Product optimization method and device, equipment and medium
By constructing feature matrix and calculating deviation measurement values, determining outlier function as the optimization goal, the problems of large computing resource consumption and low optimization efficiency caused by multi-dimensional independent evaluation are solved, and efficient unified processing and accuracy improvement of product optimization are achieved.
Patent Information
- Application Number
- CN202311147501.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-09-06
- Publication Date
- 2025-01-14
AI Technical Summary
In the process of product optimization in the prior art, multi-dimensional independent evaluation leads to high consumption of computing resources, slow computing speed, and difficult to determine optimization goals, resulting in low optimization efficiency.
By obtaining the characteristic values and change trends of each function, building a feature matrix, calculating the deviation metric value, and determining the outlier function as the optimization goal, realizing unified extraction and processing of data sources, and improving calculation speed and optimization efficiency.
It improves computing speed and optimization efficiency, reduces the consumption of computer resources, clarifies optimization goals, and improves the accuracy and efficiency of product optimization.
Smart Images
Figure CN119312235A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the fields of artificial intelligence and finance, and more specifically, to a product optimization method, device, equipment, medium and program product. Background Art
[0002] Evaluating and diagnosing products or functions and giving corresponding optimization suggestions is an important process to continuously improve the competitiveness of the product or function and enhance the user experience. In related technologies, multiple dimensions are generally evaluated for each function of a product, and the data from multidimensional data sources are analyzed independently in a single dimension. The data are separated by different data sources as a whole, resulting in inconsistent rankings of each function under each dimension, making it difficult to determine specific optimization goals. The optimization goals include using functions with excellent performance as the basis for optimization, or using functions with poor performance as functions to be optimized.
[0003] In the process of realizing the concept of the invention disclosed in the present invention, the inventors found that when multiple dimensions are evaluated independently, each dimension separately performs processes such as data source acquisition, data processing, indicator calculation, rule matching and result storage, which will consume a large amount of computer resources and the calculation speed is slow. In addition, due to the inconsistent evaluation criteria of each dimension, it is difficult to determine the specific optimization target, resulting in low optimization efficiency. Therefore, how to integrate various dimensions, uniformly extract and process data from each data source to improve the calculation speed and optimization efficiency, and reduce the consumption of computer resources is a problem that needs to be solved urgently. Summary of the invention
[0004] In view of the above problems, the present disclosure provides a product optimization method, apparatus, device, medium and program product.
[0005] One aspect of an embodiment of the present disclosure provides a product optimization method, including: determining N functions provided by a product to be optimized, where N is an integer greater than or equal to 2; obtaining M eigenvalues of each of the functions and a change trend of each feature to obtain N feature matrices, where M is an integer greater than or equal to 1; determining outlier functions among the N functions based on deviation metric values of each of the N feature matrices, where the deviation metric values indicate a degree of deviation of the corresponding function based on a metric standard; and achieving product optimization based on the outlier functions, where the outlier functions include functions whose deviation metric values are greater than or equal to a preset threshold.
[0006] According to an embodiment of the present disclosure, after obtaining the M eigenvalues of each of the functions and the changing trends of each feature, the method further includes: classifying each of the functions based on the M eigenvalues of each of the functions and the changing trends of each feature; and determining the deviation measure value of the feature matrix of each function based on the category of the function.
[0007] According to an embodiment of the present disclosure, the change trend includes an upward trend, a stable trend and a downward trend. Before classifying each of the functions, the value range of each feature is divided into at least two feature value intervals, and each of the feature value intervals corresponds to an upward trend, a stable trend and a downward trend; a classification table including S1 basic classification cells is obtained, each of the basic classification cells has a subordinate category, and each of the basic classification cells is an intersection cell between M features determined according to the feature value interval and the corresponding change trend, and S1 is an integer greater than or equal to 1; wherein, classifying each of the functions includes: based on the interval range of each feature value of each function and the change trend to which it belongs, determining the basic classification cell to which each function belongs from the classification table.
[0008] According to an embodiment of the present disclosure, determining the deviation metric value of the feature matrix of each function based on the category of the function includes: calculating the distance metric between the feature matrices of any two functions in the same category; for any function, calculating the average value of the distance metric between it and all functions in the same category as the deviation metric value.
[0009] According to an embodiment of the present disclosure, calculating the distance metric between the feature matrices of any two functions in the same category includes: for any category, when N is greater than M, calculating the Mahalanobis distance between the feature matrices of any two functions in the same category; when N is less than or equal to M, calculating the Euclidean distance between the feature matrices of any two functions in the same category.
[0010] According to an embodiment of the present disclosure, determining the outlier function among the N functions based on the deviation metric values of each of the N feature matrices includes: sorting the average values of the N functions; and taking the first K functions in the order as outlier functions, where K is an integer greater than or equal to 1 and less than or equal to N.
[0011] According to an embodiment of the present disclosure, before implementing product optimization based on the outlier function, corresponding strategy recommendation content is set for each of the basic classification cells, and the strategy recommendation content between any two basic classification cells is the same or different; wherein, implementing product optimization based on the outlier function includes: determining the strategy recommendation content of the basic classification cell to which the outlier function belongs; and optimizing based on the strategy recommendation content.
[0012] According to an embodiment of the present disclosure, setting corresponding strategy recommendation content for each of the basic classification cells includes: dividing S1 basic classification cells into S2 merged intervals, wherein each merged interval includes at least one basic classification cell, S2 is an integer greater than or equal to 1 and less than or equal to S1; setting corresponding strategy recommendation content for each of the merged intervals, wherein all basic classification cells under the same merged interval have the same strategy recommendation content.
[0013] According to an embodiment of the present disclosure, the product to be optimized includes an application to be optimized, the M feature values include a satisfaction value and an access traffic value, and obtaining the M feature values of each function and the change trend of each feature includes: obtaining the satisfaction value and the satisfaction change trend, the satisfaction value is obtained by at least one user based on the evaluation data of the corresponding function; and obtaining the access traffic value and the traffic change trend, the access traffic value is obtained by at least one user based on the access data of the corresponding function.
[0014] According to an embodiment of the present disclosure, the product to be optimized includes a software application to be optimized, and the N functions are N application functions provided by the software application.
[0015] Another aspect of an embodiment of the present disclosure provides a product optimization device, including: a function determination module, used to determine N functions provided by a product to be optimized, where N is an integer greater than or equal to 2; a feature matrix module, used to obtain M eigenvalues of each of the functions and the change trend of each feature, to obtain N feature matrices, where M is an integer greater than or equal to 1; an outlier function module, used to determine outlier functions among the N functions based on the deviation metric values of each of the N feature matrices, where the deviation metric values indicate the degree of deviation of the corresponding function based on the metric standard; and a function optimization module, used to achieve product optimization based on the outlier functions, where the outlier functions include functions whose deviation metric values are greater than or equal to a preset threshold.
[0016] Another aspect of an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors execute the method as described above.
[0017] Another aspect of the embodiments of the present disclosure further provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the method as described above.
[0018] Another aspect of the embodiments of the present disclosure further provides a computer program product, including a computer program, which implements the above method when executed by a processor.
[0019] One or more of the above embodiments have the following beneficial effects: for each function provided by the product to be optimized, the eigenvalues of each data source under M dimensions and the change trends of each feature are extracted and merged into a high-dimensional feature matrix. On the basis of obtaining the feature matrix based on the static eigenvalues and the dynamic change trends, the deviation metric values of each feature matrix are evaluated to achieve the effect of uniformly extracting and processing data from each data source, and the outlier function is quickly determined as the optimization target to carry out the optimization operation, thereby improving the calculation speed and optimization efficiency compared to the low-dimensional data processing process under a single dimension, and reducing the consumption of computer resources. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:
[0021] Figure 1 The application scenario diagram of the product optimization method according to the embodiment of the present disclosure is schematically shown;
[0022] Figure 2 A flowchart of a product optimization method according to an embodiment of the present disclosure is schematically shown;
[0023] Figure 3 Schematically shows a flow chart for determining a deviation metric value according to an embodiment of the present disclosure;
[0024] Figure 4A A classification flow chart according to an embodiment of the present disclosure is schematically shown;
[0025] Figure 4B A classification table according to an embodiment of the present disclosure is schematically shown;
[0026] Figure 5 A flowchart of calculating the average value of the distance metric according to an embodiment of the present disclosure is schematically shown;
[0027] Figure 6 The optimization flow chart according to the embodiment of the present disclosure is schematically shown;
[0028] Figure 7 A flowchart schematically shows the preset strategy suggestion content according to an embodiment of the present disclosure;
[0029] Figure 8 A flowchart of a product optimization method according to another embodiment of the present disclosure is schematically shown;
[0030] Fig. 9 The structural block diagram of the product optimization device according to the embodiment of the present disclosure is schematically shown;
[0031] Fig.10A block diagram of an electronic device suitable for implementing a product optimization method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION
[0032] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.
[0033] When planning to optimize a product to improve user experience, the difficulty lies in where to optimize. For example, a mobile banking application provides multiple application functions such as e-commerce, transfer, collection, and financial management. It is time-consuming and labor-intensive to optimize all functions, and independent evaluation of each function in multiple dimensions will produce different results. For example, from the perspective of user satisfaction, enterprise cost, enterprise revenue, transaction speed, etc., e-commerce user evaluation may rank higher, while transfer enterprise revenue is higher, which makes it difficult to determine the specific optimization target.
[0034] In some exemplary existing product optimization processes, manual analysis is usually performed on a single data source, such as obtaining user attitude data through questionnaires, and forming conclusions such as satisfaction and product optimization directions through analysis. Or, analysis is performed based on the embedded data recorded in the system background, and the objects that should be optimized in the future are simply located by comparing the views of different functions and pages. There will be at least one or more of the following problems:
[0035] First, the overall analysis results are separated by different data sources (such as questionnaires and embedded point data), lacking truly effective integration, making it difficult to improve the comprehensiveness and accuracy of diagnosis. Existing multi-dimensional perspective analysis only analyzes data from multiple sources independently, or compresses it into one dimension for calculation. It is quite difficult to calculate data from multiple sources in high-dimensional space.
[0036] Second, the analytical conclusions obtained by manual analysis methods are not standardized enough, and it is difficult to compare the analytical conclusions of different researchers on the same product or element, and there is fluctuation in reliability.
[0037] Third, the system model relies on automated analysis to display and locate problematic products or functions, but lacks in-depth advice and guidance.
[0038] Fourth, manual analysis methods require a lot of manpower and time costs, especially when the target product has a large number of functions, the marginal cost increases.
[0039] Some embodiments of the present disclosure provide a product optimization method, which extracts the eigenvalues of each data source and the change trends of each feature under M dimensions for each function provided by the product to be optimized, and fuses them into a high-dimensional feature matrix. On the basis of obtaining the feature matrix based on the static eigenvalues and the dynamic change trends, the deviation metric values of each feature matrix are evaluated to achieve the effect of uniformly extracting and processing data from each data source, and quickly determine the outlier function as the optimization target to carry out the optimization operation, thereby improving the calculation speed and optimization efficiency compared to the low-dimensional data processing process under a single dimension, and reducing the consumption of computer resources.
[0040] Figure 1 The following schematically shows an application scenario diagram of the product optimization method according to an embodiment of the present disclosure. It should be noted that: Figure 1 What is shown are merely examples to which the embodiments of the present disclosure can be applied, so as to help those skilled in the art understand the technical content of the present disclosure, but it does not mean that the embodiments of the present disclosure cannot be used in other devices, systems, environments or scenarios.
[0041] like Figure 1 As shown, the application scenario 100 according to this embodiment may include terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a medium for a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.
[0042] Users can use terminal devices 101, 102, 103 to interact with server 105 through network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).
[0043] The terminal devices 101 , 102 , and 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.
[0044] The server 105 may be a server that provides various services, such as a background management server (only an example) that provides support for websites browsed by users using the terminal devices 101, 102, and 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.
[0045] Server 105 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud computing, network services, and middleware services.
[0046] In applications such as client applications and web applications (abbreviated as APP in English), the client (i.e., front end) and the server (i.e., back end) can communicate data through network messages. For example, in the APP client, the parameters that the server needs to obtain from the client are assembled, and the network request method is called to send them to the server.
[0047] It should be noted that the product optimization method provided in the embodiment of the present invention can be executed by the terminal device 101, 102, 103 or the server 105. Accordingly, the product optimization device provided in the embodiment of the present invention can generally be set in the terminal device 101, 102, 103 or the server 105.
[0048] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.
[0049] The following will be based on Figure 1 The scene described by Figure 2 to Figure 8 The product optimization method of the embodiment of the present disclosure is described in detail.
[0050] Figure 2 A flowchart of a product optimization method according to an embodiment of the present disclosure is schematically shown.
[0051] like Figure 2 As shown, the product optimization method of this embodiment includes:
[0052] In operation S210, N functions provided by the product to be optimized are determined, where N is an integer greater than or equal to 2.
[0053] Products refer to items or services produced or provided by an enterprise or organization to meet customer needs or solve their problems. They can include software products, hardware products or other categories of products. For example, software products refer to products composed of computer programs, data and documents, etc., which are provided to customers in the form of applications, system programs or tool programs. Hardware products refer to products composed of physical materials or component structures for people to use and operate. Functions refer to what the product to be optimized can do or what effects it can provide.
[0054] In some embodiments, the product to be optimized includes a software application to be optimized, and the N functions are N application functions provided by the software application. The software application can provide services to users in the form of a client or a browser webpage in a terminal device, and a backend server is deployed to interact with the front end to implement the functions.
[0055] In operation S220 , M eigenvalues of each function and the change trend of each feature are obtained to obtain N feature matrices, where M is an integer greater than or equal to 1.
[0056] For example, the features of a function can be extracted based on the inherent attributes of the function itself, or can be extracted based on various feedback data of the function from the external environment. For example, the features obtained from the inherent attributes can include R&D costs, function content description, online time, required computer resources, etc. The features obtained from the external environment can include compatibility with terminal devices, total user usage time, enterprise costs, enterprise revenues, user satisfaction, user traffic, etc.
[0057] For example, the product to be optimized includes the application to be optimized, the M feature values include a satisfaction value and an access flow value, and obtaining the M feature values of each function and the change trend of each feature includes: obtaining a satisfaction value and a satisfaction change trend, where the satisfaction value is obtained through evaluation data of at least one user based on the corresponding function, and obtaining an access flow value and a flow change trend, where the access flow value is obtained through access data of at least one user based on the corresponding function.
[0058] For example, the satisfaction value can be obtained through statistical knowledge. Statistics refers to the organization, analysis and interpretation of the collected big data, so as to draw conclusions about the quantity, nature and relationship of related things that conform to the laws of nature, and obtain the laws and trends of related things. In the embodiments of the present disclosure, statistical methods are used to organize, analyze and interpret objectively existing behavioral data to obtain objective evaluation data that conforms to the laws of nature.
[0059] For example, descriptive statistical methods such as mean, median, standard deviation and percentile can be used to obtain customer satisfaction values. The current satisfaction value can also be compared with the mean of historical satisfaction values (such as the satisfaction values obtained in the previous three times) to obtain the change rate as the satisfaction change trend. For example, relevant complaints, service consultations, and internal knowledge retrieval records are used as inputs to the satisfaction calculation model, and the satisfaction value is given after relevant analysis and calculation. For service links that do not have customer evaluations, the actual customer operations, number of operations, operation time, stay time, and system processing and feedback time and other behavioral data in production are analyzed with reference to standard values (such as median or mean, etc.) to obtain the satisfaction value. The satisfaction calculation model can be obtained in advance based on machine learning algorithm training.
[0060] For another example, when a customer uses a certain function of the product to be optimized through a mobile phone, with the user's authorization, the customer's micro-expression images and voice audio are obtained, and then feature extraction processing is performed on the micro-expression images and voice audio to obtain micro-expression image features and voice audio features. Subsequently, the micro-expression image features and voice audio features are input into a pre-trained classification model to obtain a classification result for characterizing the customer's emotional tendency, and then the satisfaction value is obtained.
[0061] For example, the access data of each function can be obtained by using the tracking technology, thereby obtaining the access flow value, and the current access flow value can be compared with the average of the historical access flow value (such as the flow value obtained in the previous three times) to obtain the change rate as the flow change trend. It can be seen that the change trend involved in the present disclosure is represented by a numerical value.
[0062] According to the embodiments of the present disclosure, user attitudes are represented by satisfaction values, user behaviors are represented by access traffic values, and their respective change trends are comprehensively considered. Static and dynamic change data are integrated and directly calculated and analyzed in high-dimensional space in order to more accurately evaluate the degree of deviation of functions and provide more comprehensive and accurate conclusions for determining optimization goals.
[0063] In operation S230 , outlier functions among the N functions are determined based on deviation metric values of the respective N feature matrices, where the deviation metric values indicate a degree of deviation of the corresponding function based on a metric standard.
[0064] The metric can be based on units of measurement such as time, distance, density, etc., and is determined in combination with current and / or historical data of each function. Deviation means that the performance of one or more functions is inconsistent with the majority of functions, or does not meet the expected performance. For example, N functions can be taken as N elements, and the centroid element can be determined based on N feature matrices, and the outlier function can be determined based on the distance from any element to the centroid element. For another example, the density in a certain area around each element, when the density of a certain element is the smallest, it is an outlier function.
[0065] In operation S240 , product optimization is implemented based on outlier functions, where the outlier functions include functions whose deviation metric values are greater than or equal to a preset threshold.
[0066] Optimization can include code bug fixes, code refactoring, front-end page beautification, or business process simplification. The preset threshold can be set flexibly. The performance of outlier functions can be better or worse than that of most functions.
[0067] According to the embodiments of the present disclosure, for each function provided by the product to be optimized, the eigenvalues of each data source and the change trends of each feature under M dimensions are extracted and merged into a high-dimensional feature matrix. On the basis of obtaining the feature matrix based on the static eigenvalues and the dynamic change trends, the deviation metric values of each feature matrix are evaluated to achieve the effect of uniformly extracting and processing data from each data source, and quickly determine the outlier function as the optimization target to carry out the optimization operation, thereby improving the calculation speed and optimization efficiency compared to the low-dimensional data processing process under a single dimension, and reducing the consumption of computer resources.
[0068] Figure 3 The flowchart of determining the deviation metric value according to an embodiment of the present disclosure is schematically shown.
[0069] After obtaining the M eigenvalues of each function and the changing trend of each feature, such as Figure 3 As shown, this embodiment includes:
[0070] In operation S310 , each function is classified based on the M feature values of each function and the change trend of each feature.
[0071] In some embodiments, the M feature values and the changing trends of each feature can be input into a classification model to obtain a classification result of the corresponding function. The classification model can be constructed and trained using a neural network algorithm, for example. In other embodiments, classification rules can be pre-specified based on expert experience, and each rule can be matched with the M feature values and the changing trends of each feature to obtain a classification result.
[0072] In operation S320 , a deviation metric value of a feature matrix of each function is determined based on the category of the function.
[0073] For example, the deviation measurement value may be calculated under the same category, or different measurement standards may be used for different categories, or different deviation measurement units may be used for different categories.
[0074] According to the embodiments of the present disclosure, the deviation metric value is determined according to the category of the function, and the accuracy can be improved by considering the characteristics of the category.
[0075] Figure 4A The classification flow chart according to the embodiment of the present disclosure is schematically shown. Figure 4B A classification table according to an embodiment of the present disclosure is schematically shown.
[0076] The above-mentioned change trend includes an upward trend, a stable trend and a downward trend. Before classifying each function, as shown in FIG. 4 , operation S430 in this embodiment is one of the embodiments of operation S310, and specifically includes:
[0077] In operation S410, the value range of each feature is divided into at least two feature value intervals, and each feature value interval corresponds to an upward trend, a stable trend, and a downward trend.
[0078] For example, if the value range of satisfaction is 0 to 100, then 0 to 50 is the low characteristic value interval, and 51 to 100 is the high characteristic value interval. An upward trend may correspond to a change rate greater than 30%, a stable trend may correspond to a change rate within 20% to 30%, and a downward trend may correspond to a change rate greater than -10% and less than 20%. It can be understood that the values corresponding to the above intervals or trends are only examples.
[0079] In operation S420, a classification table including S1 basic classification cells is obtained, each basic classification cell has a category to which it belongs, each basic classification cell is an intersection cell between M features determined according to a feature value interval and a corresponding change trend, and S1 is an integer greater than or equal to 1.
[0080] Reference Figure 4B , forming 36 basic classification cells. High or low represents the characteristic value interval, and each interval corresponds to rising, stable and falling trends. The non-pattern filled cells are Class I functions, and the pattern filled parts are Class II functions. Class I functions indicate a good development trend. Class II functions indicate a poor situation. Specifically, points can be assigned according to the corresponding characteristics, intervals and change trends, and then the score of each cell is calculated as the basis for classification. The judgment criteria for the characteristic value intervals and each change trend are shown in Table 1 below.
[0081] Table 1
[0082] Serial number type Performance Judgment criteria 1 flow high The functional flow rate in this period is higher than the average of all functional flows 2 flow Low The functional flow rate in this period is lower than the average of all functional flows 3 flow rise The traffic of this function in this period is higher than: the average traffic of the previous three periods of this function + the actual error based on the full historical data of this function 4 flow Stablize The traffic of this function in this period is between: the average traffic of the first three periods of this function + - the standard deviation based on the full historical data of this function 5 flow decline The traffic of this function in this period is less than: the average traffic of the previous three periods of this function - the standard deviation based on the full historical data of this function 6 Satisfaction high The satisfaction with this function is higher than the average satisfaction with all functions in this period 7 Satisfaction Low The satisfaction level of this function is lower than the average of all functions in this period. 8 Satisfaction rise The satisfaction level of this function in this period is higher than: the average satisfaction level of the previous three periods of this function + the standard deviation based on the full historical data of this function 9 Satisfaction Stablize The satisfaction level of this function in this period is between: the average satisfaction level of the previous three periods of this function + - the standard deviation based on the full historical data of this function 10 Satisfaction decline The satisfaction level of this function in this period is less than: the average satisfaction level of this function in the previous three periods - the standard deviation based on the full historical data of this function
[0083] In operation S430, based on the interval range and the change trend of each characteristic value of each function, a basic classification cell to which each function belongs is determined from the classification table.
[0084] Continue to refer to Figure 4B , determine the corresponding intervals according to the satisfaction value and the flow value, and then determine the satisfaction trend and the flow trend to obtain the corresponding basic classification cell. The category corresponding to this cell is the category of the function.
[0085] According to the embodiments of the present disclosure, classification is performed by determining the cell to which each function belongs in the classification table, which can integrate multi-dimensional features, and the classification process is simple, convenient and efficient.
[0086] Figure 5 The flowchart of calculating the average value of the distance metric according to an embodiment of the present disclosure is schematically shown.
[0087] In operation S510 , a distance metric between feature matrices of any two functions in the same category is calculated.
[0088] In some embodiments, calculating the distance metric between the feature matrices of any two functions in the same category includes: for any category, when N is greater than M, calculating the Mahalanobis distance between the feature matrices of any two functions in the same category. When N is less than or equal to M, calculating the Euclidean distance between the feature matrices of any two functions in the same category. The accuracy of distance metric calculation can be improved.
[0089] Due to the requirements of the distance algorithm, it is determined whether the number of functions in each category is greater than the number of features. For example, the satisfaction value, satisfaction change trend, access flow value and flow change trend are four features. If the number of functions is greater than the number of features, the Mahalanobis distance is calculated. Otherwise, the Euclidean distance is calculated.
[0090] The distance between each function is calculated in a four-dimensional space consisting of satisfaction value, satisfaction change trend, access flow value and flow change trend. For example, the Mahalanobis distance is used to calculate the distance d between each feature matrix as shown in Formula 1: ij , thus forming a distance matrix D, where each row is the distance between a feature matrix and the other feature matrices. The Mahalanobis distance eliminates the interference of correlation between features and also eliminates the dimensional differences of each feature.
[0091]
[0092] Where ∑ is the covariance matrix between features. X is the feature matrix, i and j are both greater than or equal to 1.
[0093] For example, if the number of functions is less than or equal to the number of features, the standard Euclidean distance is used to calculate the distance between each feature matrix.
[0094]
[0095] Where V is the variance of the elements in the X feature matrix.
[0096] In operation S520 , for any function, an average value of the distance metrics between it and all functions in the same category is calculated as a deviation metric value.
[0097] In some embodiments, determining the outlier function among the N functions based on the deviation metric values of each of the N feature matrices includes: sorting the average values of the N functions, and taking the first K functions in the order as outlier functions, where K is an integer greater than or equal to 1 and less than or equal to N.
[0098] In some other embodiments, a mean value threshold may be set. If the mean value is greater than the mean value,
[0099] In some other embodiments, a mean value threshold may be set, and if the value is greater than the mean value, it is considered as an outlier function.
[0100] According to an embodiment of the present disclosure, a larger average value of the distance metric indicates that the function deviates more from the performance of other functions, and one or more outlier functions may be identified and optimized.
[0101] Figure 6 The optimization flow chart according to the embodiment of the present disclosure is schematically shown.
[0102] like Figure 6 As shown, in this embodiment, operation S620 and operation S630 are one embodiment of operation S240, and specifically include:
[0103] In operation S610, before implementing product optimization based on the outlier function, corresponding strategy suggestion content is set for each basic classification cell, and the strategy suggestion content between any two basic classification cells is the same or different.
[0104] In operation S620 , the policy suggestion content of the basic classification cell to which the outlier function belongs is determined.
[0105] In operation S630, optimization is performed based on the policy suggestion content.
[0106] According to the embodiments of the present disclosure, it is possible to quickly determine the content of strategy recommendations and improve product optimization efficiency.
[0107] Figure 7 The flowchart of the preset strategy suggestion content according to the embodiment of the present disclosure is schematically shown.
[0108] like Figure 7 As shown, this embodiment is one of the embodiments of operation S610, including:
[0109] In operation S710, S1 basic classification cells are divided into S2 merged intervals, wherein each merged interval includes at least one basic classification cell, and S2 is an integer greater than or equal to 1 and less than or equal to S1. See Table 2.
[0110] Table 2
[0111]
[0112] In operation S720, corresponding strategy suggestion content is set for each merged interval, wherein all basic classification cells in the same merged interval have the same strategy suggestion content.
[0113] Table 3
[0114] serial number suggestion G101 To be both popular and well-received, we should extract experience and share traffic G103 If there is a risk of user loss, the cause should be analyzed G301 If the box office is popular but not well received, we should analyze the reasons and be wary of traffic shock G303 There is a risk of user loss, and analysis should be done as soon as possible G102 Users are losing, and the reasons should be analyzed G104 The performance is worrying and the reasons should be analyzed as soon as possible G302 Users are losing and should be analyzed as soon as possible G304 The performance is worrying and the reasons should be analyzed as soon as possible G201 Popularity but not sales, we should look for growth opportunities G203 There is a risk of user loss, and the reasons should be analyzed G202 Users are losing, and the reasons should be analyzed G204 The performance is worrying and the reasons should be analyzed as soon as possible G401 Neither popular nor well-received, development should be observed G402 Neither popular nor profitable, observation should be combined with business strategy analysis
[0115] In some embodiments, description content may be pre-set for each basic classification cell, and descriptions of the respective conditions of the 36 basic classification cells may be given, as shown in Table 4.
[0116] Table 4
[0117]
[0118] According to an embodiment of the present disclosure, each policy recommendation content may correspond to multiple function categories, which is convenient for reducing the amount of data storage, and when multiple outlier functions correspond to the same policy recommendation content, the optimization efficiency can be improved.
[0119] Figure 8 A flowchart of a product optimization method according to another embodiment of the present disclosure is schematically shown.
[0120] like Figure 8 As shown, the product optimization method of this embodiment includes:
[0121] In operation S810, data is acquired.
[0122] The initial data table D101 is obtained by automatic collection or manual upload. The variables included in the initial data table D101 are: function name (title), satisfaction (ets), satisfaction change (etsc), access flow (pv), access flow change (pvc).
[0123] in,
[0124] ets: ETS score of the group standard "Guidelines for User Experience Evaluation of ETS Index for Financial Interface Products (T / IQA20-2023)". This indicator measures the satisfaction with the experience.
[0125] etsc: The rate of change of this period's ets compared with the average of the previous three periods' ets.
[0126] pv: The access traffic of the target product or function of experience diagnosis.
[0127] PVC: The rate of change of PV in this period compared with the average PV of the previous three periods.
[0128] In operation S820, the elements are classified. Refer to operation S310.
[0129] In operation S830, a distance matrix is obtained for each category.
[0130] In operation S840, an average value of the distance metric of each element is obtained based on the distance matrix.
[0131] In operation S850, a policy recommendation is generated for each discrete function under each category. For example, under category I functions, the satisfaction and traffic of function A are both high and rising significantly. The policy recommendation is that the function is popular and well received, and experience should be extracted and traffic should be shared. Under category II functions, the traffic of function B is high and has dropped significantly, and the satisfaction is low and falling. The policy recommendation is that the performance of the function is worrying, and the reasons should be analyzed as soon as possible.
[0132] In operation S860, the product is optimized based on the policy recommendation.
[0133] According to the embodiments of the present disclosure, an automated, standardized, and intelligent function evaluation method is established. Attitude data and behavior data that can reflect user experience are collected, static and dynamic change data are integrated, and direct calculation and analysis are performed in high-dimensional space, without information loss caused by compressed dimensions, which can provide more comprehensive and accurate conclusions. At the same time, the product functions are classified and analyzed in full combination with actual needs, and standardized and personalized status descriptions and strategic recommendations are given for outlier functions.
[0134] Based on the above product optimization method, the present disclosure also provides a product optimization device. Fig. 9 The device is described in detail.
[0135] Fig. 9 The structural block diagram of the product optimization device according to an embodiment of the present disclosure is schematically shown.
[0136] like Fig. 9 As shown, the product optimization device 900 of this embodiment includes a function determination module 910, a feature matrix module 920, an outlier function module 930, and a function optimization module 940.
[0137] The function determination module 910 may perform operation S210 to determine N functions provided by the product to be optimized, where N is an integer greater than or equal to 2.
[0138] The feature matrix module 920 may perform operation S220 to obtain M feature values of each function and a change trend of each feature to obtain N feature matrices, where M is an integer greater than or equal to 1.
[0139] The outlier function module 930 may perform operation S230 for determining an outlier function among the N functions based on a deviation metric value of each of the N feature matrices, where the deviation metric value indicates a degree of deviation of the corresponding function based on a metric standard.
[0140] The function optimization module 940 may perform operation S240 for implementing product optimization based on outlier functions, where the outlier functions include functions whose deviation metric values are greater than or equal to a preset threshold.
[0141] The product optimization device 900 includes two parts for performing the above Figure 2 to Figure 8 A module for each step of any one of the described embodiments.
[0142] It should be noted that the implementation methods, technical problems solved, functions realized, and technical effects achieved of each module / unit / sub-unit in the device part embodiment are the same or similar to the implementation methods, technical problems solved, functions realized, and technical effects achieved of each corresponding step in the method part embodiment, and will not be repeated here.
[0143] According to an embodiment of the present disclosure, any multiple modules among the function determination module 910, the feature matrix module 920, the outlier function module 930, and the function optimization module 940 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module.
[0144] According to an embodiment of the present disclosure, at least one of the function determination module 910, the feature matrix module 920, the outlier function module 930, and the function optimization module 940 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware or in an appropriate combination of any of them. Alternatively, at least one of the function determination module 910, the feature matrix module 920, the outlier function module 930, and the function optimization module 940 can be at least partially implemented as a computer program module, and when the computer program module is run, the corresponding function can be executed.
[0145] Fig.10 A block diagram of an electronic device suitable for implementing a product optimization method according to an embodiment of the present disclosure is schematically shown.
[0146] like Fig.10As shown, the electronic device 1000 according to an embodiment of the present disclosure includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage part 1008 into a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include an onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.
[0147] In RAM 1003, various programs and data required for the operation of electronic device 1000 are stored. Processor 1001, ROM 1002 and RAM 1003 are connected to each other via bus 1004. Processor 1001 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 1002 and / or RAM 1003. It should be noted that the program can also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in one or more memories.
[0148] According to an embodiment of the present disclosure, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to the bus 1004. The electronic device 1000 may further include one or more of the following components connected to the I / O interface 1005: an input portion 1006 including a keyboard, a mouse, etc. An output portion 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc. A storage portion 1008 including a hard disk, etc. And a communication portion 1009 including a network interface card such as a LAN card, a modem, etc. The communication portion 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that the computer program read therefrom is installed into the storage portion 1008 as needed.
[0149] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments. It may also exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.
[0150] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: 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), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, a computer-readable storage medium may include the ROM 1002 and / or RAM 1003 described above and / or one or more memories other than ROM 1002 and RAM 1003.
[0151] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the method provided by the embodiment of the present disclosure.
[0152] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 1001. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.
[0153] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 1009, and / or installed from the removable medium 1011. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.
[0154] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1009, and / or installed from the removable medium 1011. When the computer program is executed by the processor 1001, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.
[0155] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or product-oriented programming languages, and / or assembly / machine languages. Programming languages include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).
[0156] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0157] It will be appreciated by those skilled in the art that the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations and / or combinations are not explicitly described in the present disclosure. In particular, the features described in the various embodiments and / or claims of the present disclosure may be combined and / or combined in a variety of ways without departing from the spirit and teachings of the present disclosure. All of these combinations and / or combinations fall within the scope of the present disclosure.
[0158] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above separately, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. The scope of the present disclosure is defined by the attached claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.
Claims
1. A product optimization method, comprising: Determine N functions provided by the product to be optimized, where N is an integer greater than or equal to 2; Obtaining M eigenvalues of each function and the change trend of each feature to obtain N feature matrices, where M is an integer greater than or equal to 1; Determining an outlier function among the N functions based on a deviation metric value of each of the N feature matrices, the deviation metric value indicating a degree of deviation of the corresponding function based on a metric standard; Product optimization is achieved based on the outlier function, wherein the outlier function includes a function where the deviation metric value is greater than or equal to a preset threshold.
2. The method according to claim 1, wherein: After obtaining the M characteristic values of each function and the change trend of each characteristic, the method further includes: Classifying each of the functions based on the M feature values of each function and the change trend of each feature; A deviation measure value of a feature matrix of each function is determined based on the category of the function.
3. The method according to claim 2, wherein: The change trend includes an upward trend, a stable trend and a downward trend. Before classifying each of the functions, The value range of each feature is divided into at least two feature value intervals, and each of the feature value intervals corresponds to an upward trend, a stable trend, and a downward trend; Obtain a classification table including S1 basic classification cells, each of which has a category to which it belongs, each of which is an intersection cell between M features determined according to a feature value interval and a corresponding change trend, and S1 is an integer greater than or equal to 1; Among them, classifying each of the functions includes: Based on the interval range and the change trend of each characteristic value of each function, the basic classification cell to which each function belongs is determined from the classification table.
4. The method according to claim 2, wherein: Determining the deviation measure value of the feature matrix of each function based on the category of the function includes: Calculate the distance metric between the feature matrices of any two features of the same category; For any of the functions, the average of the distance measures between the function and all functions in the same category is calculated as the deviation measure value.
5. The method according to claim 4, wherein: The distance metric between the feature matrices of any two features of the same category is calculated as follows: for any category, When N is greater than M, the Mahalanobis distance between the feature matrices of any two functions in the same category is calculated; When N is less than or equal to M, the Euclidean distance between the feature matrices of any two features of the same category is calculated.
6. The method according to claim 4, wherein: Determining the outlier function among the N functions based on the deviation metric values of each of the N feature matrices includes: Sort the average values of N functions; The first K functions in the order of K are respectively regarded as outlier functions, where K is an integer greater than or equal to 1 and less than or equal to N.
7. The method according to claim 3, wherein: Before implementing product optimization based on the outlier function, corresponding strategy suggestion content is set for each of the basic classification cells, and the strategy suggestion content between any two basic classification cells is the same or different; Wherein, realizing product optimization based on the outlier function includes: Determining the strategy recommendation content of the basic classification cell to which the outlier function belongs; Based on the strategy, content is recommended for optimization.
8. The method according to claim 7, wherein: The corresponding strategy suggestion content for each basic classification cell includes: Divide S1 basic classification cells into S2 merged intervals, where each merged interval includes at least one basic classification cell, and S2 is an integer greater than or equal to 1 and less than or equal to S1; Corresponding strategy suggestion content is set for each of the merged intervals, wherein all basic classification cells under the same merged interval have the same strategy suggestion content.
9. The method according to any one of claims 1 to 8, wherein: The product to be optimized includes an application to be optimized, the M characteristic values include a satisfaction value and an access flow value, and obtaining the M characteristic values of each function and the change trend of each characteristic includes: Acquire the satisfaction value and the satisfaction change trend, wherein the satisfaction value is obtained through evaluation data of at least one user based on a corresponding function; and The access flow value and the flow change trend are obtained, wherein the access flow value is obtained through access data of at least one user based on a corresponding function.
10. The method according to any one of claims 1 to 8, wherein: The product to be optimized includes a software application to be optimized, and the N functions are N application functions provided by the software application.
11. A product optimization device, comprising: A function determination module, used to determine N functions provided by the product to be optimized, where N is an integer greater than or equal to 2; A feature matrix module, used to obtain M feature values of each function and the change trend of each feature, and obtain N feature matrices, where M is an integer greater than or equal to 1; An outlier function module, configured to determine an outlier function among the N functions based on a deviation metric value of each of the N feature matrices, wherein the deviation metric value indicates a degree of deviation of the corresponding function based on a metric standard; A function optimization module is used to achieve product optimization based on the outlier function, wherein the outlier function includes a function whose deviation metric value is greater than or equal to a preset threshold.
12. An electronic device comprising: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors are enabled to execute the method according to any one of claims 1 to 10.
13. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, causes the processor to execute the method according to any one of claims 1 to 10.