Campus information intelligent pushing method and system based on all-purpose card, and storage medium

Abstract

The invention relates to the technical field of information pushing, in particular to a campus information intelligent pushing method and system based on an all-purpose card and a storage medium. The method comprises the following steps: acquiring event vectors of student behaviors through an all-purpose card; acquiring an attention change degree according to the change of the event vectors of the current behavior and the neighborhood behavior of the student; obtaining a content conversion degree according to behavior similar conditions and attention change degree differences between the students and the students in the group with similar behaviors; according to the basic information difference between the student and the current similar student and the difference of the concerned condition of the current behavior in the current time period, the concerned offset degree is obtained; obtaining the current guiding degree according to the content conversion degree and the attention offset degree, and then obtaining the weight of each element in a collaborative filtering behavior matrix in a collaborative filtering algorithm. According to the method, the weights of the elements in the collaborative filtering behavior matrix are dynamically adjusted accurately in real time, so that the accuracy and efficiency of intelligent information pushing for students are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of information push technology, specifically to a method, system, and storage medium for intelligent campus information push based on a campus card. Background Technology

[0002] With the deepening of smart campus construction, universities have widely adopted "one-card" systems to integrate student behavior data, covering various scenarios such as consumption, access control, library borrowing, and attendance, providing a data foundation for campus information management. One-card data implicitly contains key information such as user preferences, lifestyle habits, and interests, creating conditions for precise campus information services (such as course notifications, competition activities, and internship / recruitment recommendations). Currently, mainstream campus recommendation systems analyze students' historical behavioral data (such as frequent library visits and consumption in specific locations) and use collaborative filtering algorithms to match similar user groups, achieving personalized push notifications of campus information.

[0003] However, existing collaborative filtering algorithms rely on historical behavioral data to model student behavior, ignoring the dynamic evolution of students' focus over time. For example, students may focus on course schedules and textbook acquisition at the beginning of the semester, and later shift their focus to competition registration, sports activities, or club information. This makes it difficult to identify and respond to shifts in attention in a timely manner, causing the content recommended to students through collaborative filtering algorithms to lag behind actual needs. Furthermore, the relevance of the pushed information is poor, affecting information delivery efficiency and service experience. Summary of the Invention

[0004] To address the technical problem that existing collaborative filtering algorithms ignore the dynamic evolution of students' attention over time, leading to difficulties in timely identification and response to shifts in attention and resulting in inaccurate information delivery, the present invention aims to provide a campus information intelligent delivery method, system, and storage medium based on a campus card system. The specific technical solution adopted is as follows: In a first aspect, one embodiment of the present invention provides a method for intelligent push of campus information based on a campus card, the method comprising the following steps: By using each student's campus card information, we can obtain the event vector of each student's behavior and the basic information of each student in real time. Based on the changes in the event vectors of each student's current behavior and the neighboring behaviors of that student, the degree of attention change for each student's current behavior is obtained. Based on each student's behavior within the current time period, obtain the behavioral similarity group for each student; based on the differences in behavioral similarity and attention changes between each student and each student within their behavioral similarity group within the current time period, obtain the content conversion degree of each student's current behavior; Based on the similarity of the event vectors of each student's current behavior with each other, obtain the current similar students of each student; based on the differences in basic information between each student and each of their current similar students, and the differences in the attention received by their current behavior in the current time period, obtain the attention offset degree of each student's current behavior. Based on the content conversion degree and attention shift degree of each student's current behavior, the current orientation degree of each student is obtained; based on the current orientation degree of each student and each student in the group with similar behavior, the weight of each element in the collaborative filtering behavior matrix of the collaborative filtering algorithm is obtained, and the current intelligent push information of each student is determined.

[0005] Furthermore, the method for obtaining the degree of change of concern is as follows: For any student, arrange the student's current behavior and the neighboring behaviors of the current behavior in chronological order to obtain the student's reference behavior sequence; The magnitude of the difference between the event vector of each behavior in the reference behavior sequence and its preceding adjacent behavior is used as the change analysis value of the corresponding behavior. The result of normalizing the mean of the change analysis values ​​is taken as the degree of attention change in the student's current behavior.

[0006] Furthermore, the method for obtaining the degree of content conversion is as follows: For any student, arrange the event vectors of the student's behavior within the current time period according to the chronological order of the corresponding behaviors to obtain the student's current behavior vector sequence; Students in the group with similar behavior to the student are all taken as target students. For any target student, the cosine similarity between the current behavior vector sequence of the student and the target student is obtained, which is taken as the current similarity between the student and the target student. The difference in the degree of change in attention between this student and the target student is used as the current reference difference between the two students. The product of the current similarity and the current reference difference is taken as the current deviation between the student and the target student. The normalized result of summing the current deviations of this student from each target student is taken as the content conversion degree of the student's current behavior.

[0007] Furthermore, the method for obtaining the degree of attention offset is as follows: For any given student, the representativeness of the student's current behavior is determined based on the occurrence of the student's current behavior within the current time period and the degree of change in the attention given to the student's current behavior. For any current similar student of the student, the magnitude of the difference between the basic information vectors of the student and the current similar student is negatively correlated and normalized, and the result is taken as the degree of similarity between the basic information of the student and the current similar student. The difference in representativeness between the current behavior of this student and that of the current similar student is taken as the representativeness difference between the current behavior of this student and that of the current similar student. The product of the similarity of the basic information and the difference represented by the current behavior is used as the attention deviation analysis value between the student and the currently similar student. The mean of the attentional bias analysis values ​​between the student and all currently similar students is used as the degree of attentional shift in the student's current behavior.

[0008] Furthermore, the method for obtaining the degree of representativeness is as follows: The ratio of the number of times the student's current behavior occurs in the current time period to the total number of times the student's behavior occurs in the current time period is used as the first feature value; The product of the first feature value and the normalized result of the negative correlation between the student's current behavior and the degree of attention change is used as the representativeness of the student's current behavior.

[0009] Furthermore, the method for obtaining the current guidance level is as follows: The product of each student's current behavior's content conversion level and attention shift level is used as each student's current orientation level.

[0010] Furthermore, the method for obtaining the behaviorally similar groups is as follows: Based on the behavior of each student within the current time period, a collaborative filtering algorithm is used to obtain groups of students whose behavior is similar to that of each student within the current time period.

[0011] Furthermore, the method for obtaining the current similar students is as follows: For any student, the normalized result of the difference between the event vector of the current behavior of the student and each other's current behavior is used as the similarity analysis value of the current behavior of the student and each other's current behavior. When the current behavior similarity analysis value is less than the preset behavior similarity threshold, the corresponding student will be regarded as the current similar student of that student.

[0012] Secondly, another embodiment of the present invention provides a campus information intelligent push system based on a smart card, the system comprising: The data acquisition module is used to obtain the event vector of each student's behavior and the basic information of each student in real time through each student's campus card information; The attention change acquisition module is used to acquire the attention change degree of each student's current behavior based on the changes in the event vector of each student's current behavior and the neighboring behaviors of the current behavior; The content conversion degree acquisition module is used to acquire the behavioral similarity group of each student based on the behavior of each student in the current time period; and to acquire the content conversion degree of each student's current behavior based on the behavioral similarity and attention change differences between each student and each student in the behavioral similarity group in the current time period. The attention offset degree acquisition module is used to obtain the current similar students of each student based on the similarity of the event vectors of each student's current behavior with each other; and to obtain the attention offset degree of each student's current behavior based on the differences in basic information between each student and each of their current similar students, as well as the differences in the attention received by the current behavior within the current time period. The data processing module is used to obtain the current orientation level of each student based on the content conversion level and attention shift level of each student's current behavior; based on the current orientation level of each student and each student in the group with similar behavior, it obtains the weight of each element in the collaborative filtering behavior matrix in the collaborative filtering algorithm, and determines the current intelligent push information for each student.

[0013] Thirdly, another embodiment of the present invention provides a computer-readable storage medium comprising: a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, it implements the steps of any of the above methods.

[0014] The present invention has the following beneficial effects: This invention first obtains the degree of attention change of each student's current behavior based on the changes in the event vectors of the behaviors of each student's current behavior and those of its neighbors, thus initially reflecting the possibility that each student's current behavior has changed. To accurately analyze whether each student's current behavior has changed, enabling accurate campus information recommendations and reducing irrelevant recommendations, the invention further obtains the content conversion degree of each student's current behavior based on the differences in behavioral similarity and attention change among students in the same behavioral similarity group within the current time period, further reflecting the change in each student's current behavior. To more accurately analyze whether each student's current behavior has changed, the invention further obtains the attention shift of each student's current behavior based on the differences in basic information between each student and each of their current similar students, as well as the differences in the attention received by their current behavior within the current time period. The degree of change further reflects the current behavior of each student; then, based on the degree of content transformation and attention shift of each student's current behavior, the current orientation degree of each student is obtained, accurately reflecting the possibility of deviation in each student's current behavior, dynamically identifying the dominant attention direction of students at the current stage, improving the relevance and timeliness of information push, and effectively avoiding problems such as delayed and irrelevant push content; furthermore, based on the current orientation degree of each student and each student in the group with similar behavior, the weight of each element in the collaborative filtering behavior matrix of the collaborative filtering algorithm is obtained, realizing the real-time dynamic adjustment of the weights in the collaborative filtering behavior matrix, effectively improving the ability of the campus management system to model the balance between students' long-term and short-term behaviors; thus, accurately and efficiently determining the current intelligent push information for each student, effectively improving the relevance and student acceptance of the pushed campus information content, and improving the student experience. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic flowchart illustrating a campus information intelligent push method based on a smart card system, provided as an embodiment of the present invention. Figure 2 This is a structural diagram of a campus information intelligent push system based on a smart card, provided as an embodiment of the present invention. Figure 3 This is a schematic diagram of a computer device provided according to an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation methods, structures, features, and effects of the campus information intelligent push method, system, and storage medium based on a smart card system proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details the specific solutions for the intelligent campus information push method, system, and storage medium based on the all-in-one card provided by this invention.

[0020] Example 1: This invention proposes a method for intelligent campus information push based on a campus card system. Please refer to [link / reference]. Figure 1 The diagram illustrates a schematic flowchart of a campus information intelligent push method based on a smart card system, according to an embodiment of the present invention. The method includes the following steps: Step S1: Obtain the event vector of each student's behavior and the basic information of each student in real time through each student's campus card information.

[0021] Specifically, the campus management system can obtain the event vector of each student's actions and each student's basic information in real time from each student's campus card information. It should be noted that an action includes consumption data (such as purchase records at print shops, supermarkets, canteens, etc.), access control card swipe records (such as entry and exit data for libraries, dormitories, teaching buildings, laboratories, etc.), book borrowing records, class attendance data, library seat reservation records, and registration data for various campus activities, competitions, lectures, etc. Each action also includes a unique student identifier (such as student ID), action type, time, location, device number, and other auxiliary information. The data related to each student's actions are constructed into a vector, namely the event vector of each action. Each event vector has the same length, and the elements at the same position in each event vector correspond to the same data type. Basic information includes the corresponding student's student ID, department, age, and gender.

[0022] To accurately analyze each student's recent behavior and recommend more appropriate behaviors in the campus management system, this embodiment only analyzes each student's non-essential behaviors. Non-essential behaviors refer to students' personal activities other than necessary activities like attending classes and eating. When a student focuses on a particular behavior, their behavior pattern usually exhibits a clear focus, frequently visiting specific locations and concentrating on related activities. If the student shifts their focus to another event due to external stimuli (such as teacher suggestions, social influences, or personal plans), their behavior pattern will change accordingly. For example, a student who previously frequently read in the library might now seek out quieter, more secluded study rooms for targeted study in preparation for a subject competition.

[0023] Step S2: Based on the changes in the event vectors of each student's current behavior and the neighboring behaviors of their current behavior, obtain the degree of attention change for each student's current behavior.

[0024] Specifically, when a student's focus shifts, their behavioral patterns gradually adjust, manifesting as a series of new behaviors different from their previous paths. If multiple consecutive behaviors deviate from the previous pattern, it indicates a substantial change in the content the student is focusing on, and the greater the deviation, the more significant the change in focus. Therefore, this embodiment obtains the degree of change in each student's current focus based on the changes in the event vectors of each student's current behavior and its neighboring behaviors. The greater the degree of change in focus, the greater the likelihood that the student's current behavior has changed. In this embodiment, the three historical behaviors most closely related to each student's current behavior in time sequence are considered as the neighboring behaviors of each student's current behavior. Implementers can set the neighboring behaviors of each student's current behavior according to actual circumstances, and this is not limited here.

[0025] Preferably, in one feasible embodiment of this invention, the method for obtaining the degree of attention change is as follows: For any student, the student's current behavior and the neighboring behaviors of the current behavior are arranged in chronological order to obtain a reference behavior sequence for the student; the magnitude of the difference between the event vector of each behavior in the reference behavior sequence and its preceding adjacent behavior is used as the change analysis value of the corresponding behavior; when the change analysis values ​​are larger, it indicates that multiple consecutive behaviors have deviated, indirectly indicating that the attention change of the student's current behavior is more obvious. Therefore, in this embodiment, the mean of the change analysis values ​​is normalized as the degree of attention change of the student's current behavior. Specifically, in this embodiment, the mean of the change analysis values ​​is normalized using a normalization function.

[0026] This allows us to obtain the degree of change in each student's current behavior.

[0027] Step S3: Based on each student's behavior within the current time period, obtain the behavioral similarity group for each student; based on the differences in behavioral similarity and attention changes between each student and each student within their behavioral similarity group within the current time period, obtain the content conversion degree of each student's current behavior.

[0028] Specifically, under similar academic stages, professional backgrounds, and lifestyles, some students often exhibit strong behavioral similarities. When there is a significant difference in the degree of attention change between a student's current behavior and that of other students with similar behaviors, it indicates a greater likelihood of a change in the content of that student's current behavior. Therefore, this embodiment first identifies the behavioral similarity group for each student based on their behavior within the current time period, accurately filtering out students with similar behaviors to each student within the current time period. Then, based on the differences in behavioral similarity and attention change between each student and each student within their behavioral similarity group within the current time period, it obtains the degree of content conversion for each student's current behavior. The greater the degree of content conversion, the more likely the student's current attention content has changed, and the more the recommended behavior for that student should consider the bias of their current behavior. It should be noted that this embodiment sets the duration of the current time period to 7 days. Implementers can adjust the length of the current time period according to actual circumstances; this is not limited here, but the end time of the current time period must be the current time.

[0029] Preferably, in one feasible implementation of this embodiment, the method for obtaining similar groups is as follows: based on the behavior of each student within the current time period, a collaborative filtering algorithm is used to obtain the behavioral similar groups of each student within the current time period. The collaborative filtering algorithm is a well-known technique and will not be described in detail here.

[0030] Preferably, in one feasible embodiment of this method, the method for obtaining the degree of content conversion is as follows: For any student, the event vectors of the student's behavior within the current time period are arranged according to the chronological order of the corresponding behaviors from beginning to end to obtain the student's current behavior vector sequence, which accurately represents the student's behavior within the current time period; students in the group with similar behaviors to the student are all taken as target students; for any target student, the cosine similarity between the current behavior vector sequence of the student and the target student is obtained as the current similarity degree between the student and the target student; it should be noted that if the lengths of the current behavior vector sequences of the student and the target student are different, the shorter current behavior vector sequence is supplemented with 0. The greater the current similarity degree, the more similar the overall behavior of the student and the target student is within the current time period; the greater the difference in the degree of attention change between the current behavior of the student and the target student, the more accurately it can be said that the student's current behavior has undergone content change. The absolute value of the difference between the student's and the target student's changes in attention is then used as the current reference difference between the two students. The product of the current similarity and the current reference difference is then used as the current deviation between the student and the target student. The greater the current deviation, the more the student's current behavior deviates. To more accurately represent the content change in the student's current behavior, the sum of the current deviations between the student and each target student is normalized, and this sum is used as the content conversion degree of the student's current behavior. In this embodiment, the sum of the current deviations between the student and each target student is normalized using the norm normalization function.

[0031] At this point, the content conversion level of each student's current behavior is obtained.

[0032] Step S4: Based on the similarity of the event vectors of each student's current behavior with each other, obtain the current similar students of each student; based on the differences in basic information between each student and each of their current similar students, and the differences in the attention received by the current behavior within the current time period, obtain the attention offset degree of each student's current behavior.

[0033] Specifically, to more accurately analyze whether each student's current behavior has changed, this embodiment further obtains each student's current similar students based on the similarity of the event vectors of each student's current behavior with every other student. The more similar a student's basic information is to their current similar students, and the less similar their current behavior is within the current time period, the more likely that the student's current behavior has changed. Furthermore, this embodiment obtains the degree of attention shift for each student's current behavior based on the differences in basic information between each student and each of their current similar students, as well as the differences in the level of attention received by their current behavior within the current time period. The greater the degree of attention shift, the more likely the student's current behavior has deviated.

[0034] Preferably, in one feasible implementation of this embodiment, the method for obtaining current similar students is as follows: For any student, the normalized result of the magnitude of the difference between the event vector of the student's current behavior and that of each other is used as the current behavior similarity analysis value of the student and each other. In this embodiment, the magnitude of the difference between the event vector of the student's current behavior and that of each other is normalized using the norm normalization function. The smaller the current behavior similarity analysis value, the more similar the student's current behavior is to that of other corresponding students. Therefore, this embodiment sets a preset behavior similarity threshold of 0.2. The implementer can set the size of the preset behavior similarity threshold according to the actual situation, which is not limited here. When the current behavior similarity analysis value is less than the preset behavior similarity threshold, the corresponding student is taken as the current similar student of that student.

[0035] At this point, we have obtained the current similar students for each student.

[0036] Preferably, in one feasible embodiment of this invention, the method for obtaining the degree of attention shift is as follows: For any student, based on the occurrence of the student's current behavior within the current time period and the degree of attention change of the student's current behavior, the representativeness of the student's current behavior is obtained; the greater the representativeness, the more meaningful and representative the student's current behavior is. Specifically, the method for obtaining the representativeness is: the ratio of the number of times the student's current behavior occurs within the current time period to the total number of the student's behaviors within the current time period is used as the first feature value; the larger the first feature value, the higher the frequency of the student's current behavior within the current time period, and the more representative the student's current behavior is; the smaller the degree of attention change of the student's current behavior, the more representative the student's current behavior is also indirectly reflected; furthermore, in this embodiment, the product of the first feature value and the negatively correlated and normalized result of the student's current behavior's attention change is used as the representativeness of the student's current behavior. Specifically, in this embodiment, the negative of the student's current behavior's attention change is used as the power of an exponential function with a natural constant as the base, and the output of this exponential function is the negatively correlated and normalized result of the student's current behavior's attention change. For any currently similar student of the given student, the magnitude of the difference between the basic information vectors corresponding to the student and the currently similar student is negatively correlated and normalized. This result is used as the similarity degree of the basic information between the student and the currently similar student. The greater the similarity degree, the more identical the basic information of the student and the currently similar student. In this embodiment, the similarity number of the magnitude of the difference between the basic information vectors corresponding to the student and the currently similar student is used as the power of an exponential function with the natural constant as the base. The output of this exponential function is the result of negatively correlated and normalized magnitude of the difference between the basic information vectors corresponding to the student and the currently similar student. It should be noted that the vector corresponding to the basic information of each student is a vector formed by converting the student's basic information into data. The length of the vector corresponding to the basic information of each student must be the same, and the basic information types corresponding to the elements at the same position in different vectors must be the same. The more similar the basic information of a student to that of a currently similar student, and the greater the difference in the representativeness of the current behavior between the two students, the more it indicates a change in the content of the student's current behavior. Therefore, the absolute value of the difference in the representativeness of the current behavior between the student and the currently similar student is taken as the difference in the representativeness of the current behavior between the two students. Then, the product of the similarity of basic information and the difference in the representativeness of current behavior is taken as the attention deviation analysis value between the student and the currently similar student. The larger the attention deviation analysis value, the greater the deviation in the current behavior reflected by the currently similar student. To accurately characterize the deviation in the student's current behavior, the mean of the attention deviation analysis values ​​between the student and all currently similar students is taken as the degree of attention shift in the student's current behavior.

[0037] At this point, the degree of attention shift for each student's current behavior is obtained.

[0038] Step S5: Based on the content conversion degree and attention shift degree of each student's current behavior, obtain the current guidance degree of each student; based on the current guidance degree of each student and each student in the group with similar behavior, obtain the weight of each element in the collaborative filtering behavior matrix in the collaborative filtering algorithm, and determine the current intelligent push information for each student.

[0039] Specifically, known collaborative filtering algorithms use time-decay weighting to model students' historical behavior, which doesn't closely reflect the actual trends of changes in students' attention over time. This makes it difficult to respond promptly to shifts in students' attention, resulting in content recommendations based on existing collaborative filtering algorithms being outdated or lacking relevance. To improve the accuracy of content recommendations, this embodiment dynamically adjusts the weights of the collaborative filtering algorithm based on the student's current focus, achieving more precise information matching. It is known that student card behavior data in a closed campus environment has similar structures, consistent temporal sequences, and strong comparability. By comparing student groups with similar current behavioral trajectories, it's possible to effectively infer students' current focus. Furthermore, the greater the degree of content transformation and attention shift in a student's current behavior, the more significant the change in that student's current behavior compared to similar student groups, indirectly indicating a stronger focus orientation. Therefore, this embodiment obtains each student's current orientation level based on the degree of content transformation and attention shift in their current behavior. The greater the current orientation level, the stronger the student's current focus orientation. The current guidance level is obtained by multiplying the content conversion level and attention shift level of each student's current behavior. The value of the current guidance level ranges from 0 to 1.

[0040] At this point, the current orientation level of each student is obtained.

[0041] To provide accurate and timely behavioral recommendations for each student and enhance their acceptance of the campus management system, this paper dynamically determines the weight of each student's corresponding element in the collaborative filtering behavior matrix within the collaborative filtering algorithm. This allows for real-time adjustment of the weights in the collaborative filtering behavior matrix, thereby accurately and efficiently determining the current intelligent push information for each student. For example, if a student's current behavior (such as entering the library or registering for a science competition) indicates a higher likelihood of a shift in focus, the corresponding element in the collaborative filtering matrix should be assigned a higher weight, making this behavior more guiding in similar user matching and recommendation results. By introducing a "behavioral guidance weighting mechanism," a sensitive response to shifts in student focus is achieved, thereby improving the accuracy and timeliness of campus information push. It should be noted that the elements in the collaborative filtering behavior matrix represent the behaviors of each student within the same behavioral similarity group during the current time period. The collaborative filtering behavior matrix and the weights of each element in the matrix are well-known technologies and will not be elaborated further.

[0042] In summary, this embodiment obtains the event vector of student behavior through the student ID card; it obtains the degree of attention change based on the changes in the event vectors of the student's current behavior and neighboring behaviors; it obtains the degree of content conversion based on the similarity of behavior and the difference in the degree of attention change between the student and students in the same behavioral group; it obtains the degree of attention shift based on the differences in basic information between the student and currently similar students, and the differences in the attention received by the current behavior within the current time period; and it obtains the current guidance degree based on the degree of content conversion and the degree of attention shift, thereby obtaining the weight of each element in the collaborative filtering behavior matrix in the collaborative filtering algorithm. This invention effectively improves the accuracy and efficiency of intelligent information push to students by dynamically adjusting the weights of elements in the collaborative filtering behavior matrix in real time and accurately.

[0043] Example 2: This invention also proposes a campus information intelligent push system based on a smart card system; please refer to [link / reference]. Figure 2 The diagram illustrates a structure of a campus information intelligent push system based on a smart card system according to an embodiment of the present invention. The system includes: a data acquisition module 10, a attention change degree acquisition module 20, a content conversion degree acquisition module 30, an attention offset degree acquisition module 40, and a data processing module 50.

[0044] The data acquisition module 10 is used to acquire the event vector of each student's behavior and the basic information of each student in real time through each student's campus card information.

[0045] The attention change acquisition module 20 is used to acquire the attention change degree of each student's current behavior based on the changes in the event vector of each student's current behavior and the neighboring behaviors of the current behavior.

[0046] The content conversion degree acquisition module 30 is used to acquire the behavioral similarity group of each student based on the behavior of each student in the current time period; and to acquire the content conversion degree of each student's current behavior based on the behavioral similarity and attention change differences between each student and each student in the behavioral similarity group in the current time period.

[0047] The attention offset degree acquisition module 40 is used to obtain the current similar students of each student based on the similarity of the event vectors of each student's current behavior with each other; and to obtain the attention offset degree of each student's current behavior based on the differences in basic information between each student and each of their current similar students, as well as the differences in the attention received by the current behavior in the current time period.

[0048] The data processing module 50 is used to obtain the current orientation level of each student based on the content conversion level and attention shift level of each student's current behavior; based on the current orientation level of each student and each student in the group with similar behavior, it obtains the weight of each element in the collaborative filtering behavior matrix in the collaborative filtering algorithm, and determines the current intelligent push information for each student.

[0049] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the campus information intelligent push system based on a smart card and the campus information intelligent push method based on a smart card provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiment, which will not be repeated here.

[0050] Example 3: The present invention also proposes a computer device, see [link to relevant documentation]. Figure 3 The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402. When the processor 402 executes the computer program 403, the computer device can execute any of the aforementioned campus information intelligent push methods based on the all-in-one card.

[0051] Example 4: The present invention also provides a computer-readable storage medium storing computer program code, which, when executed on a computer, causes the computer to perform the aforementioned method steps to implement the campus information intelligent push method based on a smart card provided in the above embodiments.

[0052] Example 5: The present invention also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to realize the campus information intelligent push method based on a smart card provided in the above embodiments.

[0053] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0054] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0055] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for intelligent push of campus information based on a campus card, characterized in that, The method includes the following steps: By using each student's campus card information, we can obtain the event vector of each student's behavior and the basic information of each student in real time. Based on the changes in the event vectors of each student's current behavior and the neighboring behaviors of that student, the degree of attention change for each student's current behavior is obtained. Based on each student's behavior within the current time period, obtain the behavioral similarity group for each student; based on the differences in behavioral similarity and attention changes between each student and each student within their behavioral similarity group within the current time period, obtain the content conversion degree of each student's current behavior; Based on the similarity of the event vectors of each student's current behavior with each other, obtain the current similar students of each student; based on the differences in basic information between each student and each of their current similar students, and the differences in the attention received by their current behavior in the current time period, obtain the attention offset degree of each student's current behavior. Based on the content conversion degree and attention shift degree of each student's current behavior, the current orientation degree of each student is obtained; based on the current orientation degree of each student and each student in the group with similar behavior, the weight of each element in the collaborative filtering behavior matrix of the collaborative filtering algorithm is obtained, and the current intelligent push information of each student is determined.

2. The campus information intelligent push method based on a smart card system as described in claim 1, characterized in that, The method for obtaining the degree of change of concern is as follows: For any student, arrange the student's current behavior and the neighboring behaviors of the current behavior in chronological order to obtain the student's reference behavior sequence; The magnitude of the difference between the event vector of each behavior in the reference behavior sequence and its preceding adjacent behavior is used as the change analysis value of the corresponding behavior. The result of normalizing the mean of the change analysis values ​​is taken as the degree of attention change in the student's current behavior.

3. The campus information intelligent push method based on a smart card system as described in claim 1, characterized in that, The method for obtaining the content conversion level is as follows: For any student, arrange the event vectors of the student's behavior within the current time period according to the chronological order of the corresponding behaviors to obtain the student's current behavior vector sequence; Students in the group with similar behavior to the student are all taken as target students. For any target student, the cosine similarity between the current behavior vector sequence of the student and the target student is obtained, which is taken as the current similarity between the student and the target student. The difference in the degree of change in attention between this student and the target student is used as the current reference difference between this student and the target student; The product of the current similarity and the current reference difference is taken as the current deviation between the student and the target student. The normalized result of summing the current deviations of this student from each target student is taken as the content conversion degree of the student's current behavior.

4. The campus information intelligent push method based on a smart card system as described in claim 1, characterized in that, The method for obtaining the degree of attention offset is as follows: For any given student, the representativeness of the student's current behavior is determined based on the occurrence of the student's current behavior within the current time period and the degree of change in the attention given to the student's current behavior. For any current similar student of the student, the magnitude of the difference between the basic information vectors of the student and the current similar student is negatively correlated and normalized, and the result is taken as the degree of similarity between the basic information of the student and the current similar student. The difference in representativeness between the current behavior of this student and that of the current similar student is taken as the representativeness difference between the current behavior of this student and that of the current similar student. The product of the similarity of the basic information and the difference represented by the current behavior is used as the attention deviation analysis value between the student and the currently similar student. The mean of the attentional bias analysis values ​​between the student and all currently similar students is used as the degree of attentional shift in the student's current behavior.

5. The campus information intelligent push method based on a smart card system as described in claim 4, characterized in that, The method for obtaining the representativeness is as follows: The ratio of the number of times the student's current behavior occurs in the current time period to the total number of times the student's behavior occurs in the current time period is used as the first feature value; The product of the first feature value and the normalized result of the negative correlation between the student's current behavior and the degree of attention change is used as the representativeness of the student's current behavior.

6. The campus information intelligent push method based on a smart card system as described in claim 1, characterized in that, The method for obtaining the current guidance level is as follows: The product of each student's current behavior's content conversion level and attention shift level is used as each student's current orientation level.

7. The campus information intelligent push method based on a smart card system as described in claim 1, characterized in that, The method for obtaining the behaviorally similar groups is as follows: Based on the behavior of each student within the current time period, a collaborative filtering algorithm is used to obtain groups of students whose behavior is similar to that of each student within the current time period.

8. The campus information intelligent push method based on a smart card system as described in claim 1, characterized in that, The method for obtaining the current similar students is as follows: For any student, the normalized result of the difference between the event vector of the current behavior of the student and each other's current behavior is used as the similarity analysis value of the current behavior of the student and each other's current behavior. When the current behavior similarity analysis value is less than the preset behavior similarity threshold, the corresponding student will be regarded as the current similar student of that student.

9. A campus information intelligent push system based on a smart card, characterized in that, The system includes: The data acquisition module is used to obtain the event vector of each student's behavior and the basic information of each student in real time through each student's campus card information; The attention change acquisition module is used to acquire the attention change degree of each student's current behavior based on the changes in the event vector of each student's current behavior and the neighboring behaviors of the current behavior; The content conversion degree acquisition module is used to acquire the behavioral similarity group of each student based on the behavior of each student in the current time period; and to acquire the content conversion degree of each student's current behavior based on the behavioral similarity and attention change differences between each student and each student in the behavioral similarity group in the current time period. The attention offset degree acquisition module is used to obtain the current similar students of each student based on the similarity of the event vectors of each student's current behavior with each other; and to obtain the attention offset degree of each student's current behavior based on the differences in basic information between each student and each of their current similar students, as well as the differences in the attention received by the current behavior within the current time period. The data processing module is used to obtain the current orientation level of each student based on the content conversion level and attention shift level of each student's current behavior; based on the current orientation level of each student and each student in the group with similar behavior, it obtains the weight of each element in the collaborative filtering behavior matrix in the collaborative filtering algorithm, and determines the current intelligent push information for each student.

10. A computer-readable storage medium comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the campus information intelligent push method based on a smart card system as described in any one of claims 1-8.

Images