optical disks; and communication units, such as network cards, modems, and wireless communication transceivers. The communication unit allows the device to exchange information / data with other devices via computer networks such as the Internet and / or various telecommunication networks.

[0074] The computing unit can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of computing units include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit executes the various methods and processes described above, such as the agile coaching method based on intervention analysis described in Example 1. For example, in some embodiments, the agile coaching method based on intervention analysis can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program can be loaded and / or installed onto a device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the computing unit, one or more steps of the agile coaching method based on intervention analysis described above can be performed. Alternatively, in other embodiments, the computing unit can be configured to execute the agile coaching method based on intervention analysis through any other suitable means (e.g., via firmware).

[0075] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0076] The program code for implementing the methods of the present application can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow charts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0077] In the context of this application, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or apparatus. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0078] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0079] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0080] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0081] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of this application can be achieved. This is not limited herein.

[0082] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0083] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An agile coaching training method based on intervention analysis, characterized in that: The following steps are involved: S1, agile development data reception; S2, quantitative basic item calculation and outlier extraction, obtains the content related to the quality process improvement target in the numerical data, performs processing analysis, and extracts the outliers of changes between adjacent iterations with iteration as the dimension; S3, semantic analysis of intervention content, obtains descriptions of team improvement behaviors, mainly from agile retrospectives and team regulations, from the textual data. After incremental analysis, semantic recognition of incremental content is performed to extract the initial and final iterations, the roles of participants, and process improvement intervention behaviors. S4, train the intelligent coaching model, receive the results of the semantic analysis of the intervention content as the intervention behavior, receive the results of the calculation of the quantitative basic items as the change trend, perform causal reasoning analysis, improve the intelligent coaching model, input the goals and initial quantitative features into the model, estimate the change trend of the measurement items, and provide process improvement intervention behavior recommendations.

2. The agile coaching training method based on intervention analysis according to claim 1, characterized in that: In step S1, the newly added agile development data is pre-classified and processed at a fixed period. The pre-classification is based on the data type and is divided into text data and numerical data. For different types of data, the processing methods are as follows: The processing method for text data is as follows: rich text type fields are archived according to timestamp, agile development stage, rich text content, and filler to form a rich text dataset R; The processing method for numerical data is: retain the original relational data structure, aggregate various types of data belonging to the agile development link by iteration, and form a numerical data set D.

3. The agile coaching training method based on intervention analysis according to claim 2, characterized in that: The specific method of step S2 is as follows: Configuring quality process improvement target data sets The target scope is limited to the attributes of the agile development link To summarize and thus evaluate the goal; Quality process improvement target data set It consists of: goals, measurement items, and supporting measurement item formulas, including a series of goals and supporting multiple measurement items; Improve target datasets according to quality process Covered in Classification, obtain the required part of the numerical data set , that is, obtaining data related to quality process improvement goals; Extracting numerical datasets Basic item dataset in ,The basic item values are the basic information contained in each R&D project; Improve target data sets based on quality process Use the matching metric formula in the metric calculation to calculate the metric value for each iteration. Analyze the changes in the values of the measurement items between adjacent iterations, calculate the growth ratio, and improve the target data set for the quality process Each target in Calculate and get the growth ratio data set ; Hypothetical target have metric items, then iterate the growth ratio dataset for each metric item Detect outliers separately and calculate the median , to represent the evaluation level of the dataset; Calculate the first quartile , the third quartile , find the interquartile range , determine the upper edge as , the lower edge is ,in, is the experience value; The initial and final iterations associated with growth rates that are higher than the upper edge and have a positive growth rate, or lower than the lower edge and have a negative growth rate, are the associated iterations with abnormal changes in the current metric item. Therefore, a quantitative outlier dataset is formed , including attributes: project unique identifier, initial and end iterations, measurement item information, measurement item value, measurement item growth rate, single outlier expression: in, A unique identifier for the project. For the initial To end the iteration, is the measurement item information, is the measurement item value; is the growth rate of the measurement item.

4. The agile coaching training method based on intervention analysis according to claim 3, characterized in that: The basic item data includes the following attributes: project unique identifier, number of people in each role of the team, project creation time, project end time, iteration cycle, and project software amount.

5. The agile coaching training method based on intervention analysis according to claim 3, characterized in that: The specific method of step S3 is as follows: First, in the rich text dataset Extract keywords from massive amounts of text and identify keywords and phrases related to process improvement plans in the text; Preserve the complete rich text of the recognized keywords or phrases; For the selected rich texts, we compare the differences in text content at the end of each iteration according to the iteration cycle. We use the word embedding model to compare the semantic similarity between the two texts to determine the newly added content as the intervention content text object for each iteration. Get the association relationship between iteration and intervention content text objects. For the same project with the same initial and final iterations, the intervention content texts are merged into intervention content text objects. Each set of association relationships contains the following attributes: project unique identifier, initial and final iterations, and intervention content text objects. Perform information extraction tasks on each set of data in the association relationship in turn; the specific steps are: perform dependency syntax analysis on the intervention content text object, extract the participant roles and measures related to process improvement by analyzing the sentence structure; incorporate the newly extracted attributes into the association relationship to form the intervention content data set ; Single intervention content expression: in, A unique identifier for the project. For the initial To end the iteration, Information about the roles of participants in process improvement, Provide information for process improvement initiatives.

6. The agile coaching training method based on intervention analysis according to claim 5, characterized in that: In step S3, for the image data in the rich text dataset, the text in the image is recognized by OCR.

7. The agile coaching training method based on intervention analysis according to claim 5, characterized in that: The specific method of step S4 is as follows: Regularly complete intervention content datasets , quantified outlier dataset , basic item dataset The three datasets are generated by processing, corresponding to the text description of process improvement intervention behavior, the quantitative expression of the more prominent changes in measurement items, and the general factors that may affect the changes in measurement items; After the data foundation is in place, the model training module conducts intelligent coaching model training, analyzes valuable process improvement intervention behaviors, and scores the impact of process improvement intervention behaviors; According to the same project unique identifier, initial and final iteration, the three data sets are connected and combined, among which the basic item data set It is based on the project to which the iteration belongs, that is, based on the unique identifier of the project, to connect and combine with the other two types of data sets; Projects are divided into three categories based on their size: small, medium, and large. The concatenated data are grouped according to project size and metric. Measures that affect the same metric for projects of the same size level are grouped together, regardless of whether the growth rate is positive or negative. The measures in the intervention content are treated as text fields. The texts of the measures in the intervention content grouped by the measurement items are compared pairwise. The texts are converted into vector representations using the word embedding method. The cosine similarity between the vectors is calculated to form the pairwise similarity angle. The measure with the highest sum of similarities within each group is calculated. Record as the first measure, record the first measure The corresponding current measurement item growth ratio is , get the first measure with Other measures of cosine similarity within the set threshold range , corresponding to the growth ratio set of the current measurement item , corresponding to the first measure Cosine similarity set , calculate the first measure The influence scoring formula for the current measurement item is: Next, obtain the second measure with the highest similarity sum among the remaining measures in the current group , repeat the above steps to calculate the second measure Score the influence of the current measurement item , until the remaining measures are 0 or the growth ratios of the remaining measures are all negative; Through the above method, the impact score is calculated for each group, and the project scale, measures, measurement items, and scores are input into the model training module to continuously improve the construction of the intelligent coaching data model.

8. An agile coaching training system based on intervention analysis, characterized in that: include: Data receiving module, agile development of data receiving; The anomaly extraction module quantifies basic item calculations and extracts anomalies, obtains content related to quality process improvement goals from the numerical data, performs processing analysis, and extracts abnormal values of changes between adjacent iterations based on iteration dimensions; The intervention analysis module obtains descriptions of team improvement behaviors, mainly from agile retrospectives and team regulations, from the textual data. After performing incremental analysis, it performs semantic recognition of incremental content to extract the initial and final iterations, the roles of participants, and the process improvement intervention behaviors. The model training module receives the results of the semantic analysis of the intervention content as the intervention behavior, receives the results of the calculation of the quantitative basic items as the change trend, performs causal inference analysis, improves the intelligent coaching model, and when the goals and initial quantitative features are input into the model, estimates the change trend of the measurement items and provides process improvement intervention behavior recommendations.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 8.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 8.

Images