A Course Recommendation Method and System Based on Big Data

Abstract

The present invention relates to the technical field of data recommendation, and specifically discloses a course recommendation method and system based on big data. The method includes obtaining the behavior status data of a target user, and acquiring the behavior status values of the same-category courses of the target user according to the behavior status data of the target user; classifying the same-category course groups based on the behavior status values of the same-category courses of the target user to obtain classification data, and uploading the classification data to a cloud control platform; obtaining the course energy efficiency data of the target course subunit in the highly concerned same-category course group, and obtaining the recommendation status value of the target course subunit based on the course energy efficiency data of the target course subunit, thereby completing the recommendation of the target course subunit. The present invention uses the target course subunit corresponding to the maximum recommendation status value in the highly concerned same-category course group as the optimal recommended course and pushes it to the target user, with a high degree of visualization.

Description

Technical Field

[0001] The present invention relates to the field of data recommendation technology, and in particular to a course recommendation method and system based on big data. Background Art

[0002] In a narrow sense, education refers to specially organized schooling; in a broader sense, it refers to social activities that influence a person's physical and mental development. The term "education" originates from Mencius's saying, "Educate the world's best talents." With the rapid development of society, today's education is no longer limited to the classroom; online education has also emerged.

[0003] In the process of promoting digital education, online education is becoming more and more popular. The existing Internet education lacks interaction between target users and the required course categories, resulting in mismatched course recommendation information for students on the Internet education side. It is unable to match target users with courses suitable for themselves, and has certain limitations. Summary of the Invention

[0004] The purpose of the present invention is to provide a course recommendation method and system based on big data. By processing the target course sub-units in the highly concerned course group of the same category, the number recommendation ratio and the number improvement ratio of the target course sub-unit are obtained. By processing the number recommendation ratio and the number improvement ratio of the target course sub-unit, the recommendation status value of the target course sub-unit is obtained. The target course sub-unit corresponding to the largest recommendation status value in the highly concerned course group of the same category is used as the optimal recommended course and pushed to the target user with a high degree of visualization.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A course recommendation method based on big data includes the following steps:

[0007] Obtain the target user's behavior status data, and obtain the target user's behavior status value for the same category of courses based on the target user's behavior status data;

[0008] Based on the behavioral status values of target users in the same category of courses, the same category course groups are classified into different levels to obtain classification data, and then uploaded to the cloud management and control platform;

[0009] It includes non-attention similar course groups, medium-attention similar course groups and high-attention similar course groups;

[0010] Obtain the course energy efficiency data of the target course sub-unit in the highly-watched course group of the same category, obtain the recommended status value of the target course sub-unit based on the course energy efficiency data of the target course sub-unit, and complete the recommendation of the target course sub-unit.

[0011] As a further solution of the present invention: The behavior state data of the target user includes the ratio of the duration of the same-category courses, the ratio of the number of the same-category courses, and the ratio of the unviewed same-category courses;

[0012] The ratio of the duration of the same-category courses is the ratio of the total duration of the same-category courses to the duration of the browsing period;

[0013] The ratio of the number of the same-category courses is the ratio of the total number of views of the same-category courses to the duration of the browsing period;

[0014] The ratio of the unviewed same-category courses is the ratio of the unviewed duration interval of the same-category courses to the duration of the browsing period.

[0015] As a further solution of the present invention: The process of obtaining the behavior state value of the target user's same-category courses is as follows:

[0016] Record the ratio of the duration of the same-category courses as Tk;

[0017] Record the ratio of the number of the same-category courses as Ts;

[0018] Record the ratio of the unviewed same-category courses as Tw;

[0019] Calculate to obtain the behavior state value Ti of the target user's same-category courses.

[0020] As a further solution of the present invention: The limit values of the preset behavior state value thresholds of the same-category courses are Ti1 and Ti2, where Ti1 < Ti2;

[0021] When Ti < Ti1, it indicates that the target user has low attention to the same-category courses, and all the same-category courses within this range are recorded as the non-attention same-category course group;

[0022] When Ti1 ≤ Ti < Ti2, it indicates that the target user has medium attention to the same-category courses, and all the same-category courses within this range are recorded as the medium-attention same-category course group;

[0023] When Ti ≥ Ti2, it indicates that the target user has high attention to the same-category courses, and the same-category courses within this range are recorded as the high-attention same-category course group.

[0024] As a further solution of the present invention: The course energy efficiency data includes the recommended ratio of the number of people in the target course subunit and the increased ratio of the number of people in the target course subunit.

[0025] As a further solution of the present invention: Record the recommended ratio of the number of people in the target course subunit as Rt;

[0026] Record the increased ratio of the number of people in the target course subunit as Rs;

[0027] Calculate the recommended status value Ri of the target course subunit.

[0028] As a further solution of the present invention: The target course subunit corresponding to the maximum recommended status value within the high - attention same - category course group is used as the optimal recommended course and pushed to the target user.

[0029] The course recommendation priority is to push to the target user in the order of the recommended status value of the target course subunit from large to small.

[0030] As a further solution of the present invention: Obtain the historical learning number of the target course subunit and the development duration of the target course subunit, calculate the ratio of the historical learning number of the target course subunit to the development duration of the target course subunit to obtain the number recommendation ratio of the target course subunit.

[0031] As a further solution of the present invention: Obtain the ratio of the number of people making progress after each course completion of the target course subunit, calculate the ratio of the ratio of the number of people making progress to the total number of learners to obtain the number increase rate of a single - time target course subunit, and sum and average the number increase rates of all times of the target course subunit within the development duration of the target course subunit to obtain the number increase ratio of the target course subunit.

[0032] As a further solution of the present invention: A course recommendation system based on big data includes:

[0033] The data acquisition module is used to obtain the behavior status data of the target user, obtain the behavior status value of the same - category courses of the target user according to the behavior status data of the target user, and upload it to the cloud control platform.

[0034] The category division module receives the behavior status value of the same - category courses transmitted by the cloud control platform. The category division module classifies the same - category course group based on the behavior status value of the same - category courses of the target user to obtain classification data, and uploads it to the cloud control platform.

[0035] The classification data includes a non - attention same - category course group, a medium - attention same - category course group, and a high - attention same - category course group.

[0036] The recommendation analysis module receives the classification data transmitted by the cloud control platform. The recommendation analysis module obtains the course energy efficiency data of the target course subunit within the high - attention same - category course group, and obtains the recommended status value of the target course subunit based on the course energy efficiency data of the target course subunit to complete the recommendation of the target course subunit.

[0037] The beneficial effects of the present invention:

[0038] (1) According to the browsing behavior of the target user for courses of the same category, the present invention obtains the ratio of the time of courses of the same category, the ratio of the number of courses of the same category, and the ratio of unviewed courses of the same category when the target user browses courses of the same category, and obtains the behavior status value of the target user's courses of the same category according to the ratio of the time of courses of the same category, the ratio of the number of courses of the same category, and the ratio of unviewed courses of the same category, that is, the behavior status value of the target user's courses of the same category is obtained from dimensions such as the total duration, the total number of browsing times, and the duration of unviewed intervals of the target user for courses of the same category, making the recognition of the target user's courses of the same category more accurate;

[0039] (2) By processing the target course sub-units within the group of courses of the same category with high attention, the present invention obtains the recommended ratio of the number of people for the target course sub-units and the increased ratio of the number of people for the target course sub-units, and obtains the recommended status value of the target course sub-units by processing the recommended ratio of the number of people for the target course sub-units and the increased ratio of the number of people for the target course sub-units. The target course sub-unit corresponding to the largest recommended status value within the group of courses of the same category with high attention is used as the optimal recommended course and pushed to the target user, with a high degree of visualization. Description of the Drawings

[0040] The present invention will be further described below in conjunction with the drawings.

[0041] Figure 1 is a flowchart of a course recommendation method based on big data according to an embodiment of the present invention;

[0042] Figure 2 is a flowchart for identifying a group of courses of the same category in a course recommendation method based on big data according to an embodiment of the present invention;

[0043] Figure 3 is a block diagram of a course recommendation system based on big data according to an embodiment of the present invention. Detailed Embodiments

[0044] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0045] Embodiment 1

[0046] Please refer to Figure 1 As shown, the present invention is a course recommendation method based on big data, including the following steps:

[0047] Obtain the behavior status data of the target user, and obtain the behavior status value of the target user's courses of the same category according to the behavior status data of the target user;

[0048] Classify the same-category course groups based on the behavioral status values of the same-category courses of the target user to obtain classification data, and upload it to the cloud control platform;

[0049] It includes the non-concerned same-category course group, the medium-concerned same-category course group, and the highly-concerned same-category course group;

[0050] Obtain the course energy efficiency data of the target course subunit within the highly-concerned same-category course group, and obtain the recommended status value of the target course subunit based on the course energy efficiency data of the target course subunit to complete the recommendation of the target course subunit.

[0051] Embodiment 2

[0052] The behavioral status data of the target user includes the same-category course time ratio, the same-category course number ratio, and the same-category unviewed ratio;

[0053] Obtain the total duration of the target user browsing the same-category courses within the browsing period, calculate the ratio of the total duration of the same-category courses to the duration of the browsing period to obtain the same-category course time ratio;

[0054] Among them, there are various classifications of course categories. For example, classified by subject: mathematics, physics, Chinese, biology, etc.;

[0055] The course category can also be classified according to different contents under the same subject. For example, algebra, geometry, calculus, etc. under the mathematics subject;

[0056] Among them, the process of obtaining the total duration of the same-category courses is as follows:

[0057] Obtain different course subunits under the same-category courses, denoted as target course subunits;

[0058] Obtain the browsing duration of the target user for different target course subunits within the browsing period, and sum up the browsing durations of all target course subunits to obtain the total duration of the same-category courses.

[0059] Obtain the total number of times the target user browses the same-category courses within the browsing period, calculate the ratio of the total number of times of browsing the same-category courses to the duration of the browsing period to obtain the same-category course number ratio;

[0060] Among them, the process of obtaining the total number of times of browsing the same-category courses is as follows:

[0061] Obtain the number of times the target user browses different target course subunits within the browsing period, and sum up the number of times of browsing all target course subunits to obtain the total number of times of browsing the same-category courses.

[0062] Obtain the interval non-browsing duration of the target user browsing courses of the same category within the browsing period, calculate the ratio of the interval non-browsing duration of the courses of the same category to the duration of the browsing period to obtain the non-browsing ratio of the same category;

[0063] Among them, the process of obtaining the browsing interval ratio of courses of the same category is as follows:

[0064] Obtain the time length between the last browsing time point of the courses of the same category by the target user within the browsing period and the end time point of the browsing period to obtain the interval non-browsing duration.

[0065] Record the ratio of courses of the same category as Tk;

[0066] Record the number ratio of courses of the same category as Ts;

[0067] Record the non-browsing ratio of the same category as Tw;

[0068] Through the formula Calculate the behavior status value Ti of the courses of the same category of the target user. Among them, d1 and d2 are both preset proportionality coefficients, and both d1 and d2 are greater than zero.

[0069] Among them, the browsing period is set by the staff according to experience, and the browsing period can be 10 days, 30 days or 60 days.

[0070] Sort the behavior status values Ti of different courses of the same category of the target user within the browsing period. The specific process is as follows:

[0071] Preset the limit values of the behavior status value thresholds of the courses of the same category as Ti1 and Ti2, where Ti1 < Ti2;

[0072] Among them, the limit values Ti1 and Ti2 of the behavior status value thresholds of the courses of the same category are an empirical value, obtained according to experience:

[0073] In the actual obtaining process, there are many groups of ratios of courses of the same category, number ratios of courses of the same category and non-browsing ratios of the same category. Process many groups of ratios of courses of the same category, number ratios of courses of the same category and non-browsing ratios of the same category to obtain the corresponding group of behavior status values of the courses of the same category. The staff identify the level of the courses of the same category according to so many groups of behavior status values of the courses of the same category, so as to obtain a correspondence between the behavior status value of the courses of the same category and the status level of the courses of the same category. Furthermore, deduce and divide the behavior status value threshold of the courses of the same category according to the status of the courses of the same category corresponding to the behavior status value of the courses of the same category, so as to obtain the limit values Ti1 and Ti2 of the behavior status value threshold of the courses of the same category. By comparing the limit values of the behavior status value threshold of the courses of the same category, the identification of the status level of the courses of the same category corresponding to the behavior status value of the courses of the same category is completed;

[0074] When Ti < Ti1, it indicates that the target user has a low attention to courses of the same category. All courses of the same category within this range are recorded as the non - attention same - category course group;

[0075] When Ti1 ≤ Ti < Ti2, it indicates that the target user has a medium attention to courses of the same category. All courses of the same category within this range are recorded as the medium - attention same - category course group;

[0076] When Ti ≥ Ti2, it indicates that the target user has a high attention to courses of the same category. Courses of the same category within this range are recorded as the high - attention same - category course group.

[0077] Embodiment 3

[0078] The target course sub - units within the high - attention same - category course group are the courses that the target user is interested in during the browsing period;

[0079] Obtain the course energy efficiency data of the target course sub - units within the high - attention same - category course group;

[0080] The course energy efficiency data includes the number recommendation ratio of the target course sub - unit and the number increase ratio of the target course sub - unit;

[0081] Obtain the historical learning number of the target course sub - unit and the development duration of the target course sub - unit. Calculate the ratio of the historical learning number of the target course sub - unit to the development duration of the target course sub - unit to obtain the number recommendation ratio of the target course sub - unit;

[0082] Among them, the historical learning number of the target course sub - unit is the total learning number of the target course sub - unit from its establishment to the present;

[0083] Among them, the development duration of the target course sub - unit is the difference between the establishment time of the target course sub - unit and the current time.

[0084] Obtain the ratio of the number of people making progress after each completion of the target course sub - unit. Calculate the ratio of the ratio of the number of people making progress to the total learning number to obtain the number increase rate of a single target course sub - unit. Sum up and take the average of the number increase rates of all times of the target course sub - unit within the development duration of the target course sub - unit to obtain the number increase ratio of the target course sub - unit;

[0085] Among them, the process of obtaining the number increase rate of the target course sub - unit is as follows:

[0086] Based on the target course sub - unit, obtain the first - exam score of the target user when taking the target course sub - unit, denoted as the reference score E0;

[0087] Obtain the exam scores of the target user in a number of target course sub-units, and traverse all the exam scores to obtain the highest exam score Emax and the lowest exam score Emin;

[0088] When the highest exam score Emax ≤ C0, it is determined that the target user has achieved non-promotional progress;

[0089] When the lowest exam score Emin > C0, it is determined that the target user has achieved improvement progress;

[0090] When Emax > C0 and Emin ≤ C0, obtain the number of times the exam score is greater than the reference score, denoted as the number of times exceeding the parameter;

[0091] If the number of times exceeding the parameter is greater than half of the number of exams, it is determined that the target user has achieved improvement progress, otherwise, it is determined that the target user has achieved non-improvement progress;

[0092] Obtain the number of people who have achieved improvement progress when a single target course sub-unit is completed, and calculate the ratio of the number of people with promotional progress to the total number of learners to obtain the population promotion rate of the target course sub-unit.

[0093] Denote the population recommendation ratio of the target course sub-unit as Rt;

[0094] Denote the population promotion ratio of the target course sub-unit as Rs;

[0095] Through the formula Calculate the recommendation status value Ri of the target course sub-unit, where k is a preset proportionality coefficient;

[0096] Among them, there are m groups of historical data, and each group of historical data includes the population recommendation ratio Rt of the target course sub-unit, the population promotion ratio Rs of the target course sub-unit, and the recommendation status value Ri of the target course sub-unit;

[0097] Use a linear model to fit the m groups of historical data, and substitute the prepared historical data into the selected fitting model for fitting, and obtain the average value of the fitting coefficients as the preset proportionality coefficient.

[0098] Take the target course sub-unit corresponding to the largest recommendation status value within the high-concern same-category course group as the optimal recommended course and push it to the target user;

[0099] The course recommendation priority is pushed to the target user in the order of the recommendation status value of the target course sub-unit from large to small.

[0100] Embodiment 4

[0101] Please refer to Figure 3 As shown, the present invention is a course recommendation system based on big data, including a data acquisition module, a category division module, a recommendation analysis module, and a cloud control platform;

[0102] The data acquisition module is used to obtain the behavior status data of the target user, obtain the behavior status values of the same-category courses of the target user according to the behavior status data of the target user, and upload them to the cloud control platform;

[0103] The category division module receives the behavior status values of the same-category courses transmitted by the cloud control platform. The category division module classifies the same-category course groups based on the behavior status values of the same-category courses of the target user to obtain classification data, and uploads them to the cloud control platform;

[0104] The classification data includes non-concerned same-category course groups, medium-concerned same-category course groups, and highly-concerned same-category course groups;

[0105] The recommendation analysis module receives the classification data transmitted by the cloud control platform. The recommendation analysis module obtains the course energy efficiency data of the target course subunit within the highly-concerned same-category course group, and obtains the recommendation status value of the target course subunit based on the course energy efficiency data of the target course subunit, and completes the recommendation of the target course subunit.

[0106] The above has described an embodiment of the present invention in detail, but the content described is only the preferred embodiment of the present invention and cannot be considered as used to limit the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the patent of the present invention.

Claims

1. A course recommendation method based on big data, characterized in that, It includes the following steps: Obtain the behavior status data of the target user, and obtain the behavior status value of the same-category courses of the target user according to the behavior status data of the target user; Based on the behavior status value of the same-category courses of the target user, classify the same-category course group into levels to obtain classification data, and upload it to the cloud control platform; The classification data includes the non-concerned same-category course group, the medium-concerned same-category course group, and the highly-concerned same-category course group; Obtain the course energy efficiency data of the target course subunit in the highly-concerned same-category course group, and obtain the recommended status value of the target course subunit based on the course energy efficiency data of the target course subunit to complete the recommendation of the target course subunit; Calculate the behavior status value Ti of the same-category courses of the target user according to the ratio of the same-category courses, the ratio of the number of same-category courses, and the ratio of unviewed same-category courses in the behavior status data of the target user; Preset the extreme values of the behavior status value threshold of the same-category courses as Ti1 and Ti2, where Ti1 < Ti2; When Ti ≥ Ti2, it means that the target user has a high degree of attention to the same-category courses. Mark the same-category courses within this range as the highly-concerned same-category course group; Obtain the course energy efficiency data of the target course subunit in the highly-concerned same-category course group; The course energy efficiency data includes the recommended ratio of the number of people in the target course subunit and the improvement ratio of the number of people in the target course subunit; Obtain the historical number of learners of the target course subunit and the development duration of the target course subunit, and calculate the ratio of the historical number of learners of the target course subunit to the development duration of the target course subunit to obtain the recommended ratio of the number of people in the target course subunit; Obtain the ratio of the number of people making progress after each course of the target course subunit is completed, calculate the ratio of the ratio of the number of people making progress to the total number of learners, obtain the number of people improvement rate of a single target course subunit, and sum and average the number of people improvement rates of all times of the target course subunit within the development duration of the target course subunit to obtain the improvement ratio of the number of people in the target course subunit; Obtain the number of people who have made improvement progress when a single target course subunit is completed, calculate the ratio of the number of people who have made improvement progress to the total number of learners, and obtain the number of people improvement rate of the target course subunit; Record the recommended ratio of the number of people in the target course subunit as Rt; Record the improvement ratio of the number of people in the target course subunit as Rs; Through the formula The recommended status value Ri of the target course subunit is calculated, where k is a preset proportionality coefficient; Among them, there are m groups of historical data, and each group of historical data includes the recommended ratio of the number of people Rt, the improvement ratio of the number of people Rs, and the recommended status value Ri of the target course subunit; Use a linear model to fit the m groups of historical data, and substitute the prepared historical data into the selected fitting model for fitting, and obtain the average value of the fitting coefficients as the preset proportional coefficient; Push the target course subunit corresponding to the largest recommended status value in the highly-concerned same-category course group as the optimal recommended course to the target user; The course recommendation priority is pushed to the target user in the order of the recommended status value of the target course subunit from large to small.

2. The method for course recommendation based on big data according to claim 1, wherein, The behavior status data of the target user includes the ratio of the same-category courses, the ratio of the number of same-category courses, and the ratio of unviewed same-category courses; The ratio of the same-category courses is the ratio of the total duration of the same-category courses to the duration of the browsing cycle; The ratio of the number of same-category courses is the ratio of the total number of views of the same-category courses to the duration of the browsing cycle; The ratio of unviewed same-category courses is the ratio of the duration of unviewed intervals of the same-category courses to the duration of the browsing cycle.

3. A course recommendation method based on big data according to claim 1, characterized in that When Ti < Ti1, it indicates that the target user has low attention to the same-category courses, and all the same-category courses within this range of the same-category courses are recorded as the non-attention same-category course group; When Ti1 ≤ Ti < Ti2, it indicates that the target user has medium attention to the same-category courses, and all the same-category courses within this range of the same-category courses are recorded as the medium-attention same-category course group.

4. A course recommendation system based on big data, which implements the method described in claim 1, characterized in that, Including: The data collection module is used to obtain the behavior status data of the target user, obtain the behavior status value of the same-category courses of the target user according to the behavior status data of the target user, and upload it to the cloud control platform; The category division module receives the behavior status value of the same-category courses transmitted by the cloud control platform. The category division module classifies the same-category course group based on the behavior status value of the same-category courses of the target user to obtain classification data, and uploads it to the cloud control platform; The classification data includes a non-attention same-category course group, a medium-attention same-category course group, and a high-attention same-category course group; The recommendation analysis module receives the classification data transmitted by the cloud control platform. The recommendation analysis module obtains the course energy efficiency data of the target course subunit within the high-attention same-category course group, and obtains the recommendation status value of the target course subunit based on the course energy efficiency data of the target course subunit, and completes the recommendation of the target course subunit.

Images