Laptop adaptation test method and system in combination with user habits

Abstract

The invention relates to a notebook adaptation test method and system in combination with user habits. According to the method, user habit data including software use preferences, input mode characteristics and environment setting behaviors in multiple scenes are collected, and a typical user habit model library is constructed through a clustering algorithm; generating a simulation operation habit sequence set based on the model library, injecting the simulation operation habit sequence set into a notebook system, and synchronously collecting resource occupation feature vectors and operation response delay data as system adaptation response data; and then performing multi-dimensional correlation analysis on the response data and the model library to obtain a user habit adaptation degree index and a key adaptation bottleneck path. And finally, a user habit adaptation optimization report is generated based on an analysis result and is used for guiding software and hardware parameter adaptation optimization of the notebook computer system, and the suitability of the system and user habits is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer device intelligent testing, and in particular to a notebook adaptation testing method and system combined with user habits. BACKGROUND

[0002] With the rapid development of information technology, notebook computers have become an indispensable tool in people's work, study and life. In the development process of notebook computers, adaptation testing has always been an important link to ensure that its performance matches user needs. However, the traditional notebook adaptation testing method has significant limitations.

[0003] Past tests have mostly revolved around standardized performance indicators of hardware, such as using specific benchmark testing programs to evaluate the operation speed of central processors, the data transmission capacity of memories, and the graphics rendering capacity of graphics processors, etc. In terms of software, the main focus is on the compatibility of common office software and basic functions of operating systems to ensure that the above-mentioned software can run stably on notebooks. However, as the application scenarios of notebook computers continue to expand and user needs become increasingly diverse, traditional testing methods have been unable to meet actual needs. Different users exhibit different habits when using notebook computers. In a work context, some users focus on word processing and frequently use specific functions of document editing software, such as formatting, content revision, etc. Users engaged in data analysis, on the other hand, tend to frequently call complex data calculation and chart generation functions in spreadsheet software. In terms of input methods, some users prefer the fast input and shortcut key operations of keyboards, some users are accustomed to gesture operations of touchpads, and some users will use specific input devices such as graphic tablets for designers and gamepads for game players. In terms of environmental settings, users will adjust screen brightness according to ambient light, change volume according to usage scenarios and personal preferences, and switch power management modes according to power and task requirements.

[0004] Therefore, due to the failure to fully consider the rich and diverse user habits, the traditional testing method often results in a mismatch between the performance of notebook computers and user needs in actual use. For example, for users who frequently perform multitasking and high-frequency use of specific software functions, the notebook may have problems such as high resource occupation and long response delay, but the traditional test cannot effectively detect such problems and cannot provide accurate guidance for the optimization of notebook systems. SUMMARY

[0005] In view of the above, to at least partially solve the deficiencies in the prior art, in a first aspect, the embodiments of the present application provide a notebook adaptation testing method combined with user habits, which comprises: Collecting multi-scene user habit data and constructing a typical user habit model library, classifying the user habit data through a clustering algorithm to obtain a plurality of typical user habit models, the typical user habit models corresponding to operation behavior characteristics of different user groups; Generating a simulated operation habit sequence set based on the typical user habit model library, the simulated operation habit sequence set being constructed according to operation preference characteristics, input timing characteristics and environmental adjustment rules in each typical user habit model, and containing user operation instructions conforming to behavior modes of different user groups; Injecting the simulated operation habit sequence set into a notebook system and synchronously collecting system adaptation response data, the system adaptation response data containing resource occupation characteristic vectors and operation response delay data; Performing multi-dimensional correlation analysis on the system adaptation response data and the typical user habit model library, calculating a user habit adaptation degree index and a key adaptation bottleneck path by comparing system adaptation response data corresponding to different typical user habit models; Generating a user habit adaptation optimization report based on the user habit adaptation degree index and the key adaptation bottleneck path.

[0006] In a second aspect, the embodiments of the present application also provide a notebook adaptation test system combined with user habits, comprising a processor, a machine readable storage medium, the machine readable storage medium being connected with the processor, the machine readable storage medium being used for storing programs, instructions or codes, and the processor being used for executing the programs, instructions or codes in the machine readable storage medium to realize the notebook adaptation test method combined with user habits.

[0007] In summary, the notebook adaptation test method and system combined with user habits provided by the embodiments of the present application can accurately depict differences of different user groups in multiple dimensions such as software use preference, input mode characteristics and environmental setting behavior by comprehensively collecting user habit data in multiple scenes and constructing a typical user habit model library using a clustering algorithm. The directional analysis of user operation behavior characteristics provides a reliable basis for subsequent tests based on real user behavior modes, greatly improving the pertinence and effectiveness of the tests. In addition, the simulated operation habit sequence set is generated based on the typical user habit model library, closely fitting the actual behavior modes of different user groups. Injecting it into the notebook system and synchronously collecting system adaptation response data including resource occupation characteristic vectors and operation response delay data can truly reflect the performance of the notebook system when facing various actual user habit operations. Compared with traditional standardized tests, the test method based on real user behavior simulation can reveal potential problems of the system in actual use.

[0008] Meanwhile, the system adaptation response data is associated with the typical user habit model library in multiple dimensions for analysis, and the user habit adaptation degree index and key adaptation bottleneck path are accurately calculated. The user habit adaptation degree index quantitatively indicates the adaptation ability of the notebook system to the specific user habit, and the key adaptation bottleneck path clearly indicates the association between the specific user operation sequence and the system component that causes the abnormal system response. This provides a key clue for in-depth understanding of the system performance bottleneck, and helps to optimize the notebook system in a targeted manner. Finally, the user habit adaptation optimization report generated based on the above analysis results, for example, contains detailed user habit type identification, system adaptation short board analysis, and targeted adjustment suggestions, etc., which provides guidance for the software and hardware parameter adaptation optimization of the notebook system, so that the optimization work can be targeted, the adaptation of the system and the user habit is effectively improved, and the performance and user experience of the notebook system in the actual use scenario are comprehensively improved, and the competitiveness of the product in the market is enhanced.

[0009] Other features and advantages of the present application will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained according to the above drawings without creative labor for those skilled in the art.

[0011] In order to more completely understand the present application and its beneficial effects, the following will be described in conjunction with the drawings, wherein the same reference numerals in the following description represent the same parts.

[0012] Figure 1 is a flowchart of a notebook adaptation test method combining user habits provided by an embodiment of the present application.

[0013] Figure 2 is an application scenario diagram of a notebook adaptation test method combining user habits provided by an embodiment of the present application.

[0014] Figure 3 is a schematic diagram of a notebook adaptation test system combining user habits provided by an embodiment of the present application. DETAILED DESCRIPTION

[0015] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative effort are within the protection scope of the present application.

[0016] Please refer to Figure 1 and Figure 2 , Figure 1 is a flowchart of a notebook adaptation test method combining user habits provided by an embodiment of the present application, Figure 2 is an application scenario diagram of a notebook adaptation test method combining user habits provided by an embodiment of the present application. The application scenario includes a system end platform for information interaction and multiple notebooks. In this embodiment, as an example, the system end platform can be a server, a server cluster or other device with large data processing and storage capabilities. The embodiment does not specifically limit this. The system end platform can receive and store various information generated by the multiple notebooks in the office scene adaptation test process, including typical user habit model data, simulated operation habit sequences, system adaptation response data (including resource occupation feature vectors and operation response delay data), and data related to user habit adaptation degree indicators and key adaptation bottleneck paths. On the other hand, the system end platform analyzes the above information in depth, such as correlating the system adaptation response data with the typical user habit model library, thereby identifying the adaptation short board and key adaptation bottleneck of the notebook system to different user groups habits. The analysis result helps to adapt and optimize the software and hardware parameters of the notebook system or give targeted adjustment suggestions, which can be used as a background for providing system updates and maintenance for the notebook. Of course, in other embodiments, the notebook local end can also perform the test method to flexibly meet the test needs in different office scenes (such as small office team local test, personal user self-defined test, etc.). The system end platform can be a background for providing system updates, maintenance and adaptation test and other functions for the notebook. In other embodiments, the method can also be performed by the notebook local end, which is not specifically limited.

[0017] As Figure 1 shown, the method includes steps S110-S150, which are described in detail below.

[0018] Step S110: Collecting user habit data in multiple scenes and constructing a typical user habit model library. The user habit data includes software use preference data, input method feature data and environment setting behavior data. The user habit data is classified by a clustering algorithm to obtain multiple typical user habit models, and the typical user habit model corresponds to the operation behavior characteristics of different user groups. In this embodiment, only as an optional example, an office scenario is taken as an example scenario throughout the whole text. In the office scenario, when collecting user habit data, the software usage preference data needs to record the information about the user starting application programs such as word processing software, spreadsheet software, and mail client, the input method feature data needs to capture the operation characteristics such as keyboard key strokes and touchpad sliding, and the environment setting behavior data needs to track the parameter adjustment conditions such as screen brightness and volume. Through the clustering algorithm, the above data is classified, and the typical user habit model formed can reflect the operation characteristics of different user groups in the office scenario, for example, some users frequently use word processing software and rely on keyboard input, and some users often use spreadsheet software and operate the touchpad more. In this embodiment, step S110 can include sub-steps S111-S114, which are described in detail as follows. Step S111: Collecting user habit data in a plurality of preset scenarios. For example, the software usage preference data includes application program startup frequency, usage time length, and function module call record, the input method feature data includes keyboard key stroke frequency, touchpad sliding trajectory, and external device operation mode, and the environment setting behavior data includes screen brightness adjustment period, volume variation rule, and power management mode switching record. As an example, the preset scenarios can include, but are not limited to, an office scenario, an entertainment scenario, and a mobile scenario, etc. In this embodiment, when collecting user habit data in the office scenario, in terms of software usage preference data, the startup frequency of application programs such as word processing software and spreadsheet software needs to be counted, the usage time length after each startup needs to be recorded, and the call record of function modules such as "paragraph setting" and "insert table" in the word processing software needs to be recorded. In terms of input method feature data, the key stroke frequency of each key of the keyboard per unit time needs to be counted, the trajectory characteristics of the touchpad sliding, including the sliding direction and path, and the click condition of the external mouse and other external device operation modes need to be recorded. In terms of environment setting behavior data, the time interval of screen brightness adjustment, i.e., the adjustment period, needs to be tracked, the volume size variation rule needs to be recorded, and the switching record of the power management mode between "balance mode" and "high performance mode" needs to be recorded. Step S112: Preprocessing the user habit data, and classifying the preprocessed user habit data by using a clustering algorithm, by calculating the similarity between user behavior feature vectors and clustering the user habit data based on the similarity.

[0019] The number of clustering clusters is determined according to the data density distribution in the clustering process, so that the clustering result can distinguish different user groups with different operation preferences; the preprocessing includes data cleaning, feature standardization and behavior event alignment, the data cleaning is used for removing abnormal operation records and repeated data, the feature standardization is used for converting habit data with different dimensions into a unified feature space, and the behavior event alignment is used for integrating related operation behaviors according to timestamps into complete behavior sequences. In the embodiment, when the user habit data collected in the office scenario is preprocessed, the data cleaning link needs to eliminate abnormal operation records such as single abnormal continuous keystrokes of the keyboard, random sliding of the touchpad without operation, and repeated data such as repeated storage of the same application program startup record. In the feature standardization link, data with different dimensions such as application program usage time (unit: minute) and keyboard keystroke frequency (unit: times / minute) are converted into a unified feature space to ensure the rationality of subsequent calculation. In the behavior event alignment link, related operation behaviors such as starting a word processing software, inputting text, and adjusting screen brightness are integrated according to timestamps to form complete behavior sequences. Then, a clustering algorithm is used for classification, the similarity between user behavior feature vectors is calculated, the number of clustering clusters is determined according to the density distribution of data in the office scenario, and the clustering result can distinguish different user groups with different operation preferences, such as user groups frequently using spreadsheet software and user groups frequently using email clients. In the embodiment, step S112 can include sub-steps S1121-S1124, which are described in detail as follows. Step S1121: converting the preprocessed user habit data into high-dimensional feature vectors, and each dimension of the feature vector corresponds to a user habit feature, and the dimensions include software usage frequency dimension, input method proportion dimension, environment setting parameter dimension and time sequence behavior transition dimension. In the embodiment, when the preprocessed user habit data in the office scenario is converted into high-dimensional feature vectors, the software usage frequency dimension corresponds to features such as word processing software startup frequency and spreadsheet software startup frequency; the input method proportion dimension corresponds to features such as keyboard input time proportion and touchpad operation time proportion; the environment setting parameter dimension corresponds to features such as screen brightness adjustment amplitude and volume adjustment amplitude; and the time sequence behavior transition dimension corresponds to behavior transition features such as transition from starting a word processing software to calling a “paragraph setting” function. Each dimension accurately corresponds to a user habit feature in the office scenario, and together constitutes a high-dimensional feature vector. Step S1122: clustering the high-dimensional feature vectors by using a density clustering algorithm, dividing the points with density into a clustering cluster by calculating the density accessibility of the data points in the feature space, and setting the clustering radius and the minimum sample number parameters, the clustering radius is determined by distance distribution statistics of the feature space, and the minimum sample number is set according to the size of the user group. In this embodiment, in the office scenario, the density of each data point in the feature space is calculated when the density clustering algorithm is used to cluster the high-dimensional feature vectors, that is, it is determined whether a data point can reach another data point through a series of data points with high enough density, and the points connected by density are divided into a cluster. The clustering radius is determined by the distance distribution between the data points in the feature space in the office scenario, to ensure that the dense and sparse areas can be reasonably distinguished. The minimum sample size is set according to the size of the user group involved in the office scenario, to ensure that each cluster can represent a certain size of user group with similar operation habits. Step S1123: Perform validity evaluation on the clustering result, and calculate the clustering similarity of each data point through the silhouette coefficient. The silhouette coefficient is used to evaluate the clustering effect by the similarity between the sample and the cluster and the similarity between the clusters, and the abnormal cluster with a silhouette coefficient lower than a set threshold is removed to obtain the effective cluster.

[0020] In this embodiment, when the clustering result in the office scenario is evaluated for validity, the clustering similarity of each data point is calculated through the silhouette coefficient. The similarity between each sample and other samples in the same cluster, i.e., the intra-cluster similarity, is calculated, and then the similarity between the sample and the samples in other clusters, i.e., the inter-cluster similarity, is calculated, and the silhouette coefficient is calculated through the two similarities. A threshold is set, and the abnormal cluster with a silhouette coefficient lower than the threshold is removed. The above abnormal cluster may be caused by the special operation habits of individual users in the office scenario, and after removal, the effective cluster that can accurately reflect the operation habits of different user groups in the office scenario is obtained. Step S1124: Name and feature label the effective cluster, assign a user group label to each effective cluster according to the common features of the user habit data in each effective cluster, the user group label is determined based on the common features of the user habit data in each effective cluster, and the feature label includes the description of the typical operation behavior of the users in each effective cluster. In this embodiment, when the effective cluster in the office scenario is named and feature labeled, the common features of the user habit data in each effective cluster are observed. If the users in a certain effective cluster frequently start the word processing software and mostly use the keyboard for input operation, the user group label of "word processing-keyboard dependent type" is assigned to the cluster. The feature label part describes the typical operation behavior of the users in the cluster in detail, such as starting the word processing software multiple times a day, each time using for a long time, and using keyboard shortcuts for text formatting, etc. Step S113: feature extraction is performed on each clustering result to obtain target habit features of the typical user habit model, the target habit features including a high-frequency software set, an input device preference weight, an environmental parameter adjustment range, and a time sequence behavior dependency relationship. The high-frequency software set is determined by sorting application calling frequencies, the input device preference weight is calculated by using proportions of different input modes, the environmental parameter adjustment range is obtained by statistical intervals of brightness, volume, and other setting values, and the time sequence behavior dependency relationship is obtained by learning transition probabilities of user operation sequences by using a Markov chain model. In the embodiment, when feature extraction is performed on each clustering result in an office scenario, the high-frequency software set is determined by sorting frequencies of calling each application program by the user in the clustering result. For example, the calling frequencies of word processing software, a mail client, and spreadsheet software are high after sorting, and the three constitute the high-frequency software set corresponding to the clustering result. The input device preference weight is obtained by calculating proportions of using different input modes such as a keyboard, a touchpad, and an external mouse. If the proportion of using the keyboard is the highest, the preference weight corresponding to the keyboard is the largest. The environmental parameter adjustment range is determined by statistical intervals of setting values such as screen brightness and volume. For example, if setting values of screen brightness are mostly in a certain interval, the interval is the adjustment range of screen brightness. The time sequence behavior dependency relationship is obtained by learning transition probabilities of a user operation sequence by using a Markov chain model. For example, transition probabilities of an operation sequence in which the user starts word processing software, calls a “save” function, and then closes the software are learned, and the time sequence behavior dependency relationship is determined. In the embodiment, step S113 can include sub-steps S1131-S1134, which are described in detail as follows. Step S1131: record the keyboard key stroke frequency, obtain the input habit features of the user by counting the triggering times of different keys and the usage frequency of combination keys in a unit time, and the combination keys include system shortcut keys and application shortcut keys.

[0021] In the embodiment, when the keyboard key stroke frequency is recorded in an office scenario, the triggering times of different keys such as letter keys, number keys, and function keys on the keyboard in a unit time are counted, and the usage frequencies of system shortcut keys such as “Ctrl+C” and “Ctrl+V” and application shortcut keys such as “Ctrl+B” (bold text) and “Ctrl+I” (italic text) are counted. By using the statistical data, the input habit features of the user can be obtained. For example, some users frequently use the “Ctrl+S” (save) shortcut key, and some users often use letter keys for text input. Step S1132: Collecting touchpad sliding track data, which includes sliding direction, sliding speed and pressure change characteristics. The sliding direction is represented by the displacement components in the horizontal and vertical directions, the sliding speed is calculated by the ratio of displacement to time, and the pressure change characteristics are recorded by the output value sequence of the touchpad pressure sensor. In this embodiment, when collecting touchpad sliding track data in an office scenario, the sliding direction is represented by the displacement components in the horizontal and vertical directions, such as a positive horizontal displacement component indicating a rightward slide and a positive vertical displacement component indicating an upward slide. The sliding speed is calculated by the ratio of displacement to time during the sliding process, reflecting the speed of the user sliding the touchpad. The pressure change characteristics are recorded by the output value sequence of the touchpad pressure sensor, with different output values corresponding to different pressing forces, reflecting the pressure change during the user's operation of the touchpad. Step S1133: Recording the external device operation mode, including the click frequency of the external mouse, the number of times the scroll wheel is used, and the key layout preference of the external keyboard. The external device operation mode is obtained by correlating the device connection state log and the input event log.

[0022] In this embodiment, when recording the external device operation mode in an office scenario, the click frequency of the external mouse is counted, i.e. the number of clicks per unit time, and the number of times the scroll wheel is used. For the external keyboard, the user's preference for the key layout is understood, such as whether they are used to using an external keyboard with a numeric keypad. By correlating the connection state log (recording when the device is connected and disconnected) and the input event log (recording the input operations of the device), the external device operation mode is determined, such as the user often connecting the external mouse and the external keyboard with a numeric keypad when dealing with a large amount of data. Step S1134: Feature fusion of input method feature data, concatenating the feature vectors of the keyboard, touchpad and external device into a unified input method feature vector for subsequent clustering analysis. In this embodiment, when performing feature fusion of input method feature data in an office scenario, the previously obtained keyboard key stroke frequency feature vector, touchpad sliding track feature vector and external device operation mode feature vector are concatenated. During the concatenation process, the dimensions of each feature vector are ensured to be reasonable, and after concatenation, a unified input method feature vector is formed, which contains comprehensive features of the user's input through the keyboard, touchpad and external device in an office scenario, and can be used for subsequent clustering analysis. Step S114: add each of the typical user habit models to a preset model library to obtain a typical user habit model library. The typical user habit model library includes a model identifier, a user group label, a target habit feature vector, and a description of an applicable scenario. The model identifier is used to uniquely distinguish different models, the user group label is used to describe the type of user corresponding to the model, the target habit feature vector is used to quantify the behavior characteristics of the model, and the description of the applicable scenario is used to describe the use scenario corresponding to the model. In this embodiment, when each typical user habit model obtained in the office scenario is added to the preset model library, a unique model identifier, such as "BM001", "BM002", etc., is assigned to each model to distinguish different models. The user group label, such as "frequent text processing type" and "email communication dominant type", describes the type of user corresponding to the model. The target habit feature vector quantifies the behavior characteristics of the model, including quantization data of high-frequency software usage and input device preferences. The description of the applicable scenario clearly indicates that the use scenario corresponding to the model is the office scenario, and finally a typical user habit model library containing the above information is formed. Step S120: generate a simulated operation habit sequence set based on the typical user habit model library. The simulated operation habit sequence set is constructed according to the operation preference characteristics, input timing characteristics, and environmental adjustment rules in each typical user habit model, and includes user operation instructions that conform to different user group behavior patterns.

[0023] The target habit feature vector of the typical user habit model can be parsed, combined with the operation node attributes of the user behavior graph, and a high-frequency software set with a usage frequency higher than a preset frequency can be extracted as an operation object, and an operation object association graph can be constructed. Based on the timing behavior dependency relationship and the path mining results of the user behavior graph, a user operation sequence generator is constructed to generate a user operation sequence containing user operation instructions that conform to the characteristics of the model. The user operation instructions include application program startup instructions, input user operation instructions, environmental adjustment instructions, and task switching instructions. The generated user operation sequence is subjected to multi-dimensional rationality verification, combined with the operation specification library of the user behavior graph to correct abnormal sequences, and the user operation sequences that pass the verification are combined into a simulated operation habit sequence set. Each user operation sequence corresponds to a typical user habit model, and the simulated operation habit sequence set contains complete operation processes of different user groups in different scenarios. In this embodiment, when step S120 is performed in an office scenario, a model corresponding to a user group label such as "word processing high frequency type" and "spreadsheet high frequency type" can be selected from a typical user habit model library, a model association network is constructed in combination with a user behavior graph (containing operation nodes such as document editing and table processing and associated relationships), and operation association logic between models is determined. Then, a target habit feature vector of the model is analyzed, a high-frequency software set (such as word processing software and spreadsheet software) is extracted in combination with software usage attributes of the operation nodes in the graph, and an operation object association graph is constructed. Then, a generator is constructed based on a time sequence behavior dependency relationship and graph path mining results (such as a high-frequency path of "starting software-inputting content-saving a document"), various operation instructions are generated, and finally a simulated operation habit sequence set is combined after multidimensional verification and correction. In this embodiment, step S120 can include sub-steps S121-S124, which are described in detail as follows. Step S121: Selecting a typical user habit model corresponding to each user group label from the typical user habit model library, and constructing a model association network in combination with a user behavior graph as a basis for generating a simulated operation habit sequence. In this embodiment, in an office scenario, models corresponding to labels such as "word processing high frequency type", "spreadsheet high frequency type", and "mail communication high frequency type" are selected from a typical user habit model library. The user behavior graph contains operation nodes such as "starting word processing software", "keyboard inputting a document", and "saving a document" and associated edges (such as "starting software" is associated with "inputting a document"). When constructing the model association network, the selected models are used as nodes, the network edges are constructed according to the coincidence degree of the operation paths corresponding to each model in the graph (such as the "word processing" model and the "mail communication" model both contain the operation path of "saving a document"), the association strength of the edges is labeled, and the model association network formed provides a basis for generating a simulated sequence subsequently. Step S122: Analyzing a target habit feature vector of a typical user habit model, extracting a high-frequency software set with a usage frequency higher than a preset frequency as an operation object in combination with the operation node attributes of the user behavior graph, and constructing an operation object association graph. In this embodiment, in an office scenario, a target habit feature vector of the "word processing high frequency type" model is analyzed, and software usage frequency data is extracted. In combination with the "usage frequency attribute" of operation nodes such as "word processing software" and "spreadsheet software" in the user behavior graph, software with a usage frequency higher than a preset frequency is screened out to form a high-frequency software set (such as word processing software, spreadsheet software, and mail software). When constructing the operation object association graph, these software are used as nodes, the edges are constructed according to the association relationship of the software operations in the graph (such as "word processing software" and "mail software" have an association of "document sending"), the association probability of the edges is labeled, and the operation object association graph construction is completed. Step S123: based on the path mining results of the time sequence behavior dependency relationship and the user behavior graph, a user operation sequence generator is constructed to generate a user operation sequence containing user operation instructions conforming to the model characteristics, the user operation instructions including application program startup instructions, input user operation instructions, environment adjustment instructions, and task switching instructions. In this embodiment, for the "word processing high frequency type" model in the office scenario, the time sequence behavior dependency relationship contains the operation interval and transition probability between "starting software", "inputting content", and "saving documents". The user behavior graph path mining results show that the high-frequency path corresponding to this model is "starting word processing software", "keyboard inputting documents", "adjusting screen brightness", and "saving documents". Based on this, the generator is constructed to generate application program startup instructions (starting word processing software), input user operation instructions (keyboard inputting documents), environment adjustment instructions (adjusting screen brightness), and task switching instructions (switching to email software), to form a user operation sequence conforming to the model characteristics.

[0024] In this embodiment, step S123 can include sub-steps S1231-S1237, which are described in detail below. Step S1231: extracting time sequence behavior dependency relationship parameters from the core feature vector of the typical user habit model, the time sequence behavior dependency relationship parameters including operation interval distribution characteristics, operation transition probability matrix, and operation duration time regularity, the operation interval distribution characteristics being obtained by statistically analyzing the time interval frequency distribution of adjacent operations in the user habit data, the operation transition probability matrix being obtained by calculating the transition probability between different operation types through a Markov chain model, and the operation duration time regularity being obtained by analyzing the duration time distribution characteristics of similar operations. In this embodiment, the time sequence behavior dependency relationship parameters are extracted from the core feature vector of the "word processing high frequency type" model in the office scenario. The operation interval distribution characteristics are obtained by statistically analyzing the time interval frequency distribution of adjacent operations such as "starting software" and "inputting content", "inputting content" and "saving documents", etc. in the user habit data of this model; the operation transition probability matrix is obtained by calculating the transition probability between different operation types such as "starting software" to "inputting content", "inputting content" to "adjusting brightness", etc. through a Markov chain model; and the operation duration time regularity is obtained by analyzing the duration time distribution characteristics of similar operations such as "keyboard inputting" and "document saving", etc. to form complete time sequence behavior dependency relationship parameters. Step S1232: constructing a time sequence behavior state transition model based on the time sequence behavior dependency relationship parameters, the time sequence behavior state transition model taking the current operation type as the state, predicting the next hop operation type through the operation transition probability matrix, determining the operation execution time point in combination with the operation interval distribution characteristics, and setting the operation execution duration time according to the operation duration time regularity, the model output containing the time sequence operation framework of operation type, execution time point, and duration time In this embodiment, the time sequence behavior state transition model is constructed based on the "word processing high frequency type" model in the office scenario. The model takes "starting word processing software" as the current state, and predicts the next operation type as "keyboard input document" through the operation transition probability matrix. Combined with the operation interval distribution characteristics, it is determined that "keyboard input" is executed at interval t1 after "starting software"; according to the operation duration law, the "keyboard input" duration is set to t2. The model finally outputs the time sequence operation framework containing information such as "operation type: keyboard input document, execution time point: t1 after starting, duration: t2". Step S1233: Create a multi-level architecture of a user operation sequence generator, which includes a feature input layer, a time sequence decision layer, and an instruction generation layer. The feature input layer receives the core feature vector of the typical user habit model, including the high-frequency software set, the input device preference weight, and the environmental parameter adjustment range.

[0025] The time sequence decision layer loads the time sequence behavior state transition model and outputs the type, timestamp, and associated entity of the next operation based on the current operation state and the parameters of the feature input layer. The instruction generation layer generates specific user operation instructions based on the output of the time sequence decision layer, including instruction identification, operation object, parameter value, and execution sequence. In this embodiment, a multi-level architecture of a user operation sequence generator is designed in the office scenario. The feature input layer receives the core feature vector of the "word processing high frequency type" model, such as the high-frequency software set (word processing, email software), the input device preference weight (keyboard x%, touchpad y%), and the environmental parameter adjustment range (brightness A-B, volume C-D). The time sequence decision layer loads the time sequence behavior state transition model and outputs the next operation type "save document", the timestamp "input after t3", and the associated entity "word processing software" based on the current "keyboard input" state and the input parameters. The instruction generation layer generates instructions accordingly (identification: CMD001, operation object: word processing software, parameter value: default save path, execution sequence: input after t3). Step S1234: Generate application operation instruction sequences through the user operation sequence generator, determine the application operation object based on the high-frequency software set, generate application startup instructions, function module calling instructions, and application closing instructions according to the operation transition probability matrix, and the time sequence of the application startup instruction conforms to the high-frequency startup period rule in the operation interval distribution characteristics. The parameter value of the function module calling instruction is determined according to the function module calling frequency in the model. In this embodiment, the generator determines the operation object based on the high-frequency software set (word processing software, email software) of the "word processing high-frequency type" model in the office scenario. According to the operation transition probability matrix, application program startup instructions (start word processing software, start email software), function module calling instructions (word processing software "paragraph setting", "insert table", email software "add attachment") and closing instructions are generated. The startup instruction timing conforms to the high-frequency startup period "morning t4-t5, afternoon t6-t7" in the operation interval distribution characteristics; the function module calling instruction parameter value (such as "line spacing: default value" in "paragraph setting") is determined according to the module calling frequency in the model. Step S1235: Generate input operation instruction sequence, select keyboard operation or touchpad operation as the main input method according to the input device preference weight, the keyboard operation instruction contains the key combination sequence and the time length parameter, the key combination sequence is generated based on the system shortcut key and the application shortcut key usage frequency in the model, and the time length parameter is set according to the operation duration law; the touchpad operation instruction contains the sliding track parameter, the click times parameter and the pressure parameter, the sliding track parameter is generated through the touchpad sliding track characteristic data in the model, and the pressure parameter reflects the touch pressure characteristic in the input device preference weight. In this embodiment, the input device preference weight of the "word processing high-frequency type" model in the office scenario is keyboard x% and touchpad y%, and the generator selects the keyboard as the main input method. The keyboard operation instruction contains key combination sequences such as "Ctrl+S" and "Ctrl+B" (generated based on the shortcut key usage frequency in the model), and the time length parameter (such as "Ctrl+S" press duration t8) is set according to the operation duration law. If touchpad operation is involved, the instruction contains sliding track parameters (such as "horizontal sliding t9 length", generated based on the model track characteristics), click times parameters (such as "double click") and pressure parameters (such as "medium pressure", reflecting the touch pressure characteristic in the preference weight). Step S1236: Generate environment adjustment instruction sequence, determine the brightness adjustment target value, the volume adjustment target value and the power mode type based on the environment parameter adjustment range, determine the trigger time point of the environment adjustment instruction according to the operation interval distribution characteristics in the time sequence behavior dependency parameter, and the trigger time point is consistent with the timing of the application program operation instruction and the input operation instruction. For example, the brightness adjustment instruction is generated at a preset time interval after the application program is started, and the volume adjustment instruction is generated in the input operation frequency reduction period In this embodiment, the model environment parameter adjustment range of the "word processing high frequency type" in the office scenario is brightness A-B, volume C-D, and power mode "balance mode". The generator determines the brightness adjustment target value (such as A1), the volume target value (such as C1), and the power mode "balance mode" according to this. In combination with the operation interval distribution characteristics, the brightness adjustment instruction is triggered at interval t10 after "starting the word processing software", and the volume adjustment instruction is triggered at the period when "the keyboard input frequency is lower than the threshold value" (such as input interval t11), ensuring that the trigger time point is consistent with the application program and input operation instruction timing. Step S1237: The application program operation instruction sequence, the input operation instruction sequence, and the environment adjustment instruction sequence are time-sequentially integrated by the time-sequential behavior state transition model, each operation instruction contained in the application program operation instruction sequence, the input operation instruction sequence, and the environment adjustment instruction sequence is sorted according to the operation timestamp, and a complete user operation sequence conforming to the behavior characteristics of the typical user habit model is formed, the user operation sequence contains the application program operation instruction, the input operation instruction, and the environment adjustment instruction arranged in order according to the timestamp. In this embodiment, the time-sequential behavior state transition model integrates the three types of instruction sequences of the "word processing high frequency type" model in the office scenario. The timestamps of each instruction are extracted, such as "starting the word processing software (t12)", "keyboard inputting a document (t12+t1)", "brightness adjustment (t12+t10)", and "saving a document (t12+t1+t3)", and sorted in ascending order according to the timestamp to form a complete user operation sequence: t12 starts the software → t12+t1 keyboard input → t12+t10 adjusts the brightness → t12+t1+t3 saves the document, which conforms to the model behavior characteristics. Step S124: The generated user operation sequence is subjected to multi-dimensional rationality verification, the abnormal sequence is corrected in combination with the operation specification library of the user behavior graph, and the user operation sequence that passes the verification is combined into a simulated operation habit sequence set, each user operation sequence corresponds to a typical user habit model, and the simulated operation habit sequence set contains complete operation processes of different user groups in different scenarios. In this embodiment, the operation sequences generated by the "word processing high frequency type" and "electronic form high frequency type" models in the office scenario are subjected to multi-dimensional verification. The logical verification refers to the operation specification library of the user behavior graph (such as "saving a document" after "inputting content"), and the abnormal sequence of "saving after closing the software" is corrected; the time-sequential verification checks whether the interval between "starting the software" and "inputting content" conforms to the high frequency interval range of the graph, and the interval abnormal sequence is corrected. The sequences corresponding to each model that pass the verification are combined to form a simulated operation habit sequence set containing complete operation processes of different user groups in the office scenario.

[0026] Step S130: inject the simulated operation habit sequence set into the notebook system, and synchronously collect system adaptation response data, wherein the system adaptation response data includes resource occupation feature vectors and operation response delay data. The resource occupation feature vectors represent real-time load states of the central processor, the memory, and the graphic processor, and the operation response delay data represents a time interval from the injection of a user operation instruction to the completion of a system function. In this embodiment, when the simulated operation habit sequence set is injected into the notebook system in an office scenario, it is necessary to ensure that the initial state of the notebook system is consistent with the system state commonly used by users in the office scenario, such as pre-installed word processing software, spreadsheet software, and the like, and that the system parameters (such as the default input method and screen resolution) conform to the conventional settings in the office scenario. The injection process needs to be performed in sequence according to the time sequence relationship of the instructions in the simulated operation habit sequence set, so as to avoid distortion of the collected response data caused by disordered execution of the instructions. When the synchronous collection system adaptation response data, the resource occupation feature vectors need to capture the load states of the central processor, the memory, and the graphic processor in real time. For example, when the instruction “start the word processing software” is injected, the load proportion of each core of the central processor, the used capacity proportion of the memory, and the rendering load proportion of the graphic processor are recorded. When the instruction “input document content and insert a table” is injected, the load changes of the above hardware resources are continuously recorded to form a feature vector composed of multiple groups of resource occupation data. The operation response delay data needs to accurately record the time interval from the injection of an instruction to the completion of a system function, such as the time from the issuance of the instruction “start the word processing software” to the complete loading and display of the software main interface, and the time from the issuance of the instruction “save the document” to the system prompt of the completion of saving. The above data needs to be associated with the corresponding operation instruction one by one to ensure the accuracy of subsequent analysis. In this embodiment, step S130 can include sub-steps S131-S136, which are described in detail below. Step S131: configure a notebook test environment, wherein the test environment includes an operating system version, a driver program version, a pre-installed application program version, and hardware configuration parameters, so that the test environment is consistent with the actual use environment of mainstream users. In this embodiment, when the notebook test environment is configured in an office scenario, the operating system version needs to be selected as the mainstream version in the office scenario to ensure that it matches the system version actually used by most users. The driver program version needs to be selected as a stable version corresponding to the hardware (such as a graphics card, a sound card, and a touchpad) to avoid abnormal hardware functions caused by an outdated or too new driver version. The pre-installed application program version needs to be consistent with the mainstream version of the frequently used software in the typical user habit model library, such as a recently updated stable version of word processing software and spreadsheet software, to ensure that the simulated operation instruction can be normally executed. The hardware configuration parameters need to be adjusted to the mainstream configuration level in the office scenario, such as setting the memory capacity to the capacity specification commonly used by most office users, selecting the mainstream solid state disk type for the hard disk and configuring a suitable partition size, setting the graphics card parameters to the performance mode commonly used in the office scenario, and avoiding the test results from deviating from the actual user usage scenario due to excessively high or low hardware configuration. Step S132: calling a user operation sequence injection tool, the user operation sequence injection tool including a software interface calling module, an input device simulation module, and an environment parameter adjusting module. The software interface calling module starts and controls the application program through the system API, the input device simulation module simulates the input signals of the keyboard and touchpad through the hardware interface, and the environment parameter adjusting module adjusts the parameters such as brightness and volume through the system setting interface. In this embodiment, when the user operation sequence injection tool is called in the office scenario, the software interface calling module needs to start and control the application program through the API of the notebook system. For example, the "start word processing software" instruction is implemented by calling the start API of the word processing software, and the "save document" instruction is implemented by calling the "save document" API of the software, so as to ensure that the starting and operation of the application program are consistent with the effect of manual operation of the user. The input device simulation module needs to simulate the input signals of the keyboard and touchpad through the hardware interface. When simulating the keyboard input, the trigger signals corresponding to the real keyboard keys need to be generated, such as generating the trigger signals of the corresponding keys when simulating the "Ctrl+S" shortcut key input; when simulating the touchpad input, signals consistent with the touchpad sliding and clicking need to be generated, such as generating the signal of the corresponding sliding track when simulating the touchpad sliding to select the text, so as to ensure that the input operation can be accurately recognized by the system. The environment parameter adjusting module needs to adjust the parameters such as brightness and volume through the system setting interface. For example, the screen brightness adjustment instruction is implemented by calling the screen brightness adjustment interface of the system, and the volume adjustment instruction is implemented by calling the volume control interface of the system, and the adjustment process needs to be consistent with the response characteristics of the original system adjustment function, so as to avoid the situation of adjustment delay or parameter abnormality. Step S133: loading the simulated operation habit sequence set into the user operation sequence injection tool, parsing the instruction type, operation object, parameter value, and time sequence information of each user operation sequence, generating an instruction execution plan, and the instruction execution plan includes the execution order and time interval of each user operation instruction. In this embodiment, after loading the simulated operation habit sequence set into the injection tool in the office scenario, the tool needs to first analyze the specific information of each user operation sequence. The instruction type analysis needs to distinguish between application startup instructions, input instructions, environment adjustment instructions, etc. For example, "start word processing software" is an application startup instruction, and "keyboard input'meeting minutes'" is an input instruction. The operation object analysis needs to clearly identify the target of the instruction, such as the operation object of the "save document" instruction being the currently edited word processing document. The parameter value analysis needs to extract the specific setting parameters in the instruction, such as the target range intermediate value in "adjust the screen brightness to the target range intermediate value". The timing information analysis needs to obtain the time interval between instructions, such as the time interval between "start word processing software" and "start inputting document content". The instruction execution plan generated based on the analysis results needs to clearly define the execution order of each user operation instruction to ensure consistency with the user's operation process. At the same time, it needs to accurately set the time interval between instructions, such as setting the time interval of the corresponding instruction according to the user habit data "start inputting x time after starting the software" to avoid excessive deviation between simulated operation and real user operation due to unreasonable time interval. Step S134: Start the user operation sequence injection tool and sequentially inject user operation instructions into the notebook system according to the instruction execution plan. During the injection process, maintain consistent time intervals with the timing behavior dependency relationship in the typical user habit model to ensure the authenticity of the operation process In this embodiment, after starting the injection tool in the office scenario, the tool needs to strictly follow the order of the instruction execution plan to inject user operation instructions. For example, follow the order of "start word processing software - keyboard input'meeting minutes' - adjust screen brightness - save document - close software" to sequentially inject, and do not skip any instruction or adjust the instruction order. During the injection process, the time interval between instructions needs to be consistent with the timing behavior dependency relationship in the typical user habit model. For example, if the model shows that the user starts inputting after starting the word processing software with an average interval of t1, then after injecting the "start word processing software" instruction, the "keyboard input'meeting minutes'" instruction needs to be injected after t1. After the user finishes inputting, adjust the screen brightness with an average interval of t2, then inject the brightness adjustment instruction after the input instruction with an interval of t2. By maintaining consistent time intervals, the simulated operation process is highly consistent with the real user operation process. Step S135: Start the response data collector while injecting user operation instructions, which collects resource occupation feature vectors through the kernel mode monitoring driver, including central processor core occupation rate, memory usage, graphics processor rendering frame rate, and storage device read-write throughput, and collects operation response delay data through the user mode monitoring program, including application startup time, window switching delay, and input response time. In this embodiment, when the response data collector is started in an office scenario, the kernel mode monitoring driver needs to directly interact with the notebook hardware kernel to collect resource occupation feature vectors in real time. When collecting the CPU core occupation rate, the load proportion of each core in a unit of time needs to be obtained to form multi-core occupation rate data. When collecting the memory usage, the used memory capacity, idle memory capacity, and memory read-write speed need to be recorded to form memory usage feature data. When collecting the GPU rendering frame rate, the number of rendering frames of the GPU in a unit of time needs to be counted, especially when executing instructions such as “insert table” and “display high-definition picture” that require graphics processing, the frame rate change needs to be recorded. When collecting the storage device read-write throughput, the read-write speed of the storage device when executing instructions such as “save document” and “open file” needs to be recorded to form storage device load data. The above data collectively constitute the resource occupation feature vector. The user mode monitoring program needs to collect operation response delay data through the system user interface. When collecting the application startup time, the time from the issuance of the application startup instruction to the complete loading and operability of the application main interface is recorded. When collecting the window switching delay, the time from the issuance of the window switching instruction to the completion of the interface refresh of the target window that becomes the active window is recorded. When collecting the input response time, the time from the issuance of the keyboard input, touchpad operation, or other instruction to the feedback of the operation result on the system interface is recorded, such as the time from the input of a text instruction to the display of the text in the document. Step S136: Align the collected resource occupation feature vector and operation response delay data with the instruction timestamp of the user operation sequence to generate a system adaptation response data set with a time marker. Each user operation instruction corresponds to a set of associated response data, and the system adaptation response data set includes multiple system adaptation response data. In this embodiment, when data alignment is performed in an office scenario, the issuance timestamp of each instruction in the user operation sequence needs to be extracted first, and then the collection timestamp corresponding to the collected resource occupation feature vector and the record timestamp corresponding to the operation response delay data are matched with the instruction timestamp. For example, the issuance timestamp of the “start word processing software” instruction is T1, and the CPU, memory, and GPU resource data collected at T1 and within a short period of time before and after T1 are taken as the resource occupation feature vector corresponding to the instruction, and the application startup time recorded from T1 is taken as the operation response delay data corresponding to the instruction. By time stamp alignment, it is ensured that each user operation instruction has a set of uniquely associated response data, such as the "keyboard input'meeting minutes'" instruction corresponding to a set of resource occupation data and input response time data during input, and the "save document" instruction corresponding to a set of resource occupation data and save response time data during saving. Integrating all instruction corresponding response data generates a time marked system adaptation response data set, which records the system response corresponding to each simulation operation instruction in the office scenario. Step S140: Multi-dimensional correlation analysis is performed on the system adaptation response data and the typical user habit model library. By comparing the system adaptation response data corresponding to different typical user habit models, a user habit adaptation degree index and a key adaptation bottleneck path are calculated. The user habit adaptation degree index is used to represent the adaptation ability of the notebook system to a specific user habit, and the key adaptation bottleneck path is used to represent the correlation between the user operation sequence and the system component that causes abnormal system response. In this embodiment, when performing multi-dimensional correlation analysis in the office scenario, the system adaptation response data needs to be grouped according to the corresponding typical user habit model, such as grouping the response data generated by the simulation operation instruction corresponding to the "word processing-keyboard dependent type" model into one group, and grouping the response data corresponding to the "spreadsheet-touchpad operation type" model into another group. When comparing different groups of data, the differences in resource occupation characteristic vector and operation response delay data are analyzed, such as comparing the central processing unit occupation rate, memory usage peak value, application startup time and other indicators in the application startup stage of the two groups of data to determine the adaptation differences of the notebook system to different user habits. When calculating the user habit adaptation degree index, the performance dimensions that users pay attention to in the office scenario need to be combined, such as application response speed, hardware resource stability, etc. The resource occupation characteristics and operation response delay data are assigned weights according to user sensitivity and then comprehensively evaluated to obtain an index that reflects the adaptation ability of the system to a specific user habit. When analyzing the key adaptation bottleneck path, the abnormal response data needs to be associated with the corresponding user operation sequence and system component, such as the abnormal increase in memory occupation rate and the increase in response delay corresponding to the "inserting a large amount of data into a spreadsheet" instruction in a group of data. The correlation between the operation sequence and the memory component needs to be located, and it is determined that the insufficient memory resource is the adaptation bottleneck to form the key adaptation bottleneck path. In this embodiment, step S140 can include sub-steps S141-S145, which are described in detail below. Step S141: Grouping the system adaptation response data including a plurality of system adaptation response data according to the corresponding typical user habit model to obtain a dedicated response data group for each typical user habit model. The dedicated response data group includes a sequence of resource occupation characteristic vectors and a sequence of operation response delay data corresponding to the typical user habit model. In this embodiment, in the office scenario, when grouping data, the typical user habit model identifier corresponding to each system adaptation response data needs to be determined first, such as "BM001" (corresponding to the "word processing-keyboard dependent" model), "BM002" (corresponding to the "spreadsheet-touchpad operation" model), and the like marked in the data. According to the model identifier, the data is classified, and a dedicated response data group is constructed for each model. The resource occupation feature vector sequence in each dedicated response data group needs to be arranged in the order of operation instruction execution, such as the dedicated group of the "BM001" model, which contains the resource occupation feature vectors corresponding to the instructions "start word processing software", "input document content", "save document", and the like in sequence; the operation response delay data sequence is also arranged in the order of instructions, containing application start time, input response time, and the like data corresponding to each instruction, to ensure that each group of data can fully reflect the system response process under the simulation operation of the corresponding model. Step S142: statistical analysis is performed on each dedicated response data group, and the resource occupation mean, resource occupation fluctuation coefficient, response delay mean, and response delay jitter rate are calculated. Among them, the resource occupation mean reflects the average load level of the system under the user habit model, the resource occupation fluctuation coefficient reflects the stability of the load, the response delay mean reflects the overall response speed, and the response delay jitter rate reflects the smoothness of the response.

[0027] In this embodiment, in the office scenario, when statistical analysis is performed on the dedicated response data group, for resource occupation mean calculation, all data of the resource occupation feature vector sequence in the dedicated response data group need to be extracted, covering the central processor core occupation rate, memory usage, graphics processor rendering frame rate, and storage device read-write throughput, and the average of all data under each dimension is calculated to obtain the resource occupation mean of each dimension, and the average of each dimension can reflect the average load level of the system under the corresponding typical user habit model, for example, the resource occupation mean of the "word processing-keyboard dependent" model dedicated group can reflect the average load situation of the system processing the high-frequency text input, document saving, and the like of this type of user. When calculating the resource occupation fluctuation coefficient, the dispersion degree of each resource dimension data needs to be determined first, and then the resource occupation mean of the corresponding dimension is combined to obtain the fluctuation coefficient by reflecting the relationship between the data dispersion degree and the mean. The smaller the fluctuation coefficient, the more stable the load change of the corresponding resource of the system under the user habit model, and the higher the stability, for example, if the memory occupation fluctuation coefficient of the "spreadsheet-touchpad operation" model dedicated group is small, it indicates that the memory load is stable when the system processes the table data editing and touchpad operation of this type of user. The response delay mean value calculation needs to extract all data of the operation response delay data sequence in the exclusive response data set, including application startup time, window switching delay, and input response time, etc. The average of each type of delay data is taken to obtain the response delay mean value of each type. The response delay mean value composed of the averages of each type can reflect the overall response speed of the system under the corresponding user habit model. For example, the response delay mean value of the "email communication-external device assisted type" model exclusive group can reflect the overall response speed of the system in processing the email sending and external mouse operation instructions of this type of user. The response delay jitter rate calculation needs to analyze the variation range of each type of response delay data. The jitter rate is obtained by measuring the degree of delay data deviating from the response delay mean value. The lower the jitter rate, the more stable the response time of the system to the operation of this type of user. For example, if the input response delay jitter rate of the "word processing-keyboard dependent type" model exclusive group is low, it means that when the user continuously inputs on the keyboard, the time difference of the system feedback for each input is small, and the response is stable. Step S143: Call the pre-constructed user habit adaptation degree index calculation model to weight and fuse the resource occupation characteristics and response delay characteristics to obtain the user habit adaptation degree index. The weights corresponding to the resource occupation characteristics and the response delay characteristics are determined according to the sensitivity of the user to different performance indicators. For example, the sensitivity of the office user to the response delay is higher than that of the resource occupation, and the sensitivity of the entertainment user to the frame rate of the graphics processing unit is higher than that of other indicators. In this embodiment, when the user habit adaptation degree index calculation model is called in the office scenario, the specific dimensions of the resource occupation characteristics and the response delay characteristics input into the model need to be determined first. The resource occupation characteristics include the statistical values of the central processing unit load, memory usage, and graphics processing unit load. The response delay characteristics include the statistical values of the application startup delay, input response delay, and window switching delay. When determining the weights, based on the performance requirement research results of the user in the office scenario, the office user pays more attention to the feedback speed of the system after issuing an instruction when processing documents, emails, and other operations. Therefore, a higher weight is assigned to the response delay characteristics, and a relatively lower weight is assigned to the resource occupation characteristics. For example, if the sensitivity of the user to the response delay is x times that of the resource occupation, the weight of the response delay characteristics is x times that of the resource occupation characteristics. The model multiplies each dimension of the resource occupation characteristics by the corresponding weight, multiplies each dimension of the response delay characteristics by the corresponding weight, and then fuses the weighted results of the two types to obtain a single user habit adaptation degree index. The higher the adaptation degree index value, the stronger the adaptation ability of the notebook system to the operation habits of the user group corresponding to the typical user habit model. For example, a high adaptation degree index of the "word processing-keyboard dependent type" model indicates that the system can better adapt to the habits of this type of user in high-frequency text input and document operation. Step S144: By comparing the user habit adaptation index of different typical user habit models, identify the typical user habit model whose adaptation index is lower than the set threshold, and determine the adaptation short board user group of the notebook system. In this embodiment, when comparing the adaptation indexes of different typical user habit models in the office scenario, the adaptation index threshold needs to be set first. The threshold is determined based on the acceptable system performance level of most users in the office scenario. If the adaptation index of a certain typical user habit model is lower than the threshold, it means that the system's adaptation ability to the operation habits of the user group corresponding to this model is insufficient. This user group is the adaptation short board user group of the notebook system. For example, by comparing the adaptation indexes of the "word processing-keyboard dependent type", "spreadsheet-touchpad operation type", and "email communication-external device assisted type" models, if the index of the "spreadsheet-touchpad operation type" model is lower than the threshold, it indicates that the system's performance in handling high-frequency table data editing, touchpad sliding to select cells, and other operations of this type of user is not acceptable to most users. This type of office user who uses spreadsheets and relies on touchpad operation is the adaptation short board user group of the system. Step S145: Perform bottleneck path analysis on the exclusive response data set corresponding to the adaptation short board user group, and locate the target operation link and target system component that cause the adaptation index to be lower than the set threshold by associating high-frequency user operation instructions in the user operation sequence with resource occupation abnormal data. For example, frequent application switching leads to high memory occupation, or high brightness adjustment leads to reduced battery life. In this embodiment, when analyzing the exclusive response data set corresponding to the adaptation short board user group in the office scenario, first extract the high-frequency operation instructions of the user operation sequence in this exclusive group. For example, the high-frequency instructions of the "spreadsheet-touchpad operation type" short board group may include "inserting multiple rows of data", "touchpad sliding to select table content", "saving large table files", etc. Next, associate the high-frequency operation instructions with the corresponding resource occupation data and check whether there are resource occupation abnormalities when each high-frequency instruction is executed. For example, when the "insert multiple rows of data" instruction is executed, if the memory occupation data is far beyond the normal range and continues to rise, it indicates that there is a resource occupation abnormality in the execution link of this instruction. When the "save large table file" instruction is executed, if the storage device read-write throughput is abnormally low and the response delay is long, it indicates that there is a storage performance abnormality in the execution link of this instruction. According to the resource occupation abnormal type, the target system component is located, the memory occupation abnormality corresponds to the memory component, the storage read-write abnormality corresponds to the storage device component, and the operation instruction link associated with the above-mentioned abnormality is the target operation link. The target operation link and the target system component jointly constitute a key adaptation bottleneck path, such as the "insert multiple rows of data" instruction (target operation link) - memory component (target system component). This path is the key bottleneck path that causes the low adaptation degree of the "spreadsheet-touchpad operation type" model. Step S150: generating a user habit adaptation optimization report based on the user habit adaptation degree index and the key adaptation bottleneck path, the user habit adaptation optimization report including user habit type identification, system adaptation short board analysis, and targeted adjustment suggestions, for guiding the software and hardware parameter adaptation optimization of the notebook system. In this embodiment, when generating a user habit adaptation optimization report in an office scenario, the user habit type identification needs to clearly correspond to each typical user habit model, such as "BM001 (word processing-keyboard dependent type)" and "BM002 (spreadsheet-touchpad operation type)", to ensure that the report can clearly distinguish different user groups.

[0028] The system adaptation short board analysis part needs to combine the adaptation degree index and the key adaptation bottleneck path of each model to explain the adaptation short board performance in detail. For example, the adaptation degree of "BM002 (spreadsheet-touchpad operation type)" is lower than the threshold value, and the short board performance is that when executing the "insert multiple rows of data" and "save large table file" instructions, the memory occupation is too high and the storage read-write is slow, resulting in long operation response delay. The targeted adjustment suggestions need to propose software and hardware optimization schemes for the key adaptation bottleneck path. In terms of hardware, it can suggest adjusting memory configuration parameters and optimizing storage device read-write mode. In terms of software, it can suggest optimizing the data processing algorithm of the spreadsheet software and improving the touchpad operation response logic. The report as a whole needs to have a clear structure, so that technical personnel can accurately carry out software and hardware parameter adaptation optimization of the notebook system according to the report content, and improve the adaptation ability of the system to the habits of each office user group. In this embodiment, step S150 can include sub-steps S151-S155, which are described in detail below. Step S151: generating an adaptation degree distribution heat map based on the user habit adaptation degree index of each typical user habit model, the horizontal axis of the adaptation degree distribution heat map being the user habit type, the vertical axis being the performance index dimension, and the color depth representing the adaptation degree, directly showing the adaptation of different user groups In this embodiment, when generating the fitness distribution heat map in the office scenario, the horizontal axis needs to list all the user habit types corresponding to the typical user habit models, such as "word processing-keyboard dependent type", "spreadsheet-touchpad operation type", "email communication-external device assisted type", and the like, and each type occupies an independent interval on the horizontal axis. The vertical axis needs to determine the performance indicator dimension, covering the resource occupation related dimension (such as central processing unit average load, memory average usage) and the response delay related dimension (such as application startup average delay, input response average delay), and each dimension occupies an independent row on the vertical axis. The color setting of the heat map needs to follow a unified rule, and the deeper the color, the higher the fitness, and the lighter the color, the lower the fitness, for example, a gradient color system from light blue to dark blue is adopted, light blue corresponds to low fitness, and dark blue corresponds to high fitness. The fitness indicators of each user habit type in each performance indicator dimension are mapped to the corresponding color according to the color rule to fill the corresponding cells of the heat map, and the formed heat map can enable the technical personnel to directly see which performance dimensions have high fitness and which dimensions have low fitness for a certain user habit type, such as "word processing-keyboard dependent type" has deep color (high fitness) in the input response delay dimension and light color (low fitness) in the memory usage dimension. Step S152: generating an operation flow optimization strategy for the trigger operation node in the key adaptation bottleneck path, the operation flow optimization strategy including adjusting software default settings, optimizing shortcut key layout, and simplifying multi-task switching steps, and adding application quick switching gesture support for the user group frequently switching applications. In this embodiment, when generating the strategy for the trigger operation node in the key adaptation bottleneck path in the office scenario, if the trigger operation node is the "insert multiple rows of data" instruction of the "spreadsheet-touchpad operation type" group, adjusting the software default settings can optimize the default cache mechanism of the "insert data" function in the spreadsheet software, reducing the temporary data storage redundancy when inserting data; optimizing the shortcut key layout can set more easily operated shortcut key combinations for the "insert multiple rows of data" function in the spreadsheet software, such as setting shortcut keys in combination with the commonly used key positions of office users. If the trigger operation node is the "frequent switching between email client and word processing software" instruction of the "email communication-external device assisted type" group, simplifying the multi-task switching steps can optimize the task switching logic of the system, reducing the process loading link when switching; adding application quick switching gesture support can set a specific sliding gesture (such as three-finger left and right sliding) through the touchpad to directly switch the two high-frequency applications in the system, without the need to click to switch through the taskbar, thereby improving the switching efficiency. Step S153: generating a hardware resource configuration optimization strategy for the resource consumption node in the key adaptive bottleneck path, the hardware resource configuration optimization strategy including a memory allocation adjustment strategy, a graphics processor parameter optimization strategy, and a storage device cache optimization strategy. In this embodiment, when generating a strategy for a resource consumption node in an office scenario, if the resource consumption node is a memory component (e.g., the "insert multiple rows of data" instruction causes abnormal memory occupation), the memory allocation adjustment strategy can set the initial memory allocation amount when the spreadsheet software is started, appropriately increase the initial allocation amount according to the memory demand of the frequently operated software, and optimize the memory recycling mechanism of the system to release the memory space that is no longer used by the software in time. If the resource consumption node is a graphics processor component (e.g., the "display high-definition table chart" instruction causes the graphics processor to be overloaded), the graphics processor parameter optimization strategy can adjust the rendering mode of the graphics processor, use a more efficient rendering algorithm for the table chart display scenario to reduce resource consumption in the rendering process, and set the performance priority of the graphics processor for the office software to ensure that the graphics processor can be allocated resources preferentially when the office software is used. If the resource consumption node is a storage device component (e.g., the "save large table file" instruction causes slow storage read and write), the storage device cache optimization strategy can increase the cache capacity of the storage device to improve the read and write speed of temporary data, and optimize the read and write scheduling algorithm of the storage device to preferentially process the file saving, opening, and other instructions of the office software to reduce the instruction queuing waiting time. Step S154: generating a software performance optimization strategy for the response delay node in the key adaptive bottleneck path. For example, the software performance optimization strategy can include specific strategies for optimizing the application startup loading process, reducing background process resource occupation, and improving interface rendering efficiency to optimize the application cold start path or increase the preloading mechanism for the user group with slow application startup. In this embodiment, when generating a strategy for a response delay node in an office scenario, if the response delay node is application startup delay (e.g., the "start large spreadsheet software" instruction responds slowly), the optimized application startup loading process can simplify unnecessary plugins and services loaded during software startup and only load core functional modules; the preloading mechanism can load part of the data required for startup in advance to shorten the waiting time of the user during actual startup according to the user habit to predict the spreadsheet software that the user is likely to start when the system is idle. If the response delay node is the input response delay (such as the feedback of the "keyboard input table data" instruction is slow), reducing the resource occupation of the background process can set the running time of the office software, automatically limit the resource use of other unnecessary background processes in the system, and ensure that the office software can preferentially obtain central processing units, memories and other resources; improving the interface rendering efficiency can optimize the interface drawing logic of the office software, reduce the rendering number of interface refreshing after input operation, and improve the input feedback speed. Step S155: integrating based on the adaptation degree distribution heat map, the adaptation short board analysis, the operation process optimization strategy, the hardware resource configuration optimization strategy and the software performance optimization strategy, and generating a user habit adaptation optimization report.

[0029] Among them, the user habit adaptation optimization report contains execution priority ranking, and the priority is determined according to the adaptation degree improvement amplitude and the implementation cost, so that each strategy can efficiently solve the core adaptation problem In this embodiment, when the user habit adaptation optimization report is generated in the office scene, the content is first organized according to the structure of "user habit type identifier-adaptation degree situation (including heat map corresponding area description)-adaptation short board analysis-corresponding optimization strategy", and each user habit type is a separate section to ensure that the report is clear and orderly. When the execution priority is ranked, the improvement amplitude of the adaptation degree after the implementation of each optimization strategy is evaluated first, such as a strategy that is expected to improve the "spreadsheet-touchpad operation type" model adaptation degree by x units after implementation, and another strategy that is expected to improve the adaptation degree by y units (x is greater than y), then the former has a higher improvement amplitude; then the implementation cost is evaluated, including hardware cost (such as the cost of increasing memory), software development cost (such as the development resources required to optimize the application startup process), and the strategy with lower cost is easier to implement. The priority is determined by combining the two, the strategy with high improvement amplitude and low implementation cost is listed as the highest priority and is executed first; the strategy with high improvement amplitude and high implementation cost is listed as the medium priority and is arranged according to the resource situation; the strategy with low improvement amplitude and high implementation cost is listed as the low priority and is considered later. For example, the "optimize spreadsheet software startup loading process" strategy (high improvement amplitude and low cost) is listed as the highest priority, and the "increase storage device cache capacity" strategy (high improvement amplitude and high cost) is listed as the medium priority, so that each strategy can efficiently solve the core adaptation problem according to the priority, and improve the adaptation ability of the system to each user group in the office scene. On the basis of the above content, such as Figure 3As shown is a notebook adaptation test system provided by an embodiment of the present application. The system includes a processor, a machine readable storage medium, an input and output device, and the like. The machine readable storage medium is connected to the processor and is used to store programs, instructions or codes. The processor is used to execute the programs, instructions or codes in the machine readable storage medium to implement the notebook adaptation test method with user habits. The notebook adaptation test system with user habits can be understood as the notebook adaptation test system with user habits provided by the present application. Figure 1 The system platform or one part of the notebook in the application scenario or the system platform or the notebook itself.

[0030] The machine readable storage medium can be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read only memory (PROM), an erasable programmable read only memory (EPROM), an electric erasable programmable read only memory (EEPROM), and the like. The machine readable storage medium is used to store programs. The processor executes the programs after receiving an execution instruction.

[0031] The processor can be an integrated circuit chip with a signal processing capability. The processor can be, but is not limited to, a general purpose processor, including a central processing unit (CPU), a network processor (NP), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), and the like.

[0032] In summary, the notebook adaptation test method and system provided by the embodiments of the present application combine user habits, collect user habit data in multiple scenarios, and use clustering algorithms to construct a typical user habit model library, which can accurately depict the differences in multiple dimensions such as software use preferences, input method characteristics, and environment setting behaviors of different user groups. The directional analysis of user operation behavior characteristics provides a reliable basis for subsequent testing based on real user behavior patterns, greatly improving the relevance and effectiveness of the test. In addition, the simulation operation habit sequence set is generated based on the typical user habit model library, which closely matches the actual behavior patterns of different user groups. Injecting it into the notebook system and synchronously collecting system adaptation response data, including resource occupation characteristic vectors and operation response delay data, can truly reflect the performance of the notebook system when facing various actual user habit operations. Compared with traditional standardized testing, this test method based on real user behavior simulation can better reveal potential problems in actual use of the system.

[0033] At the same time, the system adaptation response data and the typical user habit model library are analyzed in multiple dimensions, and the user habit adaptation degree index and the key adaptation bottleneck path are accurately calculated. The user habit adaptation degree index clearly quantifies the adaptation ability of the notebook system to specific user habits, and the key adaptation bottleneck path clearly points out the association between the specific user operation sequence and the system components that cause abnormal system response. This provides key clues for understanding system performance bottlenecks, which helps to optimize targetedly. Finally, the user habit adaptation optimization report generated based on the above analysis results, for example, contains detailed user habit type identification, system adaptation short board analysis, and targeted adjustment suggestions, etc., provides guidance for the software and hardware parameter adaptation optimization of the notebook system, so that the optimization work can be targeted, effectively improves the adaptability of the system and the user habits, and further improves the performance and user experience of the notebook system in actual use scenarios, and enhances the competitiveness of the product in the market.

[0034] It should be noted that in the data collection, the necessary means will be taken to follow the principle of legal data collection. For example, before data collection, the user is explicitly informed that the purpose of data collection is to test the notebook adaptation to improve product performance and user experience, and the explicit consent of the user is obtained. The collection process is strictly limited within the scope of the established multi-scenario, such as office, entertainment and mobile scenarios, and only collects information related to software usage preference data (including application startup frequency, usage time and function module call record), input method feature data (keyboard key stroke frequency, touchpad sliding track and external device operation mode), environment setting behavior data (screen brightness adjustment period, volume variation law and power management mode switching record), and does not collect other irrelevant or sensitive information beyond the scope. At the same time, safe and reliable data collection technology is adopted to ensure the safety of data transmission and storage process, prevent data leakage and illegal acquisition, and protect user data privacy and the legality of data collection.

[0035] In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments. The embodiments of the present application, the implementation manners and the related technical features can be combined, replaced with each other without conflict. The above is only a preferred embodiment of the present application, and does not limit the present application in any form, but any simple modification, equivalent change and modification made according to the technical essence of the present application to the above embodiments, which does not deviate from the technical solution content of the present application, still belongs to the scope of the technical solution of the present application.

Claims

1. A notebook adaptation test method incorporating user habits, characterized by, The method comprises: Collecting multi-scene user habit data and constructing a typical user habit model library, classifying the user habit data by a clustering algorithm to obtain a plurality of typical user habit models, the typical user habit models corresponding to operation behavior characteristics of different user groups, the user habit data including software use preference data, input method characteristic data and environment setting behavior data; Generating a simulated operation habit sequence set based on the typical user habit model library, the simulated operation habit sequence set being constructed according to operation preference characteristics, input timing characteristics and environment adjustment rules in each typical user habit model, and including user operation instructions conforming to behavior modes of different user groups; Injecting the simulated operation habit sequence set into a notebook system and synchronously collecting system adaptation response data, the system adaptation response data including resource occupation characteristic vectors and operation response delay data; Performing multi-dimensional correlation analysis on the system adaptation response data and the typical user habit model library, calculating a user habit adaptation degree index and a key adaptation bottleneck path by comparing system adaptation response data corresponding to different typical user habit models; Generating a user habit adaptation optimization report based on the user habit adaptation degree index and the key adaptation bottleneck path.

2. The notebook adaptation test method according to claim 1, wherein, The collecting multi-scene user habit data and constructing a typical user habit model library comprises: Collecting user habit data in a plurality of preset scenes, the software use preference data including application program startup frequency, use time length and function module call records, the input method characteristic data including keyboard key stroke frequency, touchpad sliding track and external device operation mode, and the environment setting behavior data including screen brightness adjustment period, volume variation rule and power management mode switching records; Preprocessing the user habit data and classifying the preprocessed user habit data by a clustering algorithm, clustering the user habit data based on similarity between user behavior characteristic vectors, and determining the number of clustering clusters according to data density distribution in the clustering process, so that the clustering result can distinguish user groups with different operation preferences; Extracting features from each clustering result to obtain target habit characteristics of a typical user habit model, the target habit characteristics including a high-frequency software set, an input device preference weight, an environment parameter adjustment range and a timing behavior dependency relationship; Adding each typical user habit model to a preset model library to obtain a typical user habit model library, the typical user habit model library including a model identifier, a user group label, a target habit characteristic vector and an applicable scene description. 3.The notebook adaptation test method incorporating user habits according to claim 2, wherein, The classifying the preprocessed user habit data by a clustering algorithm, clustering the user habit data based on similarity between user behavior characteristic vectors, comprises: Converting the preprocessed user habit data into high-dimensional characteristic vectors, the dimension of each characteristic vector corresponding to a user habit characteristic, and the dimensions of the characteristic vectors including software use frequency dimension, input method proportion dimension, environment setting parameter dimension and timing behavior transition dimension; The high-dimensional feature vectors are clustered by using a density clustering algorithm, the density reachability of the data points in the feature space is calculated, the points with the same density are divided into a cluster, and the clustering radius and the minimum sample number are set, the clustering radius is determined by the distance distribution statistics of the feature space, and the minimum sample number is set according to the size of the user group; The effectiveness of the clustering result is evaluated, the clustering similarity of each data point is calculated by using a silhouette coefficient, the silhouette coefficient is used to evaluate the clustering effect by using the similarity within the sample and the similarity between the clusters, and the abnormal cluster with a silhouette coefficient lower than a set threshold is removed to obtain an effective cluster; The effective cluster is named and feature-labeled, a user group label is assigned to each effective cluster according to the common feature of the user habit data in the effective cluster, the user group label is determined based on the common feature of the user habit data in the effective cluster, and the feature labeling includes a description of the typical operation behavior of the user in the effective cluster.

4. The notebook adaptation test method according to claim 1, wherein, The simulation operation habit sequence set is generated based on the typical user habit model library, including: A typical user habit model corresponding to each user group label is selected from the typical user habit model library, and a model association network is constructed by combining a user behavior graph, which is used as a generation basis for the simulation operation habit sequence; The target habit feature vector of the typical user habit model is analyzed, a high-frequency software set with a frequency higher than a preset frequency is extracted as an operation object by combining the operation node attribute of the user behavior graph, and an operation object association graph is constructed; Based on the time sequence behavior dependency relationship and the path mining result of the user behavior graph, a user operation sequence generator is constructed to generate a user operation sequence containing user operation instructions conforming to the model feature, the user operation instructions include application program start instructions, input user operation instructions, environment adjustment instructions and task switching instructions; The generated user operation sequence is subjected to multi-dimensional rationality verification, the abnormal sequence is corrected by combining the operation specification library of the user behavior graph, and the user operation sequence passing the verification is combined into a simulation operation habit sequence set, each user operation sequence corresponds to a typical user habit model, and the simulation operation habit sequence set contains complete operation processes of different user groups in different scenarios.

5. The notebook adaptation test method according to claim 4, wherein, Based on the time sequence behavior dependency relationship and the path mining result of the user behavior graph, the user operation sequence generator is constructed to generate a user operation sequence containing user operation instructions conforming to the model feature, including: The time sequence behavior dependency relationship parameters are extracted from the core feature vector of the typical user habit model, the time sequence behavior dependency relationship parameters include operation interval distribution features, operation transfer probability matrices and operation duration rules; wherein, the operation interval distribution features are obtained by statistically analyzing the time interval frequency distribution of adjacent operations in the user habit data, the operation transfer probability matrices are obtained by calculating the transfer probability between different operation types by using a Markov chain model, and the operation duration rules are obtained by analyzing the duration distribution features of similar operations; construct a time sequence behavior state transition model based on the time sequence behavior dependency relationship parameters, the time sequence behavior state transition model taking a current operation type as a state, predicting a next-hop operation type through an operation transition probability matrix, determining an operation execution time point in combination with an operation interval distribution feature, and setting a duration of operation execution according to an operation duration rule; create a multi-level architecture of a user operation sequence generator, the multi-level architecture including a feature input layer, a time sequence decision layer, and an instruction generation layer, the feature input layer receiving a core feature vector of a typical user habit model, including a high-frequency software set, an input device preference weight, and an environmental parameter adjustment range; the time sequence decision layer loading the time sequence behavior state transition model, outputting a type, a timestamp, and an associated entity of a next operation based on a current operation state and parameters of the feature input layer; and the instruction generation layer generating specific user operation instructions according to an output of the time sequence decision layer, including an instruction identifier, an operation object, a parameter value, and an execution time sequence; generate an application program operation instruction sequence through the user operation sequence generator, determine an application program operation object based on the high-frequency software set, and generate an application program start instruction, a function module call instruction, and an application program close instruction according to the operation transition probability matrix; generate an input operation instruction sequence, select a keyboard operation or a touchpad operation as a main input method according to the input device preference weight, the keyboard operation instruction including a key combination sequence and a duration parameter, the key combination sequence being generated based on system shortcut keys and application shortcut keys in the model and a frequency of use, and the duration parameter being set according to the operation duration rule; and the touchpad operation instruction including a sliding track parameter, a click count parameter, and a pressure parameter; generate an environmental adjustment instruction sequence, determine a brightness adjustment target value, a volume adjustment target value, and a power mode type based on the environmental parameter adjustment range, and determine a trigger time point of the environmental adjustment instruction in combination with an operation interval distribution feature in the time sequence behavior dependency relationship parameters; integrate the application program operation instruction sequence, the input operation instruction sequence, and the environmental adjustment instruction sequence in time sequence through the time sequence behavior state transition model, sort each operation instruction included in the application program operation instruction sequence, the input operation instruction sequence, and the environmental adjustment instruction sequence according to an operation timestamp, form a complete user operation sequence conforming to behavior features of the typical user habit model, and the user operation sequence includes the application program operation instruction, the input operation instruction, and the environmental adjustment instruction arranged in time sequence. 6.The notebook adaptation test method according to claim 1, wherein, the injection of the simulated operation habit sequence set into the notebook system, and synchronous collection of system adaptation response data, including: configure a notebook test environment, the test environment including an operating system version, a driver program version, a pre-installed application program version, and a hardware configuration parameter; invoke a user operation sequence injection tool, the user operation sequence injection tool including a software interface calling module, an input device simulation module, and an environmental parameter adjustment module; Load the simulation operation habit sequence set into a user operation sequence injection tool, parse the instruction type, operation object, parameter value and timing information of each user operation sequence, generate an instruction execution plan, and the instruction execution plan includes the execution order and time interval of each user operation instruction; Start the user operation sequence injection tool, and sequentially inject user operation instructions into the notebook system according to the instruction execution plan; Start a response data collector while the user operation instructions are being injected, the response data collector collects resource occupation feature vectors through a kernel-mode monitoring driver, including the occupation rate of each core of a central processing unit, memory usage, the rendering frame rate of a graphics processing unit and the read-write throughput of a storage device, and collects operation response delay data through a user-mode monitoring program, including application startup time, window switching delay and input response time; Align the collected resource occupation feature vectors and operation response delay data with the instruction timestamps of the user operation sequence, generate a time-labeled system adaptation response data set, each user operation instruction corresponds to a set of associated response data, and the system adaptation response data set includes multiple system adaptation response data. 7.The notebook adaptation test method incorporating user habits according to claim 6, wherein, The operation response delay data collected by the user-mode monitoring program includes application startup time, window switching delay and input response time, and includes the following steps: Record the time stamp of the issuance of the application startup instruction and the time stamp of the first rendering completion of the application main window to obtain the application startup time; Record the time stamp of the issuance of the window switching instruction and the time stamp of the target window becoming an active window and completing interface refreshing to obtain the window switching delay; Record the time stamp of the issuance of the input user operation instruction and the first feedback time stamp of the notebook system to the input user operation instruction to obtain the input response time; Perform outlier processing on the collected operation response delay data, remove abnormal fluctuations caused by transient interference by using a sliding window filtering algorithm, and retain effective data reflecting the real response characteristics of the system. 8.The notebook adaptation test method according to claim 1, wherein, The multi-dimensional correlation analysis of the system adaptation response data and the typical user habit model library is performed, the user habit adaptation degree index and the key adaptation bottleneck path are calculated by comparing the system adaptation response data corresponding to different typical user habit models, and the multi-dimensional correlation analysis includes the following steps: Group the multiple system adaptation response data according to the corresponding typical user habit models to obtain an exclusive response data group of each typical user habit model, and the exclusive response data group includes a resource occupation feature vector sequence and an operation response delay data sequence corresponding to the typical user habit model; Perform statistical analysis on each exclusive response data group, calculate the resource occupation mean value, the resource occupation fluctuation coefficient, the response delay mean value and the response delay jitter rate; Call a pre-constructed user habit adaptation degree index calculation model, weight and fuse the resource occupation features and the response delay features to obtain the user habit adaptation degree index; By comparing the user habit adaptation degree indexes of different typical user habit models, identify the typical user habit models with an adaptation degree index lower than a set threshold, and determine the adaptation short board user group of the notebook system. Perform bottleneck path analysis on the exclusive response data set corresponding to the short-board user group, and locate the target operation link and target system component that cause the adaptation degree index to be lower than the set threshold by associating the high-frequency user operation instructions in the user operation sequence with the resource occupation abnormal data; Express the association relationship of the target operation link and the target system component as a key adaptation bottleneck path, and each key adaptation bottleneck path contains a trigger operation node, a resource consumption node and a response delay node, which are used to clearly define the conduction path of the adaptation problem. 9.The notebook adaptation test method incorporating user habits according to claim 8, wherein, The user habit adaptation optimization report is generated based on the user habit adaptation degree index and the key adaptation bottleneck path, including: Generate an adaptation degree distribution heat map based on the user habit adaptation degree index of each typical user habit model, the horizontal axis of the adaptation degree distribution heat map is the user habit type, the vertical axis is the performance index dimension, and the color depth represents the adaptation degree; Generate an operation flow optimization strategy for the trigger operation node in the key adaptation bottleneck path; Generate a hardware resource configuration optimization strategy for the resource consumption node in the key adaptation bottleneck path, which includes memory allocation adjustment strategy, graphics processor parameter optimization strategy and storage device cache optimization strategy; Generate a software performance optimization strategy for the response delay node in the key adaptation bottleneck path; Integrate the adaptation degree distribution heat map, adaptation short-board analysis, operation flow optimization strategy, hardware resource configuration optimization strategy and software performance optimization strategy to generate a user habit adaptation optimization report.

10. A notebook adaptation test system incorporating user habits, characterized by, The processor, machine readable storage medium, the machine readable storage medium and the processor are connected, the machine readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the machine readable storage medium to realize the method in any one of claims 1-9.

Images