Video course watching duration statistical method and device, interest analysis system and equipment

Abstract

The invention provides a video course watching duration statistical method and device, an interest analysis system and computer equipment, and relates to the technical field of videos, and the method comprises the steps: determining a target video course bound with a user identifier according to the user identifier; collecting watching behavior data of the user on the target video course in real time; according to the watching behavior data, obtaining an initial watching duration of the target video course by the user; according to the watching behavior data, the initial watching duration is calibrated, and the single watching duration of the target video course by the user is determined; according to the method, at least one single watching time length of the target video course by the user is accumulated, and the total watching time length of the target video course by the user is determined, so that invalid watching behaviors can be effectively filtered, and watching time periods with real participation, active interaction and concentrated attention can be accurately recorded, and therefore, high-precision statistics of the effective watching time lengths can be realized, and the user experience can be improved. And the authenticity and credibility of the learning data are improved.

Description

Technical Field

[0001] This invention relates to the field of video technology, and in particular to a method, apparatus, interest analysis system, and device for calculating video course viewing time. Background Technology

[0002] With the popularization of online education, learning through video courses has become commonplace. However, platforms have revealed significant shortcomings in the accuracy of learning data statistics and analysis. Existing technologies can typically only record basic single-view durations, failing to address the complexities unique to online education. This leads to two core problems: First, the system cannot effectively identify and filter ineffective learning behaviors such as abnormally fast viewing speeds and background playback, resulting in severely distorted statistical durations that fail to accurately reflect users' effective engagement. Second, when course versions are updated, the system cannot distinguish between users rewatching old content and first-time viewing of new content, simply accumulating the viewing time, ultimately causing chaotic and inaccurate learning data that cannot support effective learning analysis.

[0003] Therefore, in view of the shortcomings of existing technical solutions, the present invention provides a method for calculating the viewing time of video courses. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, interest analysis system, and equipment for calculating video course viewing time to address the aforementioned technical problems.

[0005] On the one hand, a method for calculating video course viewing time is provided. The method includes: determining the target video course bound to the user identifier based on the user identifier; collecting user viewing behavior data of the target video course in real time, wherein the viewing behavior data includes at least one of the following: course identifier, course version number, speed adjustment record, learning interaction behavior, background switching record, start viewing time, and end viewing time; obtaining the user's initial viewing time of the target video course based on the start viewing time and end viewing time; calibrating the initial viewing time based on at least one of the following: course version number, speed adjustment record, learning interaction behavior, and background switching record, to determine the user's single viewing time of the target video course; and summing up at least one single viewing time of the user's target video course to determine the user's total viewing time of the target video course.

[0006] Optionally, real-time data collection of user viewing behavior on the target video course includes: recording the end-of-view time in response to detecting that the user actively closes the video playback interface corresponding to the target video course; and / or recording the end-of-view time in response to detecting that the duration of pausing playback of the target video course exceeds the preset pause duration; and / or recording the end-of-view time in response to detecting switching to other video courses; and / or recording the end-of-view time when the current video ends in response to detecting background playback.

[0007] Optionally, the initial viewing duration can be calibrated based on at least one of the following: course version number, speed adjustment records, learning interaction behavior, and background switching records, to determine the user's single viewing duration for the target video course. This includes: performing chapter matching calibration on the initial viewing duration based on the course version number to determine the first invalid viewing duration; performing speed adjustment calibration on the initial viewing duration based on the speed adjustment records to determine the second invalid viewing duration; performing background playback calibration on the initial viewing duration based on the learning interaction behavior and background switching records to determine the third invalid viewing duration; determining the invalid viewing duration based on the first, second, and third invalid viewing durations; and determining the user's single viewing duration for the target video course based on the initial viewing duration and the invalid viewing durations.

[0008] Optionally, based on the course version number, chapter matching calibration is performed on the initial viewing time to determine the first invalid viewing time, including: determining the chapter division information of the target video course based on the course version number; determining the chapter coverage information based on the user's viewing behavior data of the target video course; and performing chapter matching calibration on the initial viewing time based on the chapter coverage information and the chapter division information to determine the first invalid viewing time.

[0009] Optionally, based on the speed adjustment record, the initial viewing duration is calibrated by a multiplier to determine the second invalid viewing duration, including: based on the speed adjustment record, obtaining at least one viewing speed of the user and the corresponding speed duration of the viewing speed; in response to detecting that the user's viewing speed is greater than a preset speed, obtaining the speed duration corresponding to the viewing speed; in response to the speed duration corresponding to the viewing speed being greater than a preset duration, taking the speed duration of the viewing speed being greater than the preset speed as the second invalid viewing duration.

[0010] Optionally, based on the learning interaction behavior and background switching records, the initial viewing time is calibrated by background playback to determine the third invalid viewing time, including: in response to detecting that the user switches to background playback, detecting the learning interaction behavior and recording the no-interaction time; in response to the no-interaction time being longer than the preset no-interaction time, obtaining the portion of the no-interaction time exceeding the preset no-interaction time as the third invalid viewing time.

[0011] Optionally, the total viewing time of a user for a target video course is determined by summing up at least one single viewing duration of the user, including: obtaining the course version number and chapter division information viewed by the user; in response to the user viewing at least once the same version of the target video course, deduplicating and summing up the chapters of the target video course to determine the total viewing time; in response to the user viewing at least once different versions of the target video course, comparing the newly added chapters of the target video courses of different versions, and summing up the newly added chapters in full; and deduplicating and summing up common chapters of the target video courses of different versions.

[0012] On the other hand, a video course viewing duration statistics device is provided. The device includes: a first processing module for determining the target video course bound to the user identifier based on the user identifier; a second processing module for collecting user viewing behavior data of the target video course in real time, wherein the viewing behavior data includes at least one of course identifier, course version number, speed adjustment record, learning interaction behavior, background switching record, start viewing time and end viewing time; a third processing module for obtaining the user's initial viewing duration of the target video course based on the start viewing time and end viewing time; a fourth processing module for calibrating the initial viewing duration based on at least one of course version number, speed adjustment record, learning interaction behavior and background switching record to determine the user's single viewing duration of the target video course; and a fifth processing module for accumulating at least one single viewing duration of the user's target video course to determine the user's total viewing duration of the target video course.

[0013] On the other hand, an interest analysis system is provided, comprising: a storage module for storing at least one of user information, course information, viewing behavior information, single viewing duration data, cumulative viewing duration data, and total learning duration data, and establishing a multi-dimensional correlation index; an interest analysis module for analyzing user completion rate, interactivity, and group differences based on the data in the storage module, and determining user-interested courses; and a display module for visually displaying at least one of user-interested courses, learning alerts, course recommendations, and course optimization feedback.

[0014] On another front, a computer device is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it performs the following steps: determining a target video course bound to a user identifier; collecting real-time viewing behavior data of the user on the target video course, wherein the viewing behavior data includes at least one of course identifier, course version number, speed adjustment records, learning interaction behavior, background switching records, start viewing time, and end viewing time; obtaining the user's initial viewing duration for the target video course based on the start viewing time and end viewing time; calibrating the initial viewing duration based on at least one of the course version number, speed adjustment records, learning interaction behavior, and background switching records to determine the user's single viewing duration for the target video course; and summing the user's at least one single viewing duration for the target video course to determine the user's total viewing duration for the target video course.

[0015] The aforementioned video course viewing time statistics method, device, interest analysis system, and computer equipment include the following steps: First, identifying the target video course bound to the user's identifier. Second, collecting real-time user viewing behavior data for the target video course, including at least one of the following: course identifier, course version number, speed adjustment records, learning interaction behavior, background switching records, start viewing time, and end viewing time. Third, obtaining the user's initial viewing time for the target video course based on the start and end viewing times. Fourth, calibrating the initial viewing time based on at least one of the following: course version number, speed adjustment records, learning interaction behavior, and background switching records, to determine the user's single viewing time for the target video course. Fifth, summing up at least one single viewing time for the target video course to determine the user's total viewing time for the target video course. Thus, through intelligent recognition and behavior analysis algorithms, invalid viewing behavior can be effectively filtered out, and viewing periods of genuine participation, active interaction, and focused attention can be accurately recorded, thereby achieving high-precision statistics of effective viewing time and improving the authenticity and credibility of learning data. Attached Figure Description

[0016] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a method for calculating video course viewing time in one embodiment; Figure 2 This is a structural block diagram of a video course viewing time statistics device in one embodiment; Figure 3 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] It should be understood that, in the description of this application, unless the context explicitly requires it, words such as "including" or "comprising" throughout the specification should be interpreted as including rather than exclusive or exhaustive; that is, meaning "including but not limited to".

[0020] It should also be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0021] It should be noted that the terms "S1," "S2," etc., are used only for descriptive purposes and do not specifically refer to the order or sequence, nor are they intended to limit this application. They are merely for the convenience of describing the method of this application and should not be construed as indicating the sequential order of the steps. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0022] In one embodiment, such as Figure 1 As shown, a method for calculating video course viewing time is provided, including the following steps: S101: Determine the target video course bound to the user identifier based on the user identifier.

[0023] One user ID can be linked to one or more video courses.

[0024] Specifically, the target video courses could be, for example, advanced functions in high school mathematics, breakthroughs in high school English listening comprehension, basic mechanics in high school physics, and special topics in high school geometry.

[0025] Each video course includes at least one video chapter.

[0026] Specifically, user identifiers and user information are also related. A corresponding user identifier can be generated based on the user information entered by the user, and at least one corresponding video course can be generated based on the user identifier. For example, if a user enters "Grade: Senior 3, Major: Science, Learning Goal: College Entrance Examination Preparation", a user identifier S002 is generated. Correspondingly, if the user identifier is identified as S, a course corresponding to the group "Grade: Senior 3, Major: Science, Learning Goal: College Entrance Examination Preparation" will be configured. Users can adjust the courses according to their own needs.

[0027] S102: Collect user viewing behavior data for the target video course in real time. The viewing behavior data includes at least one of the following: course identifier, course version number, speed adjustment record, learning interaction behavior, background switching record, start viewing time and end viewing time.

[0028] Specifically, the speed adjustment record can include the user adjusting from 1x speed to 2x speed, or from 1x speed to 1.5x speed.

[0029] Specifically, learning interaction behaviors can include clicking on exercise links, triggering note-taking functions, submitting questions, etc.

[0030] Specifically, viewing behavior data can include the user's first viewing, course C001, version V1.0, starting at 08:30 on September 1, 2025 and ending at 09:40 on September 1, 2025, at 1.5x speed throughout, during which the user clicked the exercise link 3 times, took notes once, and did not switch to the background; the second viewing, course C001, version V1.1, with the newly added Chapter 6 "Function Extension", lasting 20 minutes, starting at 14:00 on September 3, 2025 and ending at 14:50 on September 3, 2025, at 2x speed for the first 30 minutes and 1.2x speed for the last 20 minutes, during which the user switched to the background for 18 minutes, with no background interaction, and clicked the exercise link 2 times.

[0031] S103: Based on the start and end times of viewing, obtain the user's initial viewing duration for the target video course.

[0032] Here, the start viewing time is the time when the user clicks to enter the target video course.

[0033] The end-of-view time can be either the time when the user actively closes the video or the time when the user is determined to have ended watching.

[0034] S104: Based on at least one of the following: course version number, speed adjustment record, learning interaction behavior, and background switching record, calibrate the initial viewing duration to determine the user's single viewing duration for the target video course.

[0035] Specifically, through calibration, invalid viewing behaviors such as "abnormal viewing speed" (e.g., watching for a long time at a speed more than 2.0 times that of normal learning) or "background playback" (continuous timing without interaction after switching to the background) are eliminated.

[0036] S105: Accumulate the user's at least one single viewing time for the target video course to determine the user's total viewing time for the target video course.

[0037] In one embodiment, it may further include: comparing the total viewing time with a preset viewing time limit; and in response to the total viewing time being greater than the preset viewing time limit, using the viewing time limit as the total viewing time.

[0038] The preset maximum viewing time can be 1.5 times the total duration of the latest course version.

[0039] Specifically, for the same course ID, the maximum cumulative viewing time is set to 1.5 times the total duration of the current latest version of the course, and any time exceeding the maximum will not be accumulated.

[0040] This can prevent users from maliciously accumulating study time.

[0041] It should be noted that this application, through intelligent recognition and behavior analysis algorithms, can effectively filter out invalid viewing behaviors and accurately record viewing periods of genuine participation, active interaction, and focused attention. This enables high-precision statistics on effective viewing time, which not only improves the authenticity and credibility of learning data but also provides solid data support for subsequent learning effect evaluation, course optimization, and personalized recommendations.

[0042] In some specific implementations, user viewing behavior data for the target video course is collected in real time, including: In response to detecting that a user actively closes the video playback interface corresponding to the target video course, the end viewing time is recorded; and / or, In response to the detection that the duration of pausing playback of the target video course exceeds the preset pause duration, the end viewing time is recorded; and / or, In response to detecting a switch to another video course, record the end viewing time; and / or, In response to the detection of background playback and the end of the current video playback, the end viewing time is recorded.

[0043] Specifically, the system will collect the end-of-view time when it detects that the video playback page has been actively closed; or, it will collect the end-of-view time when it detects that playback has been paused for more than a preset duration (which can be set by the platform, with a default of 30 minutes); or, it will collect the end-of-view time when it detects that the user has switched to another course; or, it will collect the end-of-view time when the current video finishes playing after the user has switched to background playback.

[0044] In this way, effective learning periods can be accurately identified, and interference from invalid data can be suppressed.

[0045] In some specific implementations, the initial viewing duration is calibrated based on at least one of the following: course version number, speed adjustment records, learning interaction behavior, and background switching records, to determine the user's single viewing duration for the target video course, including: Based on the course version number, the initial viewing time is calibrated by chapter matching to determine the first invalid viewing time; Based on the speed adjustment records, the initial viewing time was calibrated to determine the second invalid viewing time. Based on learning interaction behavior and background switching records, the initial viewing time is calibrated by background playback to determine the third invalid viewing time. The invalid viewing time is determined based on the first invalid viewing time, the second invalid viewing time, and the third invalid viewing time; The duration of a user's single viewing of a target video course is determined based on the initial viewing duration and the invalid viewing duration.

[0046] The calibration methods may include data mapping, time axis comparison, threshold judgment, persistence analysis, and time monitoring.

[0047] In one embodiment, user attention detection can also be introduced, using biometric data (such as eye tracking and facial expression recognition) to assess the user's attention level and further calibrate viewing time. For example, if a user is not in front of the screen for an extended period, the video should be considered invalid even if it is playing.

[0048] In some specific implementations, chapter matching calibration is performed on the initial viewing time based on the course version number to determine the first invalid viewing time, including: Based on the course version number, determine the chapter division information of the target video course; Based on user viewing behavior data of the target video courses, determine the chapter coverage information; Based on chapter coverage and chapter division information, the initial viewing time is calibrated by chapter matching to determine the first invalid viewing time.

[0049] The chapter division information may include the duration of each chapter and the total number of chapters.

[0050] Specifically, when a user's viewing time for Chapter 1 exceeds the length of Chapter 1, the length of Chapter 1 will be used as the standard; when a user's viewing time for Chapter 1 is equal to the length of Chapter 1, the user's viewing time will be used as the standard.

[0051] Specifically, obtain the chapter division information of the target video course. If a user only watches a portion of the chapters at a time, only the actual duration of the watched chapters is counted. For example, if a user watches the first chapter of the target video course for 20 minutes and the second chapter for 10 minutes, the time before calibration is 30 minutes and the time after calibration is still 30 minutes. If a user watches the first 5 minutes of the first chapter and does not enter the next chapter, 5 minutes are counted.

[0052] In this way, instead of relying solely on the original playback duration, the system combines chapter boundary information to dynamically determine the user's actual viewing progress within each chapter, avoiding data distortion caused by misjudgments across chapters.

[0053] In some specific implementations, based on the speed adjustment record, the initial viewing duration is calibrated by a multiplier to determine the second invalid viewing duration, including: Based on the speed adjustment records, obtain at least one viewing speed for the user and the corresponding speed duration; In response to detecting that the user's viewing speed is greater than the preset speed, obtain the duration of the speed corresponding to the viewing speed. If the duration of the viewing speed is greater than the preset duration, the duration of the viewing speed that is greater than the preset duration will be used as the second invalid viewing duration.

[0054] The preset speed can be 2x, and the preset duration can be 10 minutes.

[0055] Specifically, if a user is detected playing content at a speed greater than 2x and for a duration exceeding ten minutes, only the portion at a speed less than or equal to 2x will be counted; the portion exceeding 2x will be considered invalid.

[0056] In this way, the effectiveness of learning and content absorption are guaranteed, and the fairness of class time is maintained.

[0057] In some specific implementations, based on learning interaction behavior and background switching records, the initial viewing time is calibrated using background playback to determine the third invalid viewing time, including: In response to the detection that the user has switched to background playback, the system detects learning interaction behavior and records the duration of no interaction. If the duration of no interaction exceeds the preset duration of no interaction, the portion of the no-interaction duration exceeding the preset duration of no interaction is taken as the third invalid viewing duration.

[0058] The preset non-interaction time can be 15 minutes.

[0059] Specifically, if it is detected that a user switches to background playback and has no interaction for more than 15 minutes, only the first 15 minutes of viewing time will be counted, and the excess time will be considered invalid.

[0060] In this way, the user's genuine fragmented learning outcomes are protected, while cheating behavior of accumulating time by leaving the system running in the background for extended periods is effectively prevented, thus suppressing invalid time-farming behavior and accurately identifying the user's true learning status.

[0061] In some specific implementations, the total viewing time of the target video course is determined by summing up at least one single viewing duration of the user, including: Obtain the course version number and chapter division information viewed by the user; If a user has watched the same version of the target video course at least once, the chapters of the target video course are deduplicated and summed to determine the total viewing time. In response to the fact that a user has watched different versions of the target video course at least once, the newly added chapters of the target video courses in different versions are compared and the newly added chapters are fully added. The common chapters of the target video courses in different versions are deduplicated and accumulated.

[0062] Specifically, based on the course version number, it is determined whether the versions watched multiple times are the same. When a user watches the same version of the course multiple times, the viewing time of each chapter is accumulated each time. When a user watches different versions of the course multiple times, the versions are first distinguished based on the course version number, and then the viewing time of each new chapter in each version is fully accumulated. For chapters that are repeated with the previous version, the calculation is performed according to the chapter deduplication and accumulation logic.

[0063] For example, when a user watches the same version of the course multiple times, if the user watches Chapter 1 of the first version of the course for 20 minutes the first time (total time 30 minutes), and watches the same chapter a second time, only the remaining 10 minutes are added; if the user watches the same chapter for only 5 minutes the second time, then the total time for that chapter is 25 minutes.

[0064] This avoids duplicate timing and ensures the uniqueness and fairness of study hours.

[0065] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0066] In one embodiment, such as Figure 2 As shown, a video course viewing duration statistics device is provided. The device includes: a first processing module 201, used to determine the target video course bound to the user identifier based on the user identifier; a second processing module 202, used to collect user viewing behavior data of the target video course in real time, wherein the viewing behavior data includes at least one of course identifier, course version number, speed adjustment record, learning interaction behavior, background switching record, start viewing time and end viewing time; a third processing module 203, used to obtain the user's initial viewing duration of the target video course based on the start viewing time and end viewing time; a fourth processing module 204, used to calibrate the initial viewing duration based on at least one of course version number, speed adjustment record, learning interaction behavior and background switching record to determine the user's single viewing duration of the target video course; and a fifth processing module 205, used to accumulate at least one single viewing duration of the user's target video course to determine the user's total viewing duration of the target video course.

[0067] In a preferred embodiment of this application, the second processing module 202 is specifically configured to: record the end-of-view time in response to detecting that the user actively closes the video playback interface corresponding to the target video course; and / or, record the end-of-view time in response to detecting that the duration of pausing the playback of the target video course exceeds a preset pause duration; and / or, record the end-of-view time in response to detecting switching to other video courses; and / or, record the end-of-view time in response to detecting background playback and the current video playback ending.

[0068] In a preferred embodiment of this application, the fourth processing module 204 is specifically used to: perform chapter matching calibration on the initial viewing time according to the course version number to determine the first invalid viewing time; perform multiplier calibration on the initial viewing time according to the speed adjustment record to determine the second invalid viewing time; perform background playback calibration on the initial viewing time according to the learning interaction behavior and background switching record to determine the third invalid viewing time; determine the invalid viewing time based on the first invalid viewing time, the second invalid viewing time, and the third invalid viewing time; and determine the user's single viewing time for the target video course based on the initial viewing time and the invalid viewing time.

[0069] As a preferred implementation, in this embodiment of the application, the fourth processing module 204 is further configured to: determine the chapter division information of the target video course according to the course version number; determine the chapter coverage information according to the user's viewing behavior data of the target video course; and perform chapter matching calibration on the initial viewing time according to the chapter coverage information and the chapter division information to determine the first invalid viewing time.

[0070] In a preferred embodiment of this application, the fourth processing module 204 is further configured to: obtain at least one viewing speed and the corresponding viewing speed duration based on the speed adjustment record; in response to detecting that the user's viewing speed is greater than a preset speed, obtain the corresponding viewing speed duration; and in response to the viewing speed duration being greater than a preset duration, use the viewing speed duration greater than the preset speed as a second invalid viewing duration.

[0071] As a preferred implementation, in this embodiment of the application, the fourth processing module 204 is further configured to: detect learning interaction behavior and record no-interaction duration in response to detecting that the user has switched to background playback; and obtain the portion of the no-interaction duration exceeding the preset no-interaction duration as the third invalid viewing duration in response to the no-interaction duration exceeding the preset no-interaction duration.

[0072] In a preferred implementation, in this embodiment of the application, the fifth processing module 205 is specifically used to: obtain the course version number and chapter division information viewed by the user; in response to the user viewing at least once the same version of the target video course, deduplicating and summing the chapters of the target video course to determine the total viewing time; in response to the user viewing at least once different versions of the target video course, comparing the newly added chapters of the target video courses of different versions, and summing the newly added chapters in full; and deduplicating and summing the common chapters of the target video courses of different versions.

[0073] Specific limitations regarding the video course viewing time statistics device can be found in the limitations of the video course viewing time statistics method described above, and will not be repeated here. Each module in the aforementioned video course viewing time statistics device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0074] At the interest analysis level, existing technologies mainly rely on macro-level indicators such as viewing time, frequency, and viewing ratio for recommendations. The analysis dimensions are too simplistic. Because the model fails to incorporate key behavioral details such as user completion of core course chapters, interaction with exercises, and note-taking, and also lacks comparison with the group learning characteristics of users in the same grade and major, it cannot accurately locate the personalized interest differences of users. Therefore, this analysis method is difficult to support accurate course recommendations and effective content optimization.

[0075] In one embodiment, an interest analysis system is provided, the system comprising: The storage module is used to store at least one of the following: user information, course information, viewing behavior information, single viewing duration data, cumulative viewing duration data, and total learning duration data, and to establish a multi-dimensional relational index.

[0076] User information may include user ID, name, registration time, etc.; course information may include course ID, course name, category, etc.; and viewing behavior information may include start time, end time, etc.

[0077] Multidimensional related indexes can include single-column indexes, composite indexes, covering indexes, multi-key indexes, clustered indexes, and non-clustered indexes.

[0078] For example, a multi-dimensional association index can be established, such as "User ID - Course ID - Course Version Number - Cumulative Viewing Time", "User ID - Course Category - Cumulative Viewing Time", or "Course ID - Version Number - Core Chapters - User Viewing Completion Rate".

[0079] This facilitates subsequent data querying and retrieval, improving data reading efficiency.

[0080] The interest analysis module is used to analyze user completion rates, interactivity, and group differences based on data stored in the module, in order to determine the courses that users are interested in.

[0081] Among them, user completion rate can be viewed as completion rate, which can be analyzed using two parameters: overall completion rate and core chapter completion rate. These two parameters can be calculated by combining the chapter division information and core chapter markers in the storage module.

[0082] The core chapters can be marked by the platform's teaching and research team, such as the "formula derivation chapter" in the math course and the "grammar analysis chapter" in the English course.

[0083] Specifically, the overall completion rate can be obtained by "cumulative viewing time of the course ÷ total duration of the current version of the course × 100%".

[0084] Specifically, the completion rate of core chapters can be obtained by "cumulative viewing time of core chapters of the course ÷ total time of core chapters of the course × 100%".

[0085] For example, if a course has an "overall completion rate > 80%" and a "core chapter completion rate > 90%", it is judged as a "high-interest course" for the user; if the "overall completion rate < 30%" and the "core chapter completion rate < 20%", it is judged as a "low-interest course".

[0086] Interactivity can be defined as the correlation between interactive behaviors, which can be analyzed through interaction density.

[0087] Specifically, the interaction density can be obtained by "number of interactions ÷ cumulative viewing time".

[0088] For example, for theoretical courses (such as mathematics and physics), if the "interaction density > 0.5 times / hour" (such as clicking on exercise links twice and taking notes once per hour), the course is still judged as a "high-interest course" even if the cumulative time spent on the course is slightly lower; for practical courses (such as programming and design), if the "interaction density > 1.0 times / hour" (such as submitting practical assignments three times and initiating a question consultation once per hour), the course is judged as a "high-interest course".

[0089] Specifically, a user interest-interaction behavior association model can be constructed. For example, users who take "mathematics courses + high interaction with exercises" are labeled as "preferring exercise explanation courses"; users who take "programming courses + high interaction with homework submission" are labeled as "preferring practice-oriented courses".

[0090] Based on the "grade, major, and learning goals" in the user information, users are divided into different learning groups, such as the senior high school math intensive group or the computer science basics learning group.

[0091] Among them, group differences can be compared among groups, and can be analyzed by the benchmark value of the proportion of course classification time.

[0092] Specifically, the baseline value for the percentage of time spent on a course category can be obtained by "the total cumulative time spent on a course category by all users in a certain group ÷ the total learning time of the group × 100%".

[0093] Specifically, a baseline value for the percentage of time spent on a course category is calculated for each group. This is compared to the baseline value for an individual user and the group. If a user's percentage of time spent on a certain course category is more than 20% higher than the group's baseline value, the user is judged to have "high personal preference" for that category. If it is more than 30% lower than the group's baseline value, the user is judged to have "low personal preference".

[0094] For example, the "benchmark value of time spent on math courses" for the "senior high school math intensive group" is 40%. If user A's time spent on math courses accounts for 65%, then user A is judged to have a "personalized high preference" for math courses.

[0095] The display module is used to visually showcase at least one of the following: user-interested courses, learning alerts, course recommendations, and course optimization feedback.

[0096] Data visualization can present data from different dimensions to users and platform administrators in the form of tables and charts (bar charts, pie charts, line charts).

[0097] Specifically, what can be displayed to users may include: total personal learning time, cumulative viewing time and completion rate of each course, top 3 course categories based on personal interests, and a comparison of learning time with other users in the same group.

[0098] Specifically, what can be shown to the administrator may include: the overall course viewing time distribution on the platform, the user completion ranking of each course, the interest and preference distribution of each learning group, and a list of courses with high or low interest.

[0099] Among them, the learning warning can set warning rules and trigger corresponding interventions based on the cumulative viewing time and interest analysis results in the storage module.

[0100] For example, if a user's cumulative viewing time for a "high-interest course" is less than 30 minutes over 7 consecutive days and they do not participate in any interaction with the course, the system will automatically push a personalized reminder to the user (such as "You have not studied the 'Advanced High School Mathematics Functions' course for 7 days. Do you want me to recommend relevant review chapters for you?"); and if a user's "overall completion rate" for a "required course" (marked by the platform) is less than 50% and there are less than 10 days left before the course ends, the system will push a warning message to the administrator, who will then arrange for a tutor to provide targeted intervention (such as one-on-one Q&A and study plan development).

[0101] Among them, the course recommendation can push personalized courses to users based on the results of the "interest-interaction behavior association model" and "group comparison analysis" in the interest analysis module.

[0102] For example, math courses with a lot of problem-solving explanations can be recommended to "math users who prefer problem-solving explanations", or advanced physics courses and the top 5 physics courses with the highest user satisfaction ratings in the same group can be recommended to "personalized physics courses with a high preference".

[0103] Among these features, the course optimization feedback can transform the results of interest analysis into course optimization suggestions, which are then fed back to the platform's teaching and research team.

[0104] For example, for courses with "overall completion rate < 40%" and "core chapter completion rate < 30%", the low-completion chapters can be analyzed, and the teaching and research team can be advised to optimize the chapter content (such as simplifying the explanation of complex knowledge points and adding case demonstrations); or, for courses with "interaction density < 0.3 times / hour", it is recommended to add interactive elements (such as inserting in-class quizzes and setting up a course discussion area); or, the common characteristics of "high-interest courses" (such as chapter duration distribution and interactive element design) can be statistically analyzed to provide design references for new courses.

[0105] Thus, unlike related technologies that only analyze duration and frequency, this approach constructs an analysis system from three dimensions: viewing completion rate, interactive behavior, and group comparison. By combining the types of online education courses (theoretical / practical) with group characteristics, the accuracy of interest positioning is improved, providing a more reliable basis for personalized recommendations. At the same time, new learning alerts and course optimization feedback functions are added, forming a closed loop of "data statistics - interest analysis - teaching intervention - course iteration". This not only meets users' personalized learning needs but also provides data support for the platform's course operation and content optimization, solving the problem that related technologies only focus on "statistics and recommendations" and lack teaching assistance capabilities.

[0106] Example 1 Taking a certain high school online education platform as an example, the platform offers 12 video courses covering three categories: mathematics, English, and physics. The course information is shown in the table below (partial list of courses):

[0107] Step 1: Student registration and course information initialization.

[0108] Student B registers an account through the student information registration module, filling in personal information (grade: senior high school, major: science, learning goal: college entrance examination preparation), and the system generates a student ID of S002. The course information storage module pre-stores basic information of the above 12 courses, including course version number, core chapters, etc., and establishes association relationships.

[0109] Step 2: Observe the behavior data collection.

[0110] Student B watched course C001 (Advanced High School Mathematics Functions, Version V1.0) on the APP. The viewing behavior collection module collected the following data: First viewing: Start time 2025-09-01 08:30, end time 2025-09-01 09:40, speed 1.5x (full video), during which I clicked on the exercise link 3 times and took notes once, without switching background processes; Second viewing (Course C001 updated to V1.1, with the addition of Chapter 6 "Function Extension", duration 20 minutes): Start time 2025-09-03 14:00, end time 2025-09-03 14:50, speed 2.0x (first 30 minutes), 1.2x (last 20 minutes), during which the video was switched to the background for 18 minutes (no interaction in the background), and the exercise link was clicked twice.

[0111] Step 3: Duration calculation.

[0112] The duration calculation module combines the collected data with course information to calculate the duration: First viewing (C001V1.0): Initial duration: 70 minutes (08:30-09:40), speed 1.5x (≤2.0x, no calibration required), no invalid behavior, valid single duration 70 minutes; Chapter coverage: Watch Chapter 1 (20 minutes), Chapter 2 (30 minutes), and Chapter 3 (20 minutes), for a total viewing time of 70 minutes (C001V1.0 has a total duration of 90 minutes, with 20 minutes remaining).

[0113] Second viewing (C001V1.1): Initial duration: 50 minutes (14:00-14:50), 18 minutes of no background interaction (if it exceeds 15 minutes, only the first 15 minutes of valid background time will be counted), speedup 2.0x (≤2.0x, no calibration required). Valid single session duration: (50-18)+15=47 minutes; Chapter Coverage: Watching the old chapter 4 (25 minutes, which is the unwatched part of C001V1.0, fully added), the newly added chapter 6 (20 minutes, fully added), and the old chapter 5 (2 minutes, not finished, added 2 minutes). Total viewing time: 70 (first time) + 25 + 20 + 2 = 117 minutes (C001V1.1 total time is 110 minutes, 117 ≤ 110 × 1.5 = 165, no upper limit calibration required).

[0114] If student B also watched course C003 (Physics, V1.0), with a total viewing time of 60 minutes, and watched the other 9 courses for a total of 420 minutes, then student B's total study time is 117 + 60 + 420 = 597 minutes.

[0115] Step 4: Data storage.

[0116] The data storage module stores student B's ID (S002), personal information, viewing behavior data for courses such as C001 / C003, effective single session duration, cumulative duration, and total learning time, and establishes related indexes such as "S002-C001-V1.1-117 minutes" and "S002-Mathematics-117 minutes".

[0117] Step 5: Interest Analysis.

[0118] The interest analysis module calls up data for analysis: Completion analysis: C001 (V1.1) overall completion rate = 117 ÷ 110 × 100% ≈ 106% (calculated as 100%), core chapter completion rate = (30 + 25) ÷ (30 + 25) × 100% = 100%, judged as "high interest course"; C003 (V1.0) overall completion rate = 60 ÷ 80 × 100% = 75%, core chapter completion rate = (25 + 20) ÷ (25 + 20) × 100% = 100%, judged as "medium to high interest course".

[0119] Interactive behavior correlation analysis: The number of interactions for C001 = 3 + 2 = 5 times, with a total duration of 117 minutes. The interaction density = 5 ÷ (117 ÷ 60) ≈ 2.56 times / hour (> 0.5 times / hour). Combined with the attributes of mathematics theory courses, C001 is further confirmed as a "high-interest course".

[0120] Group Comparison Analysis: In the "Senior High School Science Entrance Examination Preparation Group" to which Student B belongs, the baseline value for the time spent on mathematics courses is 35%. Student B's time spent on mathematics courses is approximately 19.6% (117 ÷ 597 × 100%) (lower than the baseline value of 35%, but still considered high interest based on interaction density). The baseline value for physics courses is 25%. Student B's time spent on physics courses is approximately 10.1% (more than 30% lower than the baseline value, considered "low personalized preference").

[0121] Step 6: Results Display and Application Visual presentation: Show student B the total learning time of 597 minutes, completion rate of C001 of 100%, and the top 3 courses of interest (C001, C003, and English C002); show the administrator the statistical results that "the participation rate of senior high school science students in physics courses is low".

[0122] Learning alert: Student B's cumulative time spent on physics courses (C003) is more than 30% lower than the group benchmark. The system sends an alert to the administrator and suggests that the physics teacher communicate with Student B.

[0123] Course Recommendation: Based on the association model of "mathematics + interactive exercises", we recommend courses such as "In-depth Explanation of High School Mathematics Function Exercises" and "Comprehensive Problem-Solving Techniques for Senior High School Mathematics" to student B.

[0124] Course optimization feedback: The platform statistics show that the interaction density of C001 (V1.1) is 2.56 times / hour, and its characteristics of "clear chapter division and sufficient interactive exercises" are used as a design reference for new math courses; for an English course with "overall completion rate of 35%", it is recommended to optimize the way the listening chapter is explained.

[0125] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 3 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When executed by the processor, the computer program implements a method for calculating video course viewing time. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0126] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0127] Embodiments of this application also provide a computer-readable storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above-described video course viewing duration statistics method embodiments when running.

[0128] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard disk, magnetic disk, or optical disk.

[0129] Embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above-described video course viewing time statistics method embodiments.

[0130] Embodiments of this application also provide another computer program product, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above-described video course viewing duration statistics method embodiments.

[0131] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0132] The foregoing has provided a detailed description of a video course viewing time statistics method, apparatus, table parsing system, and computer equipment provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only intended to aid in understanding the method and core ideas of this application. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this application.

Claims

1. A method for video course watch time statistics, characterized in that, The method comprises: According to the user identifier, determine the target video course bound to the user identifier; Collecting user viewing behavior data of the target video course, wherein the viewing behavior data comprises at least one of course identifier, course version number, speed adjustment record, learning interaction behavior, background switching record, start watching time and end watching time; According to the start watching time and the end watching time, the initial watching time of the user for the target video course is obtained; According to at least one of the course version number, the speed adjustment record, the learning interaction behavior and the background switching record, the initial watching time is calibrated to determine the single watching time of the user for the target video course; Accumulate at least one single watching time of the user for the target video course to determine the total watching time of the user for the target video course.

2. The video lesson watch time statistics method of claim 1, wherein, The real-time collection of user viewing behavior data of the target video course comprises: In response to monitoring that the user actively closes the video playing interface corresponding to the target video course, the end watching time is recorded; and / or, In response to monitoring that the duration of pausing playing the target video course exceeds the preset pause duration, the end watching time is recorded; and / or, In response to monitoring that switching to other video courses, the end watching time is recorded; and / or, In response to monitoring that the background playing, and the current video playing ends, the end watching time is recorded.

3. The video course watch time statistics method of claim 1, wherein, The calibration of the initial watching time according to at least one of the course version number, the speed adjustment record, the learning interaction behavior and the background switching record to determine the single watching time of the user for the target video course comprises: According to the course version number, the initial watching time is chapter matching calibrated to determine the first invalid watching time; According to the speed adjustment record, the initial watching time is multiple calibrated to determine the second invalid watching time; According to the learning interaction behavior and the background switching record, the initial watching time is background hanging broadcast calibrated to determine the third invalid watching time; According to the first invalid watching time, the second invalid watching time and the third invalid watching time, the invalid watching time is determined; According to the initial watching time and the invalid watching time, the single watching time of the user for the target video course is determined.

4. The video lesson watch time statistics method of claim 3, wherein, The chapter matching calibration of the initial watching time according to the course version number to determine the first invalid watching time comprises: According to the course version number, the chapter division information of the target video course is determined; According to the user viewing behavior data of the target video course, the chapter coverage information is determined; According to the chapter coverage information and the chapter division information, the initial watching time is chapter matching calibrated to determine the first invalid watching time.

5. The video lesson watch time statistics method of claim 3, wherein, The multiple calibration of the initial watching time according to the speed adjustment record to determine the second invalid watching time comprises: According to the speed adjustment record, at least one watching speed of the user and the speed duration corresponding to the watching speed are obtained; In response to detecting that the viewing speed of the user is greater than a preset speed, a speed duration corresponding to the viewing speed is obtained; In response to the speed duration corresponding to the viewing speed being greater than a preset duration, the speed duration in which the viewing speed is greater than the preset speed is taken as a second invalid viewing duration.

6. The video lesson watch time statistics method of claim 3, wherein, The method further includes: In response to detecting that the user switches to background playing, the learning interaction behavior is detected, and a non-interaction duration is recorded; In response to the non-interaction duration being greater than a preset non-interaction duration, a part of the non-interaction duration that exceeds the preset non-interaction duration is taken as a third invalid viewing duration.

7. The video lesson watch time statistics method of claim 1, wherein, The method further includes: The course version number and chapter division information watched by the user are obtained; In response to the target video courses watched by the user at least once being of the same version, chapters of the target video courses are de-duplicated and accumulated to determine the total viewing duration; In response to the target video courses watched by the user at least once being of different versions, newly added chapters of the different versions of the target video courses are compared, and the newly added chapters are fully accumulated; Common chapters of the different versions of the target video courses are de-duplicated and accumulated.

8. A video course watch time statistics apparatus, characterized in that, The device includes: A first processing module configured to determine, according to a user identifier, a target video course bound to the user identifier; A second processing module configured to collect, in real time, viewing behavior data of the user for the target video course, wherein the viewing behavior data includes at least one of a course identifier, a course version number, speed adjustment records, learning interaction behavior, background switching records, a start viewing time, and an end viewing time; A third processing module configured to obtain, according to the start viewing time and the end viewing time, an initial viewing duration of the user for the target video course; A fourth processing module configured to calibrate, according to at least one of the course version number, the speed adjustment records, the learning interaction behavior, and the background switching records, the initial viewing duration to determine a single viewing duration of the user for the target video course; A fifth processing module configured to accumulate at least one single viewing duration of the user for the target video course to determine a total viewing duration of the user for the target video course.

9. An interest analysis system characterized by, The system includes: A storage module configured to store at least one of user information, course information, viewing behavior information, single viewing duration data, cumulative viewing duration data, and total learning duration data and establish a multi-dimensional association index; An interest analysis module configured to analyze, according to data in the storage module, user completion, interactivity, and group differences to determine a user interest course; A display module configured to visually display at least one of a user interest course, a learning warning, a course recommendation, and course optimization feedback.

10. A computer device, comprising: The device includes: A memory configured to store a computer program; a processor for executing the computer program to implement the steps of the method according to any one of claims 1 to 7.

Images