Internet behavior analysis method and system applied to smart library

Abstract

The embodiment of the invention provides an internet behavior analysis method and system applied to a smart library, and the method comprises the steps: carrying out the dynamic analysis of an internet behavior sequence of a user in the smart library, generating a behavior unit sequence containing an operation instruction unit and a resource access unit, and generating an association rule set between behavior units; performing behavior pattern evolution tracking based on the behavior unit sequence and the association rule set to obtain stage transition features and pattern variation nodes of the behavior pattern, and constructing a mapping relationship between the stage transition features and the pattern variation nodes and digital resource category attributes by combining the attribute features of the digital resources of the smart library; and according to the mapping relationship, generating an adaptation scheme of the user and the digital resource, the adaptation scheme covering the resource access path optimization suggestion and the resource association recommendation sequence, and finally pushing the adaptation scheme to the user terminal of the smart library to guide the user to perform digital resource access operation, thereby improving the resource acquisition efficiency of the user, and improving the user experience. And the smart library service quality is enhanced.

Description

Technical Field

[0001] This application relates to the field of smart library technology, and more specifically, to a method and system for analyzing internet browsing behavior in smart libraries. Background Technology

[0002] In today's digital age, smart libraries, as important venues for knowledge dissemination and information sharing, offer users a vast space for learning and research through the richness and diversity of their digital resources. With the continuous development of internet technology, users' online behavior within smart libraries is becoming increasingly complex and diverse, encompassing a range of operations from simple resource searches to in-depth learning and research.

[0003] Currently, smart libraries primarily focus on basic network traffic monitoring and simple access record statistics when managing user internet behavior. For example, they analyze the frequency and duration of users' visits to different websites to understand their overall usage of various resources. However, these existing technologies only collect and analyze superficial data on internet behavior, failing to delve deeper into the underlying patterns and potential needs behind user online behavior.

[0004] Specifically, existing technologies cannot dynamically analyze user internet behavior sequences, nor can they generate sequences of behavioral units containing operation command units and resource access units, as well as sets of association rules between these behavioral units. This makes it impossible to track the evolution of user behavior patterns and to identify transitional characteristics and variation nodes in these patterns. Furthermore, the lack of in-depth correlation analysis between user behavior patterns and digital resource attributes makes it impossible to provide users with accurate resource matching solutions, effectively guide users in accessing digital resources, fully realize the value of digital resources in smart libraries, and meet the growing personalized learning and research needs of users. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a method and system for analyzing internet behavior in smart libraries.

[0006] In conjunction with one aspect of the embodiments of this application, a method for analyzing internet browsing behavior in a smart library is provided, the method comprising: The system dynamically analyzes the internet behavior sequences of users in the smart library to generate a sequence of behavior units containing operation instruction units and resource access units, as well as a set of association rules between each behavior unit. Based on the behavioral unit sequence and the association rule set, behavioral pattern evolution is tracked to generate stage transition features and pattern mutation nodes of the behavioral pattern. By combining the digital resource attribute characteristics of the smart library, a mapping relationship is constructed between the stage transition characteristics and the pattern variation nodes and the category attributes of the digital resources; Based on the mapping relationship, an adaptation scheme for users and digital resources is generated, and the adaptation scheme includes resource access path optimization suggestions and resource association recommendation sequences; The adaptation solution is pushed to the user terminal of the smart library to guide users to access digital resources.

[0007] In conjunction with another aspect of the embodiments of this application, an internet behavior analysis system for smart libraries is provided. The internet behavior analysis system for smart libraries includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the internet behavior analysis system for smart libraries implements the steps of the aforementioned internet behavior analysis method for smart libraries.

[0008] In conjunction with another aspect of the embodiments of this application, a readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, can perform the steps of the above-described method for analyzing internet behavior in a smart library.

[0009] Compared to existing technologies, by comprehensively and deeply analyzing the internet behavior sequences of users within a smart library, it is possible to accurately generate behavioral unit sequences containing operation command units and resource access units, as well as a complex and accurate set of association rules between these behavioral units. This process breaks through the limitations of traditional methods that only collect and analyze surface-level data on internet behavior.

[0010] By tracking the evolution of behavioral patterns based on generated behavioral unit sequences and association rule sets, the system can accurately capture the stage transition characteristics and pattern mutation nodes of behavioral patterns, thereby gaining a deeper understanding of the changing patterns and development trends of users' online behavior. Combined with the rich digital resource attributes of smart libraries, a mapping relationship is constructed between stage transition characteristics and pattern mutation nodes and the category attributes of digital resources, achieving a deep correlation between users' online behavior and digital resources.

[0011] The user-digital resource matching scheme generated based on this mapping relationship can provide users with highly targeted resource access path optimization suggestions and resource association recommendation sequences, effectively guiding users to access digital resources. This not only improves the efficiency of users obtaining the resources they need and enhances the user experience, but also fully leverages the value of digital resources in smart libraries, promotes the dissemination and sharing of knowledge, and drives smart libraries towards a more intelligent and personalized direction.

[0012] To make the above-mentioned objects, features and advantages of the embodiments of this application more apparent and understandable, a detailed description will be given below in conjunction with the embodiments and the accompanying drawings. Attached Figure Description

[0013] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained in conjunction with these drawings without creative effort.

[0014] Figure 1 This invention provides a schematic diagram of the components of an internet behavior analysis system for smart libraries, as illustrated in an embodiment of this application. Figure 2 This paper illustrates a flowchart of an internet behavior analysis method for smart libraries provided in an embodiment of this application. Detailed Implementation

[0015] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. All other embodiments obtained by those skilled in the art in conjunction with the embodiments of the present application without creative effort are within the scope of protection of the present application.

[0016] The terms “first,” “second,” “third,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] Figure 1A schematic diagram of exemplary components for an internet behavior analysis system 100 applied to a smart library is shown. The internet behavior analysis system 100 for a smart library may include one or more processors 104, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. The internet behavior analysis system 100 for a smart library may also include any storage medium 106 for storing any kind of information such as code, settings, data, etc. Non-limitingly, for example, storage medium 106 may include any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any storage medium can use any technology to store information. Further, any storage medium may provide volatile or non-volatile retention of information. Further, any storage medium may represent a fixed or removable component of the internet behavior analysis system 100 for a smart library. In one case, when processor 104 executes associated instructions stored in any storage medium or combination of storage media, the internet behavior analysis system 100 for a smart library may perform any operation of the associated instructions. The Internet behavior analysis system 100 applied to smart libraries also includes one or more drive units 108 for interacting with any storage medium, such as hard disk drive units, optical disk drive units, etc.

[0018] The internet behavior analysis system 100 for smart libraries also includes input / output (I / O) 110 for receiving various inputs (via input unit 112) and providing various outputs (via output unit 114). A specific output mechanism may include a presentation device 116 and an associated graphical user interface (GUI) 118. The internet behavior analysis system 100 for smart libraries may also include one or more network interfaces 120 for exchanging data with other devices via one or more communication units 122. One or more communication buses 124 couple the components described above together.

[0019] The communication unit 122 can be implemented in any way, such as via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, or any combination thereof. The communication unit 122 may include any combination of hardwired links, wireless links, routers, gateway functions, and names applied to the Internet access behavior analysis system 100 for smart libraries, governed by any protocol or combination of protocols.

[0020] Figure 2 This application provides a flowchart illustrating a method for analyzing internet behavior in a smart library, which can be implemented by... Figure 1The Internet behavior analysis system 100 applied to the smart library shown is executed, and the detailed steps of the Internet behavior analysis method applied to the smart library are described below.

[0021] Step S110: Dynamically analyze the internet access behavior sequence of users in the smart library to generate a sequence of behavior units containing operation instruction units and resource access units, as well as a set of association rules between each behavior unit.

[0022] In this embodiment, the application scenario throughout the text is the user's behavior of accessing digital resources through a terminal device within a smart library. When a user begins to operate on the smart library's terminal, the relevant behavioral data is recorded, forming an internet access behavior sequence.

[0023] Step S111: Collect the original records of users' internet access behavior in the smart library and arrange them in order of timestamps to form an internet access behavior sequence.

[0024] When collecting raw records of user internet behavior, every action a user takes on their device can be collected, including clicks, input, and browsing, along with the timestamp of each action. These raw records cover user queries, downloads, browsing, and access to various digital resources. During the collection process, sensitive data involving user privacy, such as login account information and personal preference settings, can be processed using data anonymization techniques, replacing sensitive information with meaningless characters or codes to ensure user privacy is not leaked. These raw records are then arranged in chronological order by timestamp, forming an ordered sequence of internet behavior. For example, a query performed by the user at a certain time is recorded first, followed by browsing, then downloads, and so on, forming a complete sequence of internet behavior.

[0025] Step S112: Divide the Internet access behavior sequence into units, divide consecutive similar operation instructions into operation instruction units, and divide consecutive access to the same type of digital resources into resource access units, forming a preliminary behavior unit sequence.

[0026] After constructing the internet access behavior sequence, the unit division begins. This process requires processing the operation commands and resource access records in the sequence separately.

[0027] For example, step S1121: perform type identification on the operation instructions in the Internet access behavior sequence, and determine the operation type to which each operation instruction belongs. The operation type includes query operation, download operation and browsing operation.

[0028] The operation type is identified by analyzing the characteristics of each operation command in the sequence of online behavior. For example, when a user enters keywords in the search box and clicks the search button, the operation command is identified as a query operation; when a user clicks a resource link and selects the download option, the operation command is identified as a download operation; when a user scrolls through an opened resource page, the operation command is identified as a browsing operation.

[0029] Step S1122: Traverse the sequence of Internet access behaviors in chronological order, group the operation instructions of the same operation type that appear consecutively into a group to form an operation instruction unit. Each operation instruction unit contains the operation type and the time interval in which the operation occurred.

[0030] After determining the type of each operation instruction, the sequence of internet browsing behavior is traversed chronologically. When multiple operation instructions of the same type appear consecutively, they are grouped together to form an operation instruction unit. Each operation instruction unit explicitly records its operation type, as well as the start and end times of the group of operation instructions, thereby determining the time interval in which the operation occurred. For example, if a user performs multiple browsing operations consecutively within a certain period, these operations will be grouped into a single browsing operation instruction unit, which includes the browsing operation type and the start and end times of these browsing operations.

[0031] Step S1123: Analyze the digital resource access records in the internet access behavior sequence to determine the digital resource category corresponding to each access record.

[0032] For digital resource access records in the sequence of internet browsing behavior, the category of digital resource corresponding to each access record is determined by analyzing the resource's identification information, content tags, etc. For example, if the access record points to a paper on computer science, then the digital resource category corresponding to the access record is computer science; if it accesses a literary work, the corresponding category is literature.

[0033] Step S1124: Traverse the sequence of internet access behaviors in chronological order, group the access records of the same digital resource category that are accessed consecutively into a group to form a resource access unit. Each resource access unit contains the digital resource category and the access duration.

[0034] Similarly, the online behavior sequence is traversed chronologically. When records of consecutive accesses to the same digital resource category appear, they are grouped together to form a resource access unit. Each resource access unit records the corresponding digital resource category and the duration from the start of accessing that category of resources to the end of the access, i.e., the access duration. For example, if a user consecutively accesses multiple history books, these access records will be grouped into a history resource access unit, which includes the history category of digital resources and the total access duration.

[0035] Step S1125: Alternately arrange operation instruction units and resource access units to form a preliminary sequence of action units, wherein the operation instruction units and resource access units are arranged in chronological order.

[0036] After obtaining the operation instruction units and resource access units respectively, they are arranged alternately according to their chronological order in the internet access behavior sequence. For example, the query operation instruction unit appears first, followed by the resource access unit corresponding to that query operation, then the next operation instruction unit, and then the corresponding resource access unit, and so on, thus forming a preliminary sequence of behavior units.

[0037] Step S113: Analyze the triggering relationship between adjacent operation instruction units and resource access units, and determine the triggering condition for the previous action unit to trigger the next action unit.

[0038] Analyze adjacent behavioral units in the initial sequence to explore their triggering relationships. For example, if a query operation is immediately followed by a resource access unit, analyze how the query operation triggers access to a specific resource to determine its triggering conditions. This could be that the query keywords match the resource's tags, leading the user to access the resource. Through the analysis of a large number of adjacent behavioral units, summarize various possible triggering conditions.

[0039] Step S114: Calculate the co-occurrence frequency among each behavioral unit based on the triggering conditions, and generate an association rule set with co-occurrence frequency as the weight.

[0040] After identifying the triggering conditions, the frequency of co-occurrence between different behavioral units under these conditions is calculated. For example, under the triggering condition of "matching query keywords and resource tags," the number of times a query operation instruction unit and a resource access unit co-occur is counted, and this number is taken as their co-occurrence frequency. Then, these co-occurrence frequencies are used as weights to generate a set of association rules. Each rule in the association rule set represents the association relationship between two behavioral units, as well as the weight of that association relationship.

[0041] Step S115: Redundant units are removed from the preliminary action unit sequence, and action units with triggering relationships are retained to form the final action unit sequence containing operation instruction units and resource access units.

[0042] In the initial sequence of behavioral units, there may be some redundant units that have no triggering relationship with other behavioral units. These units are identified and eliminated by comparing them with the association rule set. Only those behavioral units that have a triggering relationship in the association rule set are retained, thus forming the final sequence of behavioral units. For example, if an operation instruction unit has no triggering relationship recorded in the association rules with any other resource access unit, then this operation instruction unit will be determined as a redundant unit and eliminated.

[0043] Step S120: Based on the behavioral unit sequence and the association rule set, perform behavioral pattern evolution tracking to generate stage transition features and pattern mutation nodes of the behavioral pattern.

[0044] With the final sequence of behavioral units and the set of association rules, we can begin to track the evolution of user behavior patterns to understand how user behavior changes.

[0045] Step S121: Extract the feature parameters of each behavior unit from the sequence of behavior units, the feature parameters including operation duration and resource access depth.

[0046] For each operation instruction unit in the action unit sequence, its operation duration is extracted, that is, the length of time from the start to the end of the operation contained in that unit. For each resource access unit, in addition to extracting the access duration, the resource access depth is also extracted. The resource access depth can be comprehensively measured by multiple dimensions such as the number of pages viewed by the user in the resource unit, the distribution of time spent on the page, and the degree of interaction with the resource content, forming a multi-dimensional feature parameter. For example, the access depth of a resource access unit can be represented by a set of values ​​consisting of the number of pages viewed, the average time spent on each page, and whether the content was marked.

[0047] Step S122: Determine the association strength between behavioral units based on the association rule set, and divide the sequence of behavioral units into multiple behavioral pattern stages according to the association strength.

[0048] The association strength between behavioral units is determined by the weights (i.e., co-occurrence frequencies) of the association relationships within the association rule set. Higher association strength indicates a closer connection between the two behavioral units. Based on the magnitude of the association strength, the sequence of behavioral units is divided into multiple behavioral pattern stages. When the association strength between behavioral units is above a certain level, they are grouped into the same stage; when the association strength is below that level, they are grouped into different stages. For example, a series of behavioral units with high association strength constitutes a behavioral pattern stage; when the association strength significantly decreases, the next behavioral pattern stage is entered.

[0049] Step S123: Calculate the rate of change of characteristic parameters of behavioral units in adjacent behavioral pattern stages, and mark the positions where the rate of change exceeds a preset threshold as stage transition points.

[0050] For two adjacent behavioral pattern stages, the changes in the characteristic parameters of each behavioral unit are calculated between the two stages to obtain the characteristic parameter change rate. For example, the difference between the operation duration of a certain operation instruction unit in the previous stage and the operation duration of the corresponding type of operation instruction unit in the next stage is calculated, and then divided by the operation duration to obtain the change rate. When this change rate exceeds a preset threshold, the position is marked as a stage transition point, indicating that the user's behavioral pattern has undergone a significant change at this position.

[0051] Step S124: Extract the behavioral unit sequence fragments before and after the stage transition point, analyze the combination changes of operation instruction units and resource access units in the fragments, and generate stage transition features.

[0052] After marking the transition point, extract behavioral unit sequence fragments within a certain range before and after the transition point. Analyze these fragments to observe the changes in the types of operation instruction units, the categories of resource access units, and their combinations before and after the transition point. For example, in the fragments before the transition point, the operation instruction units are mainly query operations, and the resource access units are mainly history-related; in the fragments after the transition point, the operation instruction units change to download operations, and the resource access units change to literature-related. Organize these features of combination changes to generate stage transition features. This feature is a multi-dimensional set that includes information on changes in operation type, resource category, and their combination relationships.

[0053] Step S125: Monitor the behavioral unit combinations in the behavioral unit sequence that do not match the association rule set, and determine the position of the first occurrence of a mismatched combination as a pattern mutation node.

[0054] The system continuously monitors combinations of behavioral units within a sequence and compares them with rules in the association rule set. When a combination of behavioral units is found to have no corresponding rule in the association rule set (i.e., a mismatch), the location where the mismatch first appears is identified as a pattern mutation node. For example, if the association rule set has not recorded a combination of a query operation instruction unit and an art resource access unit, the location where this combination first appears is marked as a pattern mutation node.

[0055] Step S126: Continuously track the sequence of behavioral units after the pattern mutation node, and record the degree of deviation between the subsequent combination of behavioral units and the original association rule set as a supplementary feature of the pattern mutation node.

[0056] After identifying the pattern mutation node, the sequence of behavioral units following that node is continuously tracked. The deviation between subsequent combinations of behavioral units and rules in the original association rule set is analyzed, and the degree of deviation is calculated. The degree of deviation can be measured by the difference between subsequent combinations and the most similar rule in the rule set, forming multi-dimensional supplementary features to enrich the information of the pattern mutation node. For example, the difference between the combination of behavioral units following the pattern mutation node and a certain rule in the association rule set is represented by multiple numerical dimensions, which together constitute supplementary features.

[0057] Step S130: Combining the digital resource attribute characteristics of the smart library, construct a mapping relationship between the stage transition characteristics and the pattern variation nodes and the category attributes of the digital resources.

[0058] To link changes in user behavior patterns with the attributes of digital resources, it is necessary to construct a mapping relationship based on the attribute characteristics of digital resources.

[0059] Step S131: Obtain the digital resource attribute characteristics of the smart library, including the subject category of the resource, the resource update cycle, and the resource access permission level.

[0060] The smart library's resource management system acquires attribute information for all digital resources. For each digital resource, its subject category is extracted, such as mathematics, physics, and chemistry; its update cycle is determined, such as monthly or quarterly updates; and its access permission level is specified, such as public access, registered user access, or access by users with specific permissions. This attribute information collectively constitutes the attribute characteristics of the digital resources, and the attribute characteristics of each resource are a multi-dimensional set.

[0061] Step S132: Perform feature quantization processing on the stage transition features to convert the stage transition features into a computable feature vector, wherein the dimension of the feature vector is consistent with the dimension of the digital resource attribute features.

[0062] Since stage transition features describe changes in behavioral patterns, they need to be converted into computable feature vectors. Based on the dimensions of the digital resource attribute features, the stage transition features are quantified. For example, if digital resource attribute features have three dimensions (subject category, update cycle, and access permission level), then the change features corresponding to these three dimensions in the stage transition features are quantified to form a 3-dimensional feature vector. The value of each dimension is set according to the specific characteristics of the feature to ensure that the feature vector can be calculated and compared with the digital resource attribute features.

[0063] Step S133: Analyze the digital resources involved in the combination of behavioral units corresponding to the pattern mutation node, extract the category attributes of these digital resources, and form a set of mutation-related resource categories.

[0064] For each combination of behavioral units corresponding to a pattern mutation node, the digital resources involved in that combination are identified. These digital resources are then analyzed to extract their category attributes, such as historical, literary, and scientific categories. These category attributes are then aggregated to form a set of mutation-related resource categories.

[0065] Step S134: Calculate the similarity between the feature vector and each digital resource attribute feature, and generate a similarity matrix. The elements in the similarity matrix represent the degree of matching between the stage transition feature and the corresponding digital resource attribute feature.

[0066] To determine the degree of matching between the phase transition features and the digital resource attribute features, it is necessary to calculate their similarity.

[0067] Step S1341: Standardize the feature vector and each digital resource attribute feature to eliminate the dimensional differences between features of different dimensions.

[0068] Since the dimensions of feature vectors and digital resource attribute features may have different units of measurement, standardization is required. For example, for the dimension in the feature vector representing changes in operation type and the dimension in the digital resource attribute feature representing the update cycle, a standardization method is used to convert their values ​​to the same numerical range, such as [0, 1], to eliminate the impact of unit of measurement differences on similarity calculation.

[0069] Step S1342: Calculate the cosine similarity value between the standardized feature vector and each digital resource attribute feature using the cosine similarity algorithm.

[0070] After standardization, the cosine similarity algorithm is used to calculate the similarity between the feature vectors and each digital resource attribute feature. Cosine similarity is calculated by taking the cosine of the angle between two vectors; a value closer to 1 indicates higher similarity, and a value closer to 0 indicates lower similarity.

[0071] Step S1343: Classify and summarize each cosine similarity value according to the category attribute of the digital resource to generate a preliminary similarity list.

[0072] The calculated cosine similarity values ​​are categorized according to the category attributes of the digital resources. For example, all similarity values ​​belonging to the historical category are grouped into one category, and those belonging to the literary category are grouped into another category, thus forming a preliminary similarity list.

[0073] Step S1344: Sort the preliminary similarity list and arrange the digital resource category attributes in descending order of similarity value.

[0074] The similarity values ​​corresponding to each category attribute in the initial similarity list are sorted, and the digital resource category attributes are arranged in descending order of similarity value in order to construct the similarity matrix later.

[0075] Step S1345: Construct a similarity matrix based on the sorting results. The rows of the matrix correspond to the feature vectors of the stage transition features, the columns of the matrix correspond to the category attributes of digital resources, and the elements in the matrix are the corresponding cosine similarity values.

[0076] Based on the sorted digital resource category attributes, a similarity matrix is ​​constructed. The number of rows in the matrix is ​​the same as the number of feature vectors (here, 1 row corresponds to the feature vector of a transitional feature at one stage), and the number of columns is the same as the number of digital resource category attributes. Each element in the matrix is ​​the cosine similarity value between the corresponding feature vector and the digital resource category attribute.

[0077] Step S1346: Normalize the similarity matrix so that the values ​​of all elements in the matrix are within a preset numerical range.

[0078] To ensure the comparability of elements in the similarity matrix, it is normalized. Each element value in the matrix is ​​transformed to a preset numerical range, such as [0, 1]. This is achieved by dividing each element value by the maximum value of all elements in the matrix, or by using other normalization methods to ensure that the element value is within this range.

[0079] Step S135: Based on the similarity matrix and the set of variant-related resource categories, establish a first mapping relationship between stage transition features and digital resource category attributes, and a second mapping relationship between pattern variant nodes and digital resource category attributes.

[0080] Based on the similarity values ​​between the feature vectors of the stage transition features in the similarity matrix and the digital resource category attributes, a first mapping relationship is established, indicating which digital resource category attributes are more closely matched with the stage transition features. Simultaneously, based on the set of variant-associated resource categories, a second mapping relationship is established, indicating which digital resource category attributes are associated with the pattern variant nodes.

[0081] Step S136: Merge the first mapping relationship and the second mapping relationship to form a comprehensive mapping relationship that includes stage transition features, pattern variation nodes and digital resource category attributes.

[0082] By integrating the first and second mapping relationships and comprehensively considering the correlation between stage transition characteristics, pattern variation nodes, and digital resource category attributes, a comprehensive mapping relationship is formed. This mapping relationship can more accurately reflect the connection between changes in user behavior patterns and digital resource category attributes.

[0083] Step S140: Generate an adaptation scheme for users and digital resources based on the mapping relationship. The adaptation scheme includes resource access path optimization suggestions and resource association recommendation sequences.

[0084] Based on the constructed comprehensive mapping relationship, a suitable digital resource adaptation solution is generated for the user.

[0085] Step S141: Extract the digital resource category attribute with the highest matching degree with the stage transition feature from the mapping relationship, and use it as the transition adaptation resource category.

[0086] In the comprehensive mapping relationship, the digital resource category attributes with the highest matching degree with the stage transition characteristics are identified and determined as the transition-adaptive resource categories. The digital resources corresponding to these category attributes best match the user's behavior patterns after the stage transition.

[0087] Step S142: Select the corresponding digital resources according to the transition adaptation resource category, analyze the storage location and access path of these digital resources in the smart library resource database, and generate the initial resource access path.

[0088] Based on the transitional adaptation resource categories, digital resources belonging to these categories are selected from the smart library's resource repository. The storage locations of these resources are analyzed, such as server storage paths and database locations, as well as the steps required to access them, such as which section of the library's official website to navigate to, which category to click, and then selecting a specific sub-item, thus generating an initial resource access path. For example, if the transitional adaptation resource category is Computer Science, the selected digital resources include computer network papers and programming tutorials, which are stored in a specific server partition of the library's resource repository. Accessing them requires first entering the "Academic Resources" section of the official website, then clicking the "Computer Science" category, and then selecting the corresponding sub-category, following the steps described above to generate the initial resource access path.

[0089] Step S143: Based on the changes in the operation instruction units in the stage transition characteristics, adjust the initial resource access path to reduce unnecessary operation steps and form a resource access path optimization suggestion.

[0090] After obtaining the initial resource access path, it is necessary to optimize it by taking into account the changes in the operation instruction units in the stage transition characteristics.

[0091] Step S1431: Analyze the changes in the operation instruction units in the stage transition characteristics, and identify duplicate operations and invalid operations in the operation instruction units.

[0092] Analyze the types, frequency, and order of operation instruction units in the phase transition characteristics to identify repetitive and invalid operations. Repetitive operations refer to the same operation instruction appearing multiple times in a short period of time, such as clicking the same search button repeatedly. Invalid operations refer to operation instructions that fail to successfully obtain resources or do not actually help access resources, such as clicking an incorrect link and then returning to the previous page.

[0093] Step S1432: Mark the path nodes corresponding to duplicate and invalid operations in the initial resource access path.

[0094] Based on the identified duplicate and invalid operations, the corresponding path nodes in the initial resource access path are located and marked. For example, if there is an operation of clicking the "Computer Science" category twice in the initial path, this is a duplicate operation, and the corresponding path node will be marked; an operation of clicking the "Literature" link and then immediately returning is an invalid operation, and its corresponding path node will also be marked.

[0095] Step S1433: Delete the marked path nodes in the initial resource access path, and directly connect adjacent valid path nodes to form a simplified resource access path.

[0096] Remove the path nodes corresponding to the marked duplicate and invalid operations from the initial path, and then directly connect the valid path nodes before and after these nodes. For example, after deleting the second node of the "Computer Science" category, directly connect the first node with the subsequent subdirectory selection node to form a simpler path.

[0097] Step S1434: Calculate the difference in operation steps between the simplified resource access path and the initial resource access path, ensuring that the difference reaches the preset optimization threshold.

[0098] The number of operation steps in the simplified resource access path is compared with the number of operation steps in the initial path, and the difference is calculated. By comparing this difference with a preset optimization threshold, it is determined whether the path simplification has achieved the expected effect. If the difference reaches or exceeds the threshold, the optimization is effective; if not, the path needs to be re-examined and further simplified.

[0099] Step S1435: Perform integrity verification on the simplified resource access path to ensure that the target digital resource can be accessed normally through the optimized path.

[0100] Simulate the operation using the simplified resource access path to check if the target digital resource can be accessed successfully. For example, click on the corresponding section, category, and subdirectory in sequence according to the optimized path to see if the required computer network papers, programming tutorials, etc., can be accurately found and opened. If the verification process reveals that the resource cannot be accessed or the wrong resource is accessed, the path needs to be corrected.

[0101] Step S1436: Add operation guidance information to the verified simplified resource access path to form the final resource access path optimization suggestion.

[0102] On the simplified resource access path that has been verified for integrity, add detailed operation guidance information for each step, such as which button to click and what information to enter. For example, next to the step of "entering the 'Academic Resources' section of the official website", add guidance such as "enter the library's official website URL in the browser address bar, and click the 'Academic Resources' option in the top navigation bar after entering the homepage", thus forming the final optimized resource access path suggestion.

[0103] Step S144: For the pattern mutation node, obtain the digital resource category attribute associated with the pattern mutation node from the mapping relationship, as the mutation adaptation resource category.

[0104] In the comprehensive mapping relationship, the digital resource category attributes associated with the pattern mutation node are identified, and these category attributes are determined as mutation-adapted resource categories. The digital resources corresponding to these category attributes are associated with the new behavioral patterns exhibited by the user at the pattern mutation node. For example, if the combination of behavioral units corresponding to the pattern mutation node involves art-related resources, then the art-related category attributes obtained from the mapping relationship are the mutation-adapted resource categories.

[0105] Step S145: Based on the mutation-adapted resource category, search for other digital resources with similar attribute characteristics to form a resource association candidate set.

[0106] Based on the attribute characteristics of the mutation-adapted resource category, other digital resources with similar attribute characteristics are searched in the resource repository of the smart library. Similar attribute characteristics can include similar subject categories, similar update cycles, and the same access permission level. For example, if the mutation-adapted resource category is oil painting, then digital resources with similar attribute characteristics to oil painting, such as watercolor painting and sculpture, are searched, and these are aggregated to form a candidate set of resource associations.

[0107] Step S146: Sort the resource association candidate set according to the access frequency of related resources in the user's historical Internet access behavior sequence, and generate a resource association recommendation sequence.

[0108] Step S1461: Extract access records for each digital resource in the resource association candidate set from the user's historical internet access behavior sequence, and count the number of accesses and the most recent access time for each digital resource.

[0109] Iterate through the user's historical internet browsing behavior sequence to find access records for each digital resource in the resource association candidate set, and count the number of times each digital resource was accessed and the time of the last access. For example, a watercolor painting tutorial in the resource association candidate set may have been accessed several times in the history, with the most recent access occurring at a certain point in time.

[0110] Step S1462: Calculate the access frequency weight of each digital resource based on the number of accesses and the time of the most recent access. The more accesses and the more recent the access time, the greater the access frequency weight.

[0111] Access frequency weights are calculated by combining the number of visits and the time of the most recent visit. A specific calculation method can be used to convert the number of visits and the time of the most recent visit into weight values, giving digital resources with more frequent visits and more recent visits a higher weight. For example, the number of visits can be converted into a base weight according to a certain ratio, and then the base weight can be adjusted according to the length of the interval between the most recent visit time and the current time. The shorter the interval, the greater the adjusted weight, ultimately yielding the access frequency weight for each digital resource.

[0112] Step S1463: Sort the digital resources in the resource association candidate set in descending order of access frequency weight to form a preliminary recommendation sequence.

[0113] Based on the calculated access frequency weights, the digital resources in the resource association candidate set are sorted, with the digital resources with the highest weights placed at the top, the next highest weights placed at the bottom, and so on, forming a preliminary recommendation sequence.

[0114] Step S1464: Check if there are duplicate digital resources in the preliminary recommendation sequence. If so, retain the digital resources with higher access frequency weights.

[0115] The initial recommendation sequence is checked for duplicate digital resources. If duplicates are found, the access frequency weights of these duplicate digital resources are compared, and only the one with the higher weight is retained, ensuring that each digital resource in the recommendation sequence is unique.

[0116] Step S1465: Adjust the length of the deduplicated preliminary recommended sequence to meet the preset sequence length requirements.

[0117] Based on actual needs, a preset length for the resource association recommendation sequence is established. If the length of the initial recommended sequence after deduplication exceeds the preset length, the digital resources of the preset length are truncated as the adjusted sequence; if the length is insufficient, some digital resources with low correlation to the mutation adaptation resource category but still relevant are added from the resource library to meet the preset length requirement.

[0118] Step S1466: Add brief descriptions of each digital resource to the adjusted recommendation sequence to form the final resource association recommendation sequence.

[0119] Add brief descriptive information to each digital resource in the adjusted recommendation sequence, such as the resource's subject matter, author, and publication date. For example, add a description next to a watercolor painting tutorial: "This tutorial introduces the basic techniques and creative steps of watercolor painting, suitable for beginners," thus forming the final resource association recommendation sequence.

[0120] Step S147: Integrate the resource access path optimization suggestions and the resource association recommendation sequence to form a user-digital resource matching scheme.

[0121] The resource access path optimization suggestions and resource association recommendation sequences are integrated and formatted according to a certain format to form a complete user-digital resource adaptation solution. For example, the adaptation solution first lists the resource access path optimization suggestions, then attaches the resource association recommendation sequences, and adds transitional explanations between the two to make the entire solution clear, concise, and easy to understand.

[0122] Step S150: Push the adaptation scheme to the user terminal of the smart library to guide the user to access digital resources.

[0123] After the adaptation plan is generated, it needs to be pushed to the user's terminal device to guide the user to perform relevant operations.

[0124] Step S151: Obtain the terminal identification information of the user logging into the Internet in the smart library. The terminal identification information is used to determine the user terminal currently being used by the user.

[0125] The smart library's network management system obtains the identification information of the terminal used by users when logging in to the internet, such as the terminal's IP address and MAC address. This identification information can uniquely identify the terminal device currently being used by the user, ensuring that the adaptation plan can be accurately pushed.

[0126] Step S152: Perform format conversion processing on the adaptation scheme to convert the adaptation scheme into a format suitable for display on the user terminal. The format includes text description and graphic guidance.

[0127] Based on the user's terminal type (such as computer, tablet, mobile phone, etc.) and operating system, the adaptation solution undergoes format conversion. The text description is adjusted to a font size, line spacing, etc., suitable for the terminal screen display; the illustrative instructions are converted to image formats supported by the terminal, such as JPG and PNG, to ensure the adaptation solution can be clearly displayed on the terminal.

[0128] Step S153: Establish a communication connection with the user terminal and send the converted adaptation scheme to the user terminal through a preset communication protocol.

[0129] Utilizing network communication technology and based on preset communication protocols (such as HTTP, HTTPS, etc.), a communication connection is established with the user terminal. After the connection is established, the converted and adapted data is sent to the user terminal, ensuring the stability and security of data transmission.

[0130] Step S154: Monitor the user terminal's reception status of the adaptation scheme. If the reception is unsuccessful, resend the adaptation scheme until the user terminal is confirmed to have received it successfully.

[0131] During the transmission of the adaptation plan, the receiving status of the user terminal is monitored in real time. Successful reception is determined by the confirmation message returned by the receiving terminal. If no confirmation message is received for an extended period or a reception failure message is received, the adaptation plan is retransmitted, and the monitoring process is repeated until the user terminal is confirmed to have successfully received the data.

[0132] Step S155: When the adaptation scheme is displayed on the user terminal, set interactive prompts to prompt the user to perform digital resource access operations according to the adaptation scheme.

[0133] When the adaptation solution is displayed on the user's terminal, interactive prompts are also set, such as pop-up prompts or floating windows. These prompts may include messages like "You have new resource access suggestions, would you like to view them?", guiding users to pay attention and perform digital resource access operations according to the adaptation solution.

[0134] Step S156: Continuously receive operation response information from the user terminal, determine whether the user has followed the adaptation plan based on the response information, and if not, push the key content of the adaptation plan again.

[0135] After the adaptation plan is pushed out, we continuously receive operation response information from user terminals, such as whether the user clicked the links in the adaptation plan or followed the path instructions. Based on this response information, we determine whether the user has followed the adaptation plan. If we find that the user has not followed the adaptation plan, such as deviating from the optimized access path, we push the key content of the adaptation plan again, such as guidance information for crucial steps, to remind the user to follow the suggestions.

[0136] Through the above series of steps, the online behavior of users within the smart library is analyzed, and corresponding adaptation solutions are generated and pushed to user terminals. This effectively guides users to access the digital resources they need more conveniently, improving their user experience in the smart library. Throughout the process, each step is closely linked, forming a complete closed loop from the collection of behavioral data to the final solution delivery, ensuring the feasibility and effectiveness of the technical solution.

[0137] In practical applications, each step can be optimized and adjusted according to specific circumstances. For example, when dividing behavior units, more refined operation type classification criteria can be introduced to improve the accuracy of unit division; when calculating similarity, multiple similarity algorithms can be tried, and the most suitable one can be selected to improve the accuracy of mapping relationships. Simultaneously, with the continuous accumulation of user behavior data and the constant updating of digital resources, it is necessary to regularly update and maintain the association rule set and mapping relationships to ensure the timeliness and applicability of the entire analysis method.

[0138] Furthermore, to further enhance the personalization of the adaptation solution, the recommended resources and access paths can be more precisely adjusted by combining users' historical behavior data and personal preference information (while protecting privacy). For example, for users who frequently download resources, downloadable resources can be prioritized in the resource association recommendation sequence; for users who are accustomed to using shortcuts, more shortcut guidance can be added to the access path optimization suggestions.

[0139] In terms of technical implementation, each step can be accomplished through corresponding software modules. For example, the behavior data collection module is responsible for collecting raw records of users' online behavior; the behavior unit segmentation module is responsible for segmenting the online behavior sequence into units; the pattern evolution tracking module is responsible for generating stage transition features and pattern mutation nodes; the mapping relationship construction module is responsible for establishing the mapping relationship between behavior pattern changes and digital resource category attributes; the adaptation scheme generation module is responsible for generating resource access path optimization suggestions and resource association recommendation sequences; and the scheme push module is responsible for pushing the adaptation scheme to the user terminal, etc. These modules communicate and exchange data through data interfaces, collaboratively completing the entire process of online behavior analysis and adaptation scheme push.

[0140] In terms of data storage, a dedicated database needs to be established to store user online behavior data, digital resource attribute data, association rule sets, mapping relationships, and other information. The database design should consider data security, integrity, and scalability, employing appropriate data storage structures and indexing methods to improve data query and access efficiency, thereby meeting the data processing speed requirements of the entire analysis method.

[0141] During system operation, a comprehensive monitoring and logging mechanism is also required. Real-time monitoring of the operational status of each module is necessary to promptly identify and resolve any faults or problems that arise. Detailed logs of all system operations and data processing procedures must be maintained for subsequent auditing, analysis, and troubleshooting. Log entries should include information such as operation time, operation content, and processing results, and log data should be backed up regularly to prevent data loss.

[0142] In summary, the internet behavior analysis method for smart libraries described in this embodiment provides users with personalized resource access solutions by comprehensively analyzing and deeply mining users' internet behavior and combining it with the attribute characteristics of digital resources. This can effectively improve users' efficiency and experience in obtaining digital resources, and also help smart libraries better manage and utilize digital resources, thereby improving service quality.

[0143] This application provides a computer-readable storage medium storing computer-executable instructions. When a processor executes these computer-executable instructions, it implements the methods provided in any of the embodiments described above.

[0144] The aforementioned computer-readable storage media can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0145] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in application-specific integrated circuits (ASICs). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0146] This application provides a computer program product, including a computer program that, when executed by a processor, implements the method provided in any embodiment of this application.

[0147] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope of this application is indicated by the claims.

[0148] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for analyzing internet behavior in a smart library, characterized in that, The method includes: The system dynamically analyzes the internet behavior sequences of users in the smart library to generate a sequence of behavior units containing operation instruction units and resource access units, as well as a set of association rules between each behavior unit. Based on the behavioral unit sequence and the association rule set, behavioral pattern evolution is tracked to generate stage transition features and pattern mutation nodes of the behavioral pattern. By combining the digital resource attribute characteristics of the smart library, a mapping relationship is constructed between the stage transition characteristics and the pattern variation nodes and the category attributes of the digital resources; Based on the mapping relationship, an adaptation scheme for users and digital resources is generated, and the adaptation scheme includes resource access path optimization suggestions and resource association recommendation sequences; The adaptation solution is pushed to the user terminal of the smart library to guide users to access digital resources.

2. The internet behavior analysis method for smart libraries according to claim 1, characterized in that, The process of dynamically parsing the internet access behavior sequences of users within the smart library generates a sequence of behavior units containing operation instruction units and resource access units, as well as a set of association rules between these behavior units, including: Collect raw records of users' online behavior in the smart library and arrange them in order of timestamps to form an online behavior sequence; The sequence of internet access behaviors is divided into units. Continuous similar operation instructions are divided into operation instruction units, and continuously accessed digital resources of the same type are divided into resource access units, thus forming a preliminary sequence of behavior units. Analyze the triggering relationship between adjacent operation instruction units and resource access units to determine the triggering condition for the preceding action unit to trigger the following action unit; Based on the triggering conditions, the co-occurrence frequency among each behavioral unit is calculated, and an association rule set with co-occurrence frequency as the weight is generated; The preliminary sequence of behavioral units is subjected to redundant unit removal processing, and behavioral units with triggering relationships are retained to form the final sequence of behavioral units containing operation instruction units and resource access units.

3. The internet behavior analysis method for smart libraries according to claim 1, characterized in that, The step of tracking behavioral pattern evolution based on the behavioral unit sequence and the association rule set, generating stage transition features and pattern mutation nodes for behavioral patterns, includes: Extract feature parameters for each behavior unit from the sequence of behavior units, the feature parameters including operation duration and resource access depth; The association strength between behavioral units is determined based on the association rule set, and the sequence of behavioral units is divided into multiple behavioral pattern stages according to the association strength. Calculate the rate of change of characteristic parameters of behavioral units in adjacent behavioral pattern stages, and mark the positions where the rate of change exceeds a preset threshold as stage transition points; Extract behavioral unit sequence fragments before and after the stage transition point, analyze the combination changes of operation instruction units and resource access units in the fragments, and generate stage transition features; Monitor the combinations of behavioral units in the sequence of behavioral units that do not match the association rule set, and determine the position of the first occurrence of a mismatched combination as a pattern mutation node; The sequence of behavioral units following the pattern mutation node is continuously tracked, and the degree of deviation between the subsequent combination of behavioral units and the original set of association rules is recorded as a supplementary feature of the pattern mutation node.

4. The internet behavior analysis method for smart libraries according to claim 1, characterized in that, The process of combining the digital resource attribute characteristics of the smart library to construct a mapping relationship between the stage transition characteristics and the pattern variation nodes and the category attributes of the digital resources includes: The digital resource attribute characteristics of the smart library are obtained, including the subject category of the resource, the resource update cycle, and the resource access permission level. The stage transition features are subjected to feature quantization processing to convert them into computable feature vectors, and the dimension of the feature vectors is consistent with the dimension of the digital resource attribute features. Analyze the digital resources involved in the behavioral unit combinations corresponding to the pattern mutation nodes, extract the category attributes of these digital resources, and form a mutation-related resource category set; Calculate the similarity between the feature vector and each digital resource attribute feature to generate a similarity matrix. The elements in the similarity matrix represent the degree of matching between the stage transition feature and the corresponding digital resource attribute feature. Based on the similarity matrix and the set of variant-related resource categories, a first mapping relationship between stage transition features and digital resource category attributes, and a second mapping relationship between pattern variant nodes and digital resource category attributes are established. By integrating the first mapping relationship and the second mapping relationship, a comprehensive mapping relationship is formed that includes stage transition characteristics, pattern variation nodes, and digital resource category attributes.

5. The internet behavior analysis method for smart libraries according to claim 4, characterized in that, The step of calculating the similarity between the feature vector and the attribute features of each digital resource, and generating a similarity matrix, includes: The feature vectors and the attributes of each digital resource are standardized to eliminate the dimensional differences between features of different dimensions. The cosine similarity algorithm is used to calculate the cosine similarity value between the standardized feature vector and each digital resource attribute feature; Each cosine similarity value is categorized and summarized according to the category attributes of the digital resources to generate a preliminary similarity list; The preliminary similarity list is sorted, and the attributes of each digital resource category are arranged in descending order of similarity value; A similarity matrix is ​​constructed based on the ranking results. The rows of the matrix correspond to the feature vectors of the transition features of the stage, the columns of the matrix correspond to the category attributes of the digital resources, and the elements in the matrix are the corresponding cosine similarity values. The similarity matrix is ​​normalized so that the values ​​of all elements in the matrix are within a preset numerical range.

6. The internet behavior analysis method for smart libraries according to claim 1, characterized in that, The process of generating a user-digital resource matching scheme based on the mapping relationship includes resource access path optimization suggestions and resource association recommendation sequences, comprising: Extract the digital resource category attribute that has the highest matching degree with the stage transition feature from the mapping relationship, and use it as the transition adaptation resource category; Based on the transitional adaptation resource categories, corresponding digital resources are selected, and the storage location and access path of these digital resources in the smart library resource repository are analyzed to generate initial resource access paths. Based on the changes in the operation instruction units in the aforementioned stage transition characteristics, the initial resource access path is adjusted to reduce unnecessary operation steps and form resource access path optimization suggestions. For the pattern mutation node, the digital resource category attribute associated with the pattern mutation node is obtained from the mapping relationship and used as the mutation adaptation resource category; Based on the aforementioned mutation-adapted resource category, other digital resources with similar attribute characteristics are searched to form a resource association candidate set; Based on the frequency of access to related resources in the user's historical online behavior sequence, the resource association candidate set is sorted to generate a resource association recommendation sequence; By integrating the resource access path optimization suggestions and the resource association recommendation sequence, a user-digital resource matching scheme is formed.

7. The internet behavior analysis method for smart libraries according to claim 6, characterized in that, The initial resource access path is adjusted based on the changes in the operation instruction units in the phase transition characteristics, reducing unnecessary operation steps and forming resource access path optimization suggestions, including: Analyze the changes in operation instruction units in the transition characteristics of the aforementioned stages to identify repetitive and invalid operations within the operation instruction units; Mark the path nodes corresponding to duplicate and invalid operations in the initial resource access path; Remove the marked path nodes from the initial resource access path and connect adjacent valid path nodes directly to form a simplified resource access path. Calculate the difference in operation steps between the simplified resource access path and the initial resource access path, ensuring that the difference reaches the preset optimization threshold; Perform integrity verification on the simplified resource access path to ensure that the target digital resource can be accessed normally through the optimized path; Add operation guidance information to the verified simplified resource access paths to form the final resource access path optimization suggestions.

8. The internet behavior analysis method for smart libraries according to claim 6, characterized in that, The step of sorting the candidate set of resource associations based on the frequency of access to related resources in the user's historical internet browsing behavior sequence to generate a resource association recommendation sequence includes: The access records of each digital resource in the resource association candidate set are extracted from the user's historical online behavior sequence, and the number of accesses and the most recent access time of each digital resource are counted. The access frequency weight of each digital resource is calculated based on the number of accesses and the time of the most recent access. The more accesses and the more recent the access time, the greater the access frequency weight. The digital resources in the resource association candidate set are sorted in descending order of access frequency weight to form a preliminary recommendation sequence; Check if there are duplicate digital resources in the initial recommendation sequence. If so, retain the digital resources with higher access frequency weights. The length of the initial recommended sequence after deduplication is adjusted to meet the preset sequence length requirements; Brief descriptions of each digital resource are added to the adjusted recommendation sequence to form the final resource association recommendation sequence.

9. The internet behavior analysis method for smart libraries according to claim 1, characterized in that, The step of pushing the adaptation solution to the user terminal of the smart library to guide users to access digital resources includes: Obtain the terminal identification information of the user logging into the Internet in the smart library. The terminal identification information is used to determine the user terminal currently being used by the user. The adaptation scheme is converted into a format suitable for display on the user terminal. The format includes text description and graphic guidance. Establish a communication connection with the user terminal and send the converted adaptation scheme to the user terminal through a preset communication protocol; Monitor the user terminal's reception status of the adaptation plan. If the reception is unsuccessful, resend the adaptation plan until the user terminal is confirmed to have received it successfully. When displaying the adaptation solution on the user terminal, set interactive prompts to guide the user to access digital resources according to the adaptation solution; It continuously receives operation response information from user terminals, determines whether the user has followed the adaptation plan based on the response information, and if not, pushes the key content of the adaptation plan again.

10. An internet behavior analysis system for smart libraries, characterized in that, The internet behavior analysis system for smart libraries includes a machine-readable storage medium and a processor. The machine-readable storage medium stores machine-executable instructions. When the processor executes the machine-executable instructions, the internet behavior analysis system for smart libraries implements the internet behavior analysis method for smart libraries as described in any one of claims 1-9.

Images