System and method for monitoring feature iterations of software products

Abstract

This application discloses a functional iteration monitoring system and method for software products. The system includes: a requirement data collection module for collecting and preprocessing multi-source requirement information about the software product to obtain structured user requirements to be processed; a behavior-requirement correlation verification module for determining the authenticity score of user requirements to be processed based on a preset mapping model between user self-reported requirements and actual operational behaviors; updating user requirements to be processed with authenticity scores greater than a preset threshold to a preset requirement pool; a requirement priority decision module for calculating the priority scores of each target user requirement in the preset requirement pool and ranking them to obtain a user requirement priority list; and a feedback module for iterating the software product based on the priority order of the priority list, identifying requirement / problem data for new functions in the iteration, and feeding it back to the requirement data collection module. Using this application can save development resources and improve product iteration efficiency.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a system and method for monitoring the functional iteration of software products. Background Technology

[0002] In the field of standardized software management, enterprises typically need to continuously iterate the functionality of software products (such as collaborative intelligent operation systems) in smart parks based on customer needs to meet the ever-changing business requirements of customers in smart parks and enhance the competitiveness of products. For example, as business develops, new functional modules need to be added to the collaborative intelligent operation system to improve the business capabilities of the software product.

[0003] In related technologies, by collecting customer feedback and product manager feedback on customer-related needs, it is possible to understand customers' user experience with existing functions and their needs for new functions.

[0004] However, the current software feature iteration process faces problems such as fragmented user requirement collection, difficulty in verifying the authenticity of requirements, and subjective decision-making on requirement priority. This leads to low-value or unrealistic development requirements being included in the iteration plan, thus wasting development resources and reducing the efficiency of software product iteration. Summary of the Invention

[0005] This application provides a functional iteration monitoring system for software products. To provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. This summary is not intended as a general description, nor is it intended to identify key / important components or describe the scope of protection of these embodiments. Its sole purpose is to present some concepts in a simple form as a prelude to the detailed description that follows.

[0006] In a first aspect, embodiments of this application provide a functional iteration monitoring system for software products, the system comprising:

[0007] The module includes a requirement data collection module, a behavior-requirement correlation verification module, a requirement prioritization decision module, and a feedback module; among which,

[0008] The requirement data collection module is used to periodically collect and preprocess user operation behavior data, requirement data, various work order data and implementation survey data about software products, to obtain structured user requirements to be processed, and store them in a preset user requirement knowledge base;

[0009] The behavior-demand association verification module is used to determine the authenticity score of the user's demand to be processed based on a preset mapping model between the user's self-reported demand and the actual operation behavior; and to update the user's demand to be processed with an authenticity score greater than a preset threshold to a preset demand pool; wherein, the mapping model is established based on historical data and user behavior patterns;

[0010] The demand priority decision module is used to calculate the priority score of each target user demand in the preset demand pool, sort them by priority, and obtain a user demand priority list.

[0011] The feedback module is used to iterate the software product based on the priority order of the user needs priority list, and to collect user behavior parameters of the new functions in each iteration in real time for data analysis to identify the needs / problems in the new functions and feed them back to the requirements data collection module.

[0012] Optionally, collect and preprocess user operation behavior data, requirement data, various work order data, and implementation survey data related to the software product to obtain structured user requirements to be processed, including:

[0013] User behavior data is collected in real time through data tracking code embedded in software products;

[0014] Extract user feedback data from the customer service system via API interface;

[0015] Extract various types of work order data submitted by customers from the work order system;

[0016] Collect and conduct survey data through online questionnaire platforms or survey tools;

[0017] Data cleaning and formatting were performed on user operation behavior data, demand data, various work order data, and implementation survey data to obtain preprocessed multiple functional demand data.

[0018] Extract keywords and key phrases related to feature iteration from each feature requirement data;

[0019] Semantic analysis is performed on keywords and key phrases to identify the requirement category for each functional requirement data;

[0020] Based on the requirement category of each functional requirement data, functional requirement data of similar requirement categories are grouped together to obtain structured user requirements to be processed.

[0021] Optionally, based on a pre-defined mapping model between user-reported needs and actual operational behaviors, a authenticity score for the user's needs to be processed is determined, including:

[0022] The mapping model between user needs to be processed and preset user self-described needs is matched with the actual operation behavior to identify the functional modules to be iterated related to the user needs to be processed.

[0023] Statistics on user frequency, usage duration, and number of errors encountered in the feature modules to be iterated;

[0024] Eliminate the dimensional differences in usage frequency, usage duration, and number of errors encountered to obtain frequency score, duration score, and error reverse score;

[0025] The frequency score, duration score, and error feedback score are summed to obtain the authenticity score of the user's needs to be processed.

[0026] Optionally, the user requirements to be processed include the requirement text and the requirement category;

[0027] Matching the user needs to be processed and the preset user self-described needs with the mapping model of actual operational behavior includes:

[0028] Load a pre-defined mapping model between user-defined requirements and actual operational behaviors. The mapping model includes an associated requirement semantic vector matrix and a function directory tree.

[0029] Preprocess the requirement text to obtain a word sequence, and then convert the sequence into a requirement semantic vector of a preset dimension;

[0030] Using the demand category as the key, a search is performed in the demand semantic vector matrix to obtain the central demand semantic vector of the demand category;

[0031] Based on the functional directory tree and the semantic vector of the central requirement, the functional modules to be iterated are identified that are related to the user requirements to be processed.

[0032] Optionally, the feature directory tree includes multiple leaf nodes, each of which represents a basic feature module in the software product.

[0033] Based on the functional directory tree and the semantic vector of the central requirement, the functional modules to be iterated are identified that are related to the user requirements to be processed, including:

[0034] Traverse the functional module node information of each leaf node in the functional directory tree;

[0035] Calculate the first cosine similarity between the functional module node information of each leaf node and the central requirement semantic vector;

[0036] Recall all target leaf nodes whose first cosine similarity is greater than a preset first similarity threshold, determine the functional modules of each target leaf node, and obtain multiple candidate functional modules;

[0037] The second cosine similarity between the semantic vector of the computation center requirement and the semantic vector of each candidate functional module;

[0038] If the maximum second cosine similarity is greater than the preset second similarity threshold, the candidate functional module corresponding to the maximum second cosine similarity is taken as the functional module to be iterated related to the user's needs to be processed; the preset second similarity threshold is greater than the preset first similarity threshold.

[0039] Optionally, a mapping model between pre-defined user-defined needs and actual operational behaviors can be generated by following these steps:

[0040] Multiple historical requirement-function records within a preset historical time period are obtained and preprocessed from the preset requirement management library to obtain multiple valid historical samples; each valid historical sample consists of a pair of historical requirement text and the historical final online function module.

[0041] Semantic vector transformation is performed on the historical demand text in each valid historical sample to obtain the historical demand semantic vector;

[0042] The module name, function description, and menu path of the final online function module in each valid historical sample are concatenated into semantic text to obtain the semantic vector of the historical module.

[0043] Based on historical requirement semantic vectors and historical module semantic vectors, a requirement semantic vector matrix and a hierarchical functional directory tree are constructed.

[0044] By associating the semantic vector matrix of requirements with the hierarchical functional directory tree, a mapping model between pre-defined user-described requirements and actual operational behaviors is obtained.

[0045] Optionally, based on historical requirement semantic vectors and historical module semantic vectors, a requirement semantic vector matrix and a hierarchical functional directory tree are constructed, including:

[0046] All historical demand semantic vectors are fed into a pre-defined unsupervised clustering algorithm to converge multiple historical demand categories.

[0047] For each historical demand category, the average of all historical demand semantic vectors for that category is used as the category center vector for that category.

[0048] Each historical demand category is combined with its category center vector into key-value pairs for storage, resulting in a demand semantic vector matrix.

[0049] Read the existing menu-submenu-function hierarchy structure in the software product;

[0050] Create leaf nodes at different levels based on the menu-submenu-function hierarchy;

[0051] In the leaf nodes at different levels, the historical functional module node information of the functional modules corresponding to the leaf nodes and the historical module semantic vector corresponding to each historical final online functional module are persistently stored to obtain a hierarchical functional directory tree.

[0052] Optionally, the priority score of each target user's demand in the preset demand pool is calculated and sorted by priority to obtain a user demand priority list, including:

[0053] Determine the number of customers covered by each target user's needs in the pre-set demand pool, the degree of matching of behavioral data, and the degree of matching of strategic goals;

[0054] Based on the preset corporate strategic information and business objectives, weights are assigned to the number of customers covered by each target user's needs, the degree of matching of behavioral data, and the degree of matching with strategic objectives.

[0055] Based on the number of customers covered for each target user's needs, the degree of matching of behavioral data, the degree of matching of strategic goals, and the weights assigned to them, the priority score of each target user's needs is calculated.

[0056] Each target user's requirement is sorted according to its priority score to obtain a priority list of user requirements.

[0057] Optionally, determine the number of customers covered by each target user's needs in the preset demand pool, the degree of matching of behavioral data, and the degree of matching of strategic goals, including:

[0058] Extract the first demand category and first keyword set for each target user's demand from the preset demand pool;

[0059] Acquire historical event tracking data and historical customer service / work order / survey texts related to user behavior within a preset period for the software product;

[0060] From historical tracking data, find the first user who has made a preset number of valid clicks on the function indicated by the first demand category;

[0061] From historical customer service / work order / survey texts, find target texts that match the first keyword set, and identify the second user who submitted the target text;

[0062] The first user and the second user are deduplicated to obtain the number of customers covered by each target user's needs in the preset demand pool;

[0063] Obtain the first behavior data for each first user;

[0064] By analyzing the first behavioral data of each first user, the matching degree of behavioral data of each target user's needs in the preset needs pool is statistically determined.

[0065] Based on a pre-set corporate strategy text library and a first set of keywords, word-level matching is performed to generate the strategic target matching degree of each target user's needs in the pre-set demand pool.

[0066] Secondly, a method for monitoring the functional iteration of software products, the method comprising:

[0067] Regularly collect and preprocess user operation behavior data, requirement data, various work order data and implementation survey data about software products to obtain structured user requirements to be processed;

[0068] Based on a pre-defined mapping model between user self-reported needs and actual operational behaviors, the authenticity score of the user needs to be processed is determined; user needs with authenticity scores greater than a pre-defined threshold are updated to a pre-defined needs pool; the mapping model is established based on historical data and user behavior patterns.

[0069] Calculate the priority score of each target user's demand in the preset demand pool, sort them by priority, and obtain a user demand priority list;

[0070] Based on the priority order of the user needs priority list, the software product is iterated in terms of function, and user behavior parameters of the new functions in each iteration are collected in real time for data analysis to identify the needs / problems in the new functions. This data is then fed back to the research tool, and the process of regularly collecting and preprocessing user operation behavior data, requirement data, various work order data, and implementation research data about the software product continues.

[0071] In this embodiment, on the one hand, the requirement data collection module can periodically collect and preprocess multi-source data to generate structured user requirements to be processed. The behavior-requirement correlation verification module, based on a preset mapping model, can filter out low-value or false requirements by determining the authenticity score of the requirements. The requirement priority decision module prioritizes requirements according to quantitative indicators to ensure that high-value requirements are developed first. Through the collaborative work of the above modules, the problems of fragmented requirement collection, difficulty in verifying the authenticity of requirements, and subjectivity in requirement priority decision-making can be effectively solved, thereby improving the utilization efficiency of development resources and enhancing the efficiency and quality of product iteration. On the other hand, by adopting a preset mapping model between user self-described requirements and actual operational behaviors, this model can identify the functional modules to be iterated related to user requirements through the requirement semantic vector matrix and functional directory tree. Then, by combining the authenticity score of the requirements with the functional modules to be iterated, the authenticity of the requirements can be effectively verified, ensuring that the development team focuses on functions that are truly valuable to users, thereby improving the market competitiveness and user satisfaction of the software product.

[0072] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0073] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0074] Figure 1 This is a schematic diagram of a system architecture for a functional iteration monitoring system for software products, provided in an embodiment of this application.

[0075] Figure 2 This is a schematic flowchart illustrating a structured process for determining user needs to be processed, as provided in an embodiment of this application.

[0076] Figure 3 This is a schematic block diagram illustrating the process of determining a candidate functional module according to an embodiment of this application;

[0077] Figure 4 This is a schematic flowchart illustrating a process for generating an authenticity score, as provided in an embodiment of this application.

[0078] Figure 5 This is a schematic diagram illustrating the model construction process of a mapping model between preset user self-described needs and actual operational behaviors provided in an embodiment of this application;

[0079] Figure 6 This is a flowchart illustrating a method for monitoring the functional iteration of a software product, as provided in an embodiment of this application.

[0080] Figure 7 This is a schematic diagram of the result interface of a user requirement priority list provided in an embodiment of this application;

[0081] Figure 8 This is a flowchart illustrating a training method for a mapping model between preset user self-described needs and actual operational behaviors provided in an embodiment of this application.

[0082] Figure 9 This is a schematic diagram of a functional iteration monitoring device for software products provided in an embodiment of this application. Detailed Implementation

[0083] The following description and accompanying drawings fully illustrate specific embodiments of this application to enable those skilled in the art to practice them.

[0084] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0085] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of systems and methods consistent with some aspects of this application as detailed in the appended claims.

[0086] In the description of this application, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship.

[0087] Currently, by collecting customer feedback and product manager feedback on customer-related needs, we can understand customers' user experience with existing features and their needs for new features.

[0088] The inventors realized that the current software function iteration process faces problems such as fragmented collection of user requirements, difficulty in verifying the authenticity of requirements, and subjective decision-making on requirement priority. This leads to low-value or unrealistic development requirements being included in the iteration plan, thereby wasting development resources and reducing the efficiency of software product iteration.

[0089] To address the aforementioned technical issues, in this embodiment, on one hand, the requirement data collection module periodically collects and preprocesses multi-source data to generate structured user requirements to be processed. The behavior-requirement correlation verification module, based on a preset mapping model, filters out low-value or false requirements by determining their authenticity scores. The requirement priority decision module prioritizes requirements according to quantitative indicators to ensure high-value requirements are developed first. Through the collaborative work of these modules, the problems of fragmented requirement collection, difficulty in verifying requirement authenticity, and subjective requirement priority decisions can be effectively solved, thereby improving the utilization efficiency of development resources and enhancing the efficiency and quality of product iteration. On the other hand, by adopting a preset mapping model between user-described requirements and actual operational behaviors, this model can identify the functional modules to be iterated related to user requirements through a requirement semantic vector matrix and a functional directory tree. Then, by combining the authenticity scores of the required functional modules, the authenticity of the requirements can be effectively verified, ensuring that the development team focuses on functions that are truly valuable to users, thereby improving the market competitiveness and user satisfaction of the software product. The following describes this in detail using exemplary embodiments.

[0090] Please see Figure 1 , Figure 1This is a schematic diagram of a system architecture for a function iteration monitoring system for software products, provided in an embodiment of this application. The system includes: a requirement data collection module, a behavior-requirement correlation verification module, a requirement priority decision module, and a feedback module. The requirement data collection module is responsible for collecting requirement data from various sources, including feedback, work order systems, and market research. The collected data is processed into structured requirements for subsequent analysis and processing. The behavior-requirement correlation verification module verifies the authenticity of requirements based on user behavior data. If the authenticity score of a requirement reaches or exceeds a preset threshold, the requirement is added to a requirement pool. The requirement priority decision module prioritizes the requirements in the requirement pool. The prioritized requirements generate a priority list to guide subsequent product iterations and feature development. After a new feature is developed and launched, the feedback module collects user behavior data on the new feature, such as usage frequency and satisfaction. The collected feedback data is fed back to the requirement data collection module to evaluate the effectiveness of the new feature and provide a basis for future iterations.

[0091] In some embodiments of this application, a requirement data collection module is used to periodically collect and preprocess user operation behavior data, requirement data, various work order data, and implementation survey data related to the software product to obtain structured user requirements to be processed, which are then stored in a preset user requirement knowledge base; a behavior-requirement association verification module is used to determine the authenticity score of the user requirements to be processed based on a preset mapping model between user self-reported requirements and actual operation behavior; and to update the user requirements to be processed corresponding to authenticity scores greater than a preset threshold to a preset requirement pool; wherein, the mapping model is established based on historical data and user behavior patterns; a requirement priority decision module is used to calculate the priority score of each target user requirement in the preset requirement pool and sort them by priority to obtain a user requirement priority list; a feedback module is used to iterate the functions of the software product based on the priority order of the user requirement priority list, and to collect user behavior parameters of the new functions of each iteration in real time for data analysis to identify the requirement / problem data existing in the new functions, and to feed them back to the requirement data collection module.

[0092] User behavior data comprises various actions generated by users while using the software product, such as clicks, scrolling, and dwell time. This data reflects users' actual usage habits and preferences. Requirement data consists of user or customer requests or suggestions for software functionality improvements, which may originate from customer feedback, market research, etc. Ticket data comprises service requests or problem reports submitted by customers, containing specific questions or requirements regarding software functionality. Implementation survey data is collected during software implementation and may include user feedback on the implementation process, requirement changes, etc. Pending user requirements are collected and structured user requirements that have not yet been validated and prioritized. The mapping model is a data model used to verify the authenticity of the requirements.

[0093] In some embodiments of this application, the specific process of collecting and preprocessing user operation behavior data, requirement data, various work order data, and implementation survey data related to software products to obtain structured user requirements to be processed includes: collecting user operation behavior data in real time through data tracking code embedded in the software product; extracting user feedback requirement data from the customer service system through API interfaces; extracting various work order data submitted by customers from the work order system; collecting implementation survey data through online questionnaire platforms or survey tools; performing data cleaning and formatting on the user operation behavior data, requirement data, various work order data, and implementation survey data to obtain preprocessed multiple functional requirement data; extracting keywords and key phrases related to functional iteration from each functional requirement data; performing semantic analysis on the keywords and key phrases to identify the requirement category of each functional requirement data; and grouping functional requirement data of similar requirement categories into one category according to the requirement category of each functional requirement data to obtain structured user requirements to be processed.

[0094] In this embodiment, the collection and integration of user data through multiple channels ensures the comprehensiveness and diversity of demand information, thus providing a solid data foundation for product iteration. Simultaneously, through semantic analysis of keywords and key phrases, the system can automatically identify and classify user needs, which greatly improves the automation level of demand identification and reduces the need for manual intervention.

[0095] Specifically, the process of performing semantic analysis on keywords and key phrases to identify the requirement category of each functional requirement data point includes: selecting a suitable pre-trained language model, such as BERT, GPT, or its variants, to capture the semantic information of keywords and key phrases; fine-tuning the pre-trained language model to adapt it to semantic analysis tasks in a specific domain, such as functional requirements analysis for software products; inputting keywords and key phrases into the semantic analysis model to extract their semantic feature vectors. The semantic feature vectors represent the position of keywords and key phrases in the semantic space, thus reflecting their semantic content; providing one or more representative keywords and key phrases, along with their corresponding semantic feature vectors, for each preset requirement category; calculating the similarity between the semantic feature vector of the requirement data to be classified and the semantic feature vector of the preset category using cosine similarity or other similarity metrics; and mapping the requirement data to be classified to the most similar requirement category based on the similarity score.

[0096] For example Figure 2 As shown, Figure 2 This application provides a schematic flowchart of a structured process for determining user needs to be processed. First, multi-source data is collected, including user operation logs, customer service feedback, work order systems, and online surveys. The collected data is then integrated to form a demand data set. Next, the collected demand data set is cleaned and formatted. Keywords and key phrases related to feature iterations are extracted from the formatted data; these keywords and phrases represent the core content of user needs. A pre-trained language model (such as BERT or GPT) is fine-tuned to adapt to the semantic analysis task of a specific domain. The keywords and phrases are input into the fine-tuned model to extract their semantic feature vectors. The cosine similarity between the semantic feature vectors of the demand data to be classified and the semantic feature vectors of the preset demand categories is calculated. Based on the cosine similarity score, the demand data to be classified is mapped to the most similar demand category. Functional demand data of similar demand categories are grouped together, ultimately outputting structured demand data.

[0097] In some embodiments of this application, the specific process of determining the authenticity score of a user's demand to be processed based on a preset mapping model of user self-reported needs and actual operational behavior includes: matching the user's demand to be processed with the preset mapping model of user self-reported needs and actual operational behavior to identify the functional modules to be iterated related to the user's demand to be processed; counting the frequency of use, duration of use, and number of errors encountered by the user on the functional modules to be iterated; eliminating the difference in units of the frequency of use, duration of use, and number of errors encountered to obtain a frequency score, duration score, and error reverse score; and summing the frequency score, duration score, and error reverse score to obtain the authenticity score of the user's demand to be processed.

[0098] Among them, the functional modules to be iterated are the functional parts that need to be updated or added to the software based on user needs.

[0099] The user requirements to be processed include the requirement text and the requirement category.

[0100] Specifically, the process of matching the user needs to be processed and the preset user self-described needs with the mapping model of actual operation behavior includes: loading the preset user self-described needs and actual operation behavior mapping model, which includes an associated requirement semantic vector matrix and a function directory tree; preprocessing the requirement text to obtain a word sequence and converting the sequence into a requirement semantic vector of preset dimensions; using the requirement category as the key, searching in the requirement semantic vector matrix to obtain the central requirement semantic vector of the requirement category; and determining the functional modules to be iterated based on the function directory tree and the central requirement semantic vector.

[0101] The requirement semantic vector matrix is ​​a data structure containing semantic information for different requirement categories, represented as vectors, with each vector representing the semantic features of the corresponding requirement category. The function directory tree is a hierarchical data model representing all possible functional modules in the software product. Each node in the tree represents a functional module, used to find the corresponding functional iteration point based on requirements.

[0102] The functional directory tree includes multiple leaf nodes, each representing a basic functional module in the software product.

[0103] Specifically, the process of determining the iterative functional modules related to the user's needs based on the functional directory tree and the central requirement semantic vector includes: traversing the functional module node information of each leaf node in the functional directory tree; calculating the first cosine similarity between the functional module node information of each leaf node and the central requirement semantic vector; recalling all target leaf nodes whose first cosine similarity is greater than a preset first similarity threshold, determining the functional modules of each target leaf node, and obtaining multiple candidate functional modules; calculating the second cosine similarity between the central requirement semantic vector and the module semantic vector of each candidate functional module; when the maximum second cosine similarity is greater than a preset second similarity threshold, selecting the candidate functional module corresponding to the maximum second cosine similarity as the iterative functional module related to the user's needs; the preset second similarity threshold is greater than the preset first similarity threshold.

[0104] For example Figure 3 As shown, Figure 3This is a schematic flowchart illustrating the process of determining candidate functional modules provided in this application. First, the user's requirements to be processed are obtained, including the user's requirement text and requirement category. A requirement semantic vector matrix and a functional directory tree are loaded. The user's requirement text is preprocessed, converting it into a word sequence, and then further converted into a requirement semantic vector. Using the requirement category as the key, the corresponding central requirement semantic vector is retrieved from the requirement semantic vector matrix. The leaf nodes of the functional directory tree represent specific functional modules in the software. These nodes are traversed to find the functional modules most relevant to the user's requirements. For each leaf node in the functional directory tree, the first cosine similarity between its functional module node information and the central requirement semantic vector is calculated. The calculated first cosine similarity is checked to see if it is greater than a preset threshold. If the similarity is greater than the threshold, the leaf node is recalled as part of the candidate functional module list. If the similarity is not greater than the threshold, the node is discarded and not considered a candidate functional module.

[0105] For example Figure 4 As shown, Figure 4 This application provides a schematic flowchart of a process for generating an authenticity score. After obtaining candidate functional modules, for each candidate functional module, the second cosine similarity between the central requirement semantic vector and the module semantic vector of that module is calculated. The maximum second cosine similarity is checked to see if it exceeds another preset threshold. If the maximum second similarity exceeds the threshold, the candidate functional module is determined to be an iterative functional module related to user needs. For the determined iterative functional modules, the frequency of user use, usage duration, and number of errors encountered on that module are statistically analyzed. The statistical data is standardized to eliminate the influence of different units, resulting in a frequency score, a duration score, and an error reverse score. The frequency score, duration score, and error reverse score are summed to obtain the authenticity score of the user's needs.

[0106] In one possible implementation, the mapping model assumes that the requirement semantic vector matrix contains the central vectors for different requirement categories, and the function directory tree contains all the functional modules of the software product. The user requirement to be processed includes the requirement text "Add export function," which, after preprocessing and vectorization, yields a vector representation [0.1, 0.2, 0.3, ...]. The requirement category to be processed includes "Function Enhancement," and its central vector [0.4, 0.5, 0.1, ...] is obtained by combining the requirement semantic vector matrix. The function directory tree is traversed, and the cosine similarity between each leaf node (functional module) and the central requirement vector is calculated to find the most relevant functional module, such as "Data Export." For the "Data Export" function, the average daily usage frequency is 5 times, the average usage time is 30 seconds, and the number of errors is 0. The difference in units of usage frequency, usage time, and the number of errors is eliminated to obtain a frequency score, a duration score, and an error reverse score. The frequency score, duration score, and error reverse score are summed to obtain the authenticity score of the user requirement to be processed. Among them, the frequency score for eliminating dimensional differences The calculation expression is:

[0107]

[0108] in, This is the frequency score after logarithmic scaling. It is the natural logarithm function, used for logarithmic scaling of frequencies; It refers to the frequency of use;

[0109]

[0110] in, and These are the minimum and maximum frequency scores based on historical experience.

[0111] Among them, the duration score for eliminating dimensional differences The calculation expression is:

[0112]

[0113] in, This is the duration score after logarithmic scaling. It refers to the duration of use;

[0114]

[0115] in, and These are the minimum and maximum duration scores based on historical experience and pre-set parameters;

[0116] Among them, the error reverse score for eliminating dimensional differences The calculation expression is:

[0117]

[0118] in, It is the initial error reverse score. It is the error rate, which is obtained by calculating the ratio between the number of errors and the total number of uses;

[0119]

[0120] in, These are the minimum and maximum values ​​of the incorrect reverse score based on historical experience.

[0121] In some embodiments of this application, the specific process of generating a mapping model between preset user self-described requirements and actual operational behaviors includes: obtaining and preprocessing multiple historical requirement-function records within a preset historical time period from a preset requirement management library to obtain multiple valid historical samples; each valid historical sample is composed of a pair of historical requirement text and a historical final online function module; performing semantic vector conversion on the historical requirement text in each valid historical sample to obtain a historical requirement semantic vector; concatenating the module name, function description, and menu path of the historical final online function module in each valid historical sample into semantic text to obtain a historical module semantic vector; constructing a requirement semantic vector matrix and a hierarchical function directory tree based on the historical requirement semantic vector and the historical module semantic vector; associating the requirement semantic vector matrix and the hierarchical function directory tree to obtain a mapping model between preset user self-described requirements and actual operational behaviors.

[0122] The preset requirements management library is a storage system used to store a collection of historical requirements and function records. These records include information such as user-submitted requirements, the processing of those requirements, and the final implemented functions. The historical time period refers to the time range of historical data considered when constructing the mapping model. Historical requirements-function records are specific records of user-submitted requirements within a specific time period and how those requirements were ultimately transformed into software functions.

[0123] Specifically, the process of constructing a requirement semantic vector matrix and a hierarchical functional directory tree based on historical requirement semantic vectors and historical module semantic vectors includes: feeding all historical requirement semantic vectors into a pre-defined unsupervised clustering algorithm to converge multiple historical requirement categories; for each historical requirement category, averaging all historical requirement semantic vectors of that category as the category center vector; combining each historical requirement category with its category center vector into key-value pairs for storage to obtain the requirement semantic vector matrix; reading the original menu-submenu-functional hierarchy structure of the software product; creating leaf nodes at different levels based on the menu-submenu-functional hierarchy structure; and persistently storing the historical functional module node information of the corresponding functional module and the historical module semantic vector corresponding to each historically launched functional module in the leaf nodes at different levels to obtain the hierarchical functional directory tree.

[0124] For example Figure 5 As shown, Figure 5 This is a schematic diagram illustrating the model construction process of a pre-defined mapping model between user-defined needs and actual operational behaviors, as provided in this application. The data source layer contains a requirement management library, which stores all requirement-related data, including historical requirements and software function records. In the model construction layer, historical data is extracted from the requirement management library. This data includes user requirement texts and the final deployed functional modules. The extracted historical requirement texts are cleaned to remove noise and irrelevant information in preparation for semantic analysis. The cleaned requirement texts are converted into semantic vectors, and an unsupervised clustering algorithm is used to cluster these semantic vectors to discover similarities between requirements and form different requirement categories. For each requirement category, the average of all requirement semantic vectors is calculated to obtain the center vector for that category. Each requirement category and its corresponding center vector are stored in a matrix to form a requirement semantic vector matrix. Metadata such as the name, description, and menu path of the functional modules are concatenated and converted into module semantic vectors. Based on the functional structure of the software product, a hierarchical functional directory tree is constructed. In the functional directory tree, leaf nodes are created for each functional module, and the semantic vectors of the functional modules are stored in these nodes. The module semantic vectors are injected into the corresponding nodes of the functional directory tree. By linking the requirement semantic vector matrix and the functional directory tree, a final mapping model is constructed, which can map the user's self-described requirements to the corresponding functional modules.

[0125] In some embodiments of this application, the specific process of calculating the priority score of each target user need in the preset demand pool and sorting them by priority to obtain a user demand priority list includes: determining the number of customers covered, behavioral data matching degree, and strategic goal matching degree of each target user need in the preset demand pool; assigning weights to the number of customers covered, behavioral data matching degree, and strategic goal matching degree of each target user need based on preset enterprise strategic information and business goal information; calculating the priority score of each target user need based on the number of customers covered, behavioral data matching degree, strategic goal matching degree, and their assigned weights; and sorting each target user need according to the size of its priority score to obtain a user demand priority list.

[0126] Specifically, the process of determining the number of customers covered, behavioral data matching degree, and strategic goal matching degree for each target user's needs in the preset demand pool includes: extracting the first demand category and first keyword set for each target user's needs in the preset demand pool; obtaining historical event tracking data and historical customer service / work order / survey texts recording user behavior within a preset period for the software product; finding the first user from the historical event tracking data who has a preset number of valid clicks on the function indicated by the first demand category; finding the target text matching the first keyword set from the historical customer service / work order / survey texts, and identifying the second user who submitted the target text; deduplicating the first user and the second user to obtain the number of customers covered for each target user's needs in the preset demand pool; obtaining the first behavioral data for each first user; calculating the behavioral data matching degree for each target user's needs in the preset demand pool based on the first behavioral data of each first user; and performing word-level matching based on the preset enterprise strategic text library and the first keyword set to generate the strategic goal matching degree for each target user's needs in the preset demand pool.

[0127] Among these, the number of covered customers refers to the number of different customers who mention or pay attention to a specific need. Behavioral data matching degree is the degree to which a user's actual behavioral data matches a specific need, reflecting the relevance of user behavior to the need. Strategic goal matching degree is the degree of consistency between the need and the company's strategic goals, reflecting the contribution of the need to the company's strategic goals. The priority score is calculated based on the number of covered customers, behavioral data matching degree, strategic goal matching degree, and their weights, and is used to measure the importance of the need. The user need priority list is a list obtained by sorting all target user needs according to their priority scores, used to guide feature iteration and resource allocation. The first need category is the main need category extracted from the need text. The first keyword set is the keywords extracted from the need text, used to describe the core content of the need. Historical tracking data refers to data preset in the software to record user behavior. Historical customer service / work order / survey text is text data collected from customer service records, work order systems, and market research. The first user refers to the user related to a specific need category in the historical tracking data. The second user refers to the user related to a specific keyword set in the customer service / work order / survey text. The default corporate strategy text library is a text library that stores corporate strategic goals and related information.

[0128] Specifically, the statistical process for behavioral data matching is as follows: First, behavioral data includes user clickstream, page dwell time, and function usage frequency within the software product. Define behavioral metrics related to the target user's needs. For example, if the need is to add a report export function, relevant metrics include the user's dwell time on the report page and the number of times the export button is clicked. For each target user need, calculate the statistical values ​​(such as mean, median, standard deviation, etc.) of its relevant behavioral metrics. Quantify the correlation between these statistical values ​​and the need to obtain the behavioral data matching degree. The higher the matching degree, the stronger the correlation between user behavior and the need.

[0129] Specifically, the process of generating strategic goal alignment scores includes: matching extracted keywords with words in a strategic text library. This can be achieved using natural language processing techniques such as simple string matching, stemming, and synonym recognition. For each keyword, its frequency of occurrence or relevance score in the strategic text library is calculated. The scores of all keywords are then aggregated to obtain the strategic goal alignment score for each target user's needs. A higher alignment score indicates a stronger consistency between the user's needs and the company's strategic goals.

[0130] In this embodiment, on the one hand, the requirement data collection module can periodically collect and preprocess multi-source data to generate structured user requirements to be processed. The behavior-requirement correlation verification module, based on a preset mapping model, can filter out low-value or false requirements by determining the authenticity score of the requirements. The requirement priority decision module prioritizes requirements according to quantitative indicators to ensure that high-value requirements are developed first. Through the collaborative work of the above modules, the problems of fragmented requirement collection, difficulty in verifying the authenticity of requirements, and subjectivity in requirement priority decision-making can be effectively solved, thereby improving the utilization efficiency of development resources and enhancing the efficiency and quality of product iteration. On the other hand, by adopting a preset mapping model between user self-described requirements and actual operational behaviors, this model can identify the functional modules to be iterated related to user requirements through the requirement semantic vector matrix and functional directory tree. Then, by combining the authenticity score of the requirements with the functional modules to be iterated, the authenticity of the requirements can be effectively verified, ensuring that the development team focuses on functions that are truly valuable to users, thereby improving the market competitiveness and user satisfaction of the software product.

[0131] Please see Figure 6 This document provides a flowchart illustrating a method for monitoring the functional iteration of a software product, as described in an embodiment of this application. Figure 6 As shown, the detection method in this application embodiment may include the following steps:

[0132] S101: Regularly collect and preprocess user operation behavior data, requirement data, various work order data and implementation survey data about software products to obtain structured user requirements to be processed;

[0133] S102, Based on a preset mapping model of user self-reported needs and actual operational behaviors, determine the authenticity score of the user needs to be processed; update the user needs to be processed corresponding to authenticity scores greater than a preset threshold to a preset needs pool; wherein, the mapping model is established based on historical data and user behavior patterns;

[0134] S103, calculate the priority score of each target user's demand in the preset demand pool, sort the priorities, and obtain a user demand priority list;

[0135] The result interface of the user demand priority list is as follows: Figure 7 As shown.

[0136] S104. Based on the priority order of the user demand priority list, the software product is iterated in terms of function, and user behavior parameters of the new functions in each iteration are collected in real time for data analysis to identify the demand / problem data in the new functions. The data is fed back to the research tool, and the steps of regularly collecting and preprocessing user operation behavior data, demand data, various work order data and implementation research data about the software product are continued.

[0137] In this embodiment, on the one hand, the requirement data collection module can periodically collect and preprocess multi-source data to generate structured user requirements to be processed. The behavior-requirement correlation verification module, based on a preset mapping model, can filter out low-value or false requirements by determining the authenticity score of the requirements. The requirement priority decision module prioritizes requirements according to quantitative indicators to ensure that high-value requirements are developed first. Through the collaborative work of the above modules, the problems of fragmented requirement collection, difficulty in verifying the authenticity of requirements, and subjectivity in requirement priority decision-making can be effectively solved, thereby improving the utilization efficiency of development resources and enhancing the efficiency and quality of product iteration. On the other hand, by adopting a preset mapping model between user self-described requirements and actual operational behaviors, this model can identify the functional modules to be iterated related to user requirements through the requirement semantic vector matrix and functional directory tree. Then, by combining the authenticity score of the requirements with the functional modules to be iterated, the authenticity of the requirements can be effectively verified, ensuring that the development team focuses on functions that are truly valuable to users, thereby improving the market competitiveness and user satisfaction of the software product.

[0138] Please see Figure 8 This document provides a flowchart illustrating a training method for a mapping model between pre-defined user-defined needs and actual operational behaviors, as described in an embodiment of this application. Figure 8 As shown, the process includes the following steps:

[0139] S201: Obtain and preprocess multiple historical requirement-function records within a preset historical time period from the preset requirement management library to obtain multiple valid historical samples; each valid historical sample consists of a pair of historical requirement text and historical final online function module.

[0140] S202, perform semantic vector transformation on the historical demand text in each valid historical sample to obtain the historical demand semantic vector;

[0141] S203, concatenate the module name, function description, and menu path of the last online function module in each valid historical sample into semantic text to obtain the semantic vector of the historical module;

[0142] S204, based on historical requirement semantic vectors and historical module semantic vectors, constructs a requirement semantic vector matrix and a hierarchical functional directory tree;

[0143] S205, associate the requirement semantic vector matrix with the hierarchical functional directory tree to obtain a mapping model between the preset user self-described requirements and actual operational behaviors.

[0144] In this embodiment, a mapping model between pre-defined user-described needs and actual operational behaviors is adopted. This model can identify the functional modules to be iterated related to user needs through a semantic vector matrix of needs and a functional directory tree. Then, by combining the functional modules to be iterated with the calculation of the authenticity score of the needs, the authenticity of the needs can be effectively verified, ensuring that the development team focuses on the functions that are truly valuable to users, thereby improving the market competitiveness and user satisfaction of the software product.

[0145] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0146] Please see Figure 9 This illustration shows a schematic diagram of a functional iteration monitoring device for a software product provided in an exemplary embodiment of this application. The device can be implemented as all or part of an electronic device through software, hardware, or a combination of both. The device 1 includes a multi-source data collection unit 10, a requirement update unit 20, a requirement sequencing unit 30, and a functional iteration feedback unit 40.

[0147] The multi-source data collection unit 10 is used to periodically collect and preprocess user operation behavior data, demand data, various work order data and implementation survey data about software products to obtain structured user demands to be processed.

[0148] The requirement update unit 20 is used to determine the authenticity score of the user's requirement to be processed based on a preset mapping model between user self-reported requirements and actual operation behavior; and to update the user's requirement to be processed corresponding to the authenticity score greater than the preset threshold to the preset requirement pool; wherein, the mapping model is established based on historical data and user behavior patterns.

[0149] The demand sorting unit 30 is used to calculate the priority score of each target user demand in the preset demand pool and sort them by priority to obtain a user demand priority list.

[0150] The Functional Iteration Feedback Unit 40 is used to iterate the software product based on the priority order of the user requirement priority list, and to collect user behavior parameters of the new functions in each iteration in real time for data analysis to identify the requirements / problems in the new functions, feed them back to the research tool, and continue to perform the steps of regularly collecting and preprocessing user operation behavior data, requirement data, various work order data and implementation research data about the software product.

[0151] It should be noted that the functional iteration monitoring device for software products provided in the above embodiments is only illustrated by the division of the above functional modules when executing the functional iteration monitoring method for software products. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the functional iteration monitoring device for software products provided in the above embodiments and the functional iteration monitoring method embodiments for software products belong to the same concept, and the implementation process is detailed in the method embodiments, which will not be repeated here.

[0152] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0153] In this embodiment, on the one hand, the requirement data collection module can periodically collect and preprocess multi-source data to generate structured user requirements to be processed. The behavior-requirement correlation verification module, based on a preset mapping model, can filter out low-value or false requirements by determining the authenticity score of the requirements. The requirement priority decision module prioritizes requirements according to quantitative indicators to ensure that high-value requirements are developed first. Through the collaborative work of the above modules, the problems of fragmented requirement collection, difficulty in verifying the authenticity of requirements, and subjectivity in requirement priority decision-making can be effectively solved, thereby improving the utilization efficiency of development resources and enhancing the efficiency and quality of product iteration. On the other hand, by adopting a preset mapping model between user self-described requirements and actual operational behaviors, this model can identify the functional modules to be iterated related to user requirements through the requirement semantic vector matrix and functional directory tree. Then, by combining the authenticity score of the requirements with the functional modules to be iterated, the authenticity of the requirements can be effectively verified, ensuring that the development team focuses on functions that are truly valuable to users, thereby improving the market competitiveness and user satisfaction of the software product.

[0154] This application also provides a computer-readable medium storing program instructions thereon, which, when executed by a processor, implement the functional iteration monitoring method for a software product provided in the above-described method embodiments. This application also provides a computer program product containing instructions that, when run on a computer, cause the computer to execute the functional iteration monitoring method for a software product described in the above-described method embodiments.

[0155] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program for monitoring the functional iteration of a software product can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The storage medium for the program for monitoring the functional iteration of a software product can be a magnetic disk, optical disk, read-only memory, or random access memory, etc.

[0156] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A system for functional iteration monitoring of a software product, characterized by, The system comprises: a demand data collection module, a behavior-demand association verification module, a demand priority decision module, and a feedback module; wherein the demand data collection module is configured to periodically collect and preprocess user operation behavior data, demand data, various types of work order data, and implementation survey data related to a software product, to obtain structured to-be-processed user demands, and store the to-be-processed user demands in a preset user demand knowledge base; the behavior-demand association verification module is configured to determine a authenticity score of the to-be-processed user demands based on a preset mapping model of user self-reported demands and actual operation behaviors, and update to-be-processed user demands corresponding to authenticity scores greater than a preset threshold to a preset demand pool; wherein the mapping model is established based on historical data and user behavior patterns; the behavior-demand association verification module is configured to determine a authenticity score of the to-be-processed user demands based on a preset mapping model of user self-reported demands and actual operation behaviors, and update to-be-processed user demands corresponding to authenticity scores greater than a preset threshold to a preset demand pool; wherein the mapping model is established based on historical data and user behavior patterns; the behavior-demand association verification module is configured to determine a authenticity score of the to-be-processed user demands based on a preset mapping model of user self-reported demands and actual operation behaviors, and update to-be-processed user demands corresponding to authenticity scores greater than a preset threshold to a preset demand pool; wherein the mapping model is established based on historical data and user behavior patterns; the behavior-demand association verification module is configured to determine a authenticity score of the to-be-processed user demands based on a preset mapping model of user self-reported demands and actual operation behaviors, and update to-be-processed user demands corresponding to authenticity scores greater than a preset threshold to a preset demand pool; wherein the mapping model is established based on historical data and user behavior patterns; the demand priority decision module is configured to calculate a priority score of each target user demand in the preset demand pool, and sort the target user demands according to the priority scores, to obtain a user demand priority list; 2. The system of claim 1, wherein, the feedback module is configured to perform function iteration on the software product based on the priority order of the user demand priority list, and collect user behavior parameters of each new function in real time for data analysis, to identify demand / problemdata existing in the new function, and feed back to the demand data collection module. the collection and preprocessing of user operation behavior data, demand data, various types of work order data, and implementation survey data related to a software product to obtain structured to-be-processed user demands comprises: collecting user operation behavior data in real time through data burying codes embedded in the software product; extracting demand data fed back by users from a customer service system through an API interface; extracting various types of work order data submitted by customers from a work order system; collecting implementation survey data through an online questionnaire platform or a survey tool; performing data cleaning and data formatting on the user operation behavior data, the demand data, the various types of work order data, and the implementation survey data, to obtain a plurality of pieces of preprocessed functional demand data; extracting keywords and key phrases related to function iteration from each piece of functional demand data; performing semantic analysis on the keywords and key phrases to identify the demand category of each piece of functional demand data; 3. The system of claim 1, wherein, grouping functional demand data of similar demand categories into one category according to the demand category of each piece of functional demand data, to obtain structured to-be-processed user demands. the to-be-processed user demands comprise demand texts and demand categories; The matching of the to-be-processed user demand and the preset user self-reported demand with the mapping model of actual operation behaviors comprises: loading a mapping model of preset user self-reported demand and actual operation behaviors, the mapping model comprising an associated demand semantic vector matrix and a function directory tree; preprocessing the demand text to obtain a word sequence, and converting the word sequence into a demand semantic vector of a preset dimension; retrieving in the demand semantic vector matrix with the demand category as a key to obtain a central demand semantic vector of the demand category; determining a to-be-iterated function module related to the to-be-processed user demand based on the function directory tree and the central demand semantic vector.

4. The system of claim 3, wherein, The function directory tree comprises a plurality of leaf nodes, each leaf node being used to represent an original function module in a software product; The determination of the to-be-iterated function module related to the to-be-processed user demand based on the function directory tree and the central demand semantic vector comprises: traversing function module node information of each leaf node in the function directory tree; calculating a first cosine similarity between the function module node information of each leaf node and the central demand semantic vector; recalling target leaf nodes with a first cosine similarity greater than a preset first similarity threshold value to determine a function module of each target leaf node and obtain a plurality of candidate function modules; calculating a second cosine similarity between the central demand semantic vector and a module semantic vector of each candidate function module; in a case where a maximum second cosine similarity is greater than a preset second similarity threshold value, taking the candidate function module corresponding to the maximum second cosine similarity as the to-be-iterated function module related to the to-be-processed user demand; the preset second similarity threshold value is greater than the preset first similarity threshold value.

5. The system of claim 1, wherein, The following steps are used to generate a mapping model of preset user self-reported demand and actual operation behaviors, comprising: obtaining and preprocessing a plurality of historical demand-function records in a preset historical time period from a preset demand management library to obtain a plurality of effective historical samples; each effective historical sample is composed of a historical demand text and a historical final online function module; performing semantic vector conversion on the historical demand text in each effective historical sample to obtain a historical demand semantic vector; splicing a module name, a function description and a menu path of the historical final online function module in each effective historical sample into a semantic text to obtain a historical module semantic vector; constructing a demand semantic vector matrix and a hierarchical function directory tree based on the historical demand semantic vector and the historical module semantic vector; associating the demand semantic vector matrix and the hierarchical function directory tree to obtain a mapping model of preset user self-reported demand and actual operation behaviors.

6. The system of claim 5, wherein, The construction of the demand semantic vector matrix and the hierarchical function directory tree based on the historical demand semantic vector and the historical module semantic vector comprises: feeding all the historical demand semantic vectors into a preset unsupervised clustering algorithm to converge a plurality of historical demand categories; For each historical demand category, the average of all historical demand semantic vectors for that historical demand category is taken as the category center vector for that historical demand category; Each historical demand category and its category center vector are combined into key-value pairs and stored to obtain a demand semantic vector matrix; Read the existing menu-submenu-function hierarchy structure in the software product; Based on the menu-submenu-function hierarchy, create leaf nodes at different levels; In the leaf nodes at different levels, the historical functional module node information of the functional module corresponding to the leaf node and the semantic vector of the historical module corresponding to the historical final online functional module are persistently stored to obtain a hierarchical functional directory tree.

7. The system of claim 1, wherein, The priority score of each target user's demand in the preset demand pool is calculated and sorted by priority to obtain a user demand priority list, including: Determine the number of customers covered by each target user's needs in the pre-set demand pool, the degree of matching of behavioral data, and the degree of matching of strategic goals; Based on preset corporate strategic information and business objective information, weights are assigned to the number of customers covered by each target user's needs, the degree of matching of behavioral data, and the degree of matching with strategic objectives; Based on the number of customers covered, the degree of matching of behavioral data, the degree of matching of strategic goals, and the weights assigned to each target user need, a priority score is calculated for each target user need. Each target user need is sorted according to its priority score to obtain a user need priority list.

8. The system of claim 7, wherein, The determination of the number of customers covered by each target user's needs in the preset demand pool, the behavioral data matching degree, and the strategic goal matching degree includes: Extract the first demand category and first keyword set for each target user's demand from the preset demand pool; Acquire historical data of user behavior recorded within a preset period for the software product, as well as historical customer service / work order / survey text. From the historical data, find the first user who has made a preset number of valid clicks on the function indicated by the first demand category; From the historical customer service / work order / survey texts, find the target text that matches the first keyword set, and determine the second user who submitted the target text; The first user and the second user are deduplicated to obtain the number of customers covered by each target user's needs in the preset demand pool; Obtain the first behavior data for each first user; Based on the first behavior data of each first user, the matching degree of behavior data of each target user's needs in the preset demand pool is statistically analyzed. Based on a pre-defined corporate strategy text library and the first keyword set, word-level matching is performed to generate the strategic target matching degree of each target user's needs in the pre-defined demand pool.

9. A method for functional iteration monitoring of a software product implemented using the system of any one of claims 1-8, characterized by, The method includes: Regularly collect and preprocess user operation behavior data, requirement data, various work order data and implementation survey data about software products to obtain structured user requirements to be processed; Determine the authenticity score of the to-be-processed user demand based on a preset mapping model of user self-reported needs and actual operation behaviors; update the to-be-processed user demand corresponding to the authenticity score greater than a preset threshold to a preset demand pool; wherein the mapping model is established based on historical data and user behavior patterns; Calculate the priority score of each target user demand in the preset demand pool for priority sorting to obtain a user demand priority list; Based on the priority order of the user demand priority list, perform function iteration on the software product, and collect user behavior parameters of each iteration of new functions in real time for data analysis to identify demand / problem data existing in the new functions, feed back to the research tool, and continue to perform the steps of periodically collecting and preprocessing user operation behavior data, demand data, various work order data and implementation research data of the software product.

Images