Online course intelligent recommendation method and system based on big data

Abstract

The invention relates to the technical field of big data intelligent recommendation, in particular to an online course intelligent recommendation method and system based on big data. According to the method, the sensitive sub-features for each user side can be analyzed based on the change of each sub-feature of the behavior tendency feature of the user side, the online course intelligent recommendation efficiency is improved, sensitive tendency categories are distinguished by analyzing the course sensitive tendency, and the online course pushing frequency of each user side is recommended in a customized manner. According to the method, whether the pushing mode is adjusted or not is analyzed, the accessed online course category can be determined by determining the online course page accessed by the user side, and the course pushing frequency for each online course category can be effectively adjusted through each accessed online course category, so that the efficiency of the pushing frequency is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of big data intelligent recommendation, and particularly to an intelligent recommendation method and system for online courses based on big data. Background Art

[0002] With the rapid development of Internet technology, the online education industry has risen rapidly. The flexibility and convenience of online courses have made them favored by the majority of teachers and students. Online education platforms have become an important way for many learners to acquire knowledge. Since the number of courses on online education platforms has increased sharply, covering various fields and disciplines, providing learners with a wide range of choices, it has also led to students facing a vast amount of course information and being difficult to select the most suitable courses for themselves, resulting in the growth of users' personalized needs. Different learners have different learning styles, including visual, auditory, and problem-solving types. In addition, learners' knowledge levels, learning abilities, and learning progress are also uneven. The traditional "one-size-fits-all" teaching mode and course recommendation methods can no longer meet the diverse learning needs of students, and personalized recommendation systems have become increasingly important. The problem of how to customize suitable learning resources for each learner is also becoming more and more urgent to solve.

[0003] Chinese Patent Publication No.: CN115423569B discloses a recommendation method based on big data and a recommendation system based on big data. For the obtained e-commerce business data to be recommended; by determining the e-commerce business data knowledge vectors of no less than two types of the e-commerce business data to be recommended, it is convenient to perform several explorations of the target e-commerce business data in the pre-configured e-commerce business data set through the e-commerce business data knowledge vectors of no less than two types; thus, based on the e-commerce business data knowledge vectors of no less than two types, several selections are made in the pre-configured e-commerce business data set, and the target e-commerce business data more related to the e-commerce business data to be recommended can be selected; therefore, through the knowledge vectors of no less than two types, a large number of specified e-commerce business data sets can be effectively selected, which can not only improve the efficiency of selecting the target e-commerce business data but also improve the accuracy of selecting the target e-commerce business data; thus, data can be pushed more accurately, and the accuracy and reliability of data push can be improved.

[0004] However, the following problems still exist in the prior art:

[0005] In the prior art, the differences in user-side behavior tendency characteristics are not considered, resulting in different user-side sensitive tendency categories. Due to the balanced push of online courses for each user side by big data, the placement frequency of courses with user-side behavior tendencies is inappropriate, leading to the problem of low efficiency in intelligent recommendation of online courses. Summary of the Invention

[0006] To this end, the present invention provides an intelligent recommendation method and system for online courses based on big data, so as to overcome the problem in the prior art that due to the failure to consider the differences in the behavioral tendency characteristics of the user side, different types of behavioral tendencies of the user side occur, and due to the balanced push of online courses for each user side by big data, the placement frequency of courses with behavioral tendencies of the user side is inappropriate, resulting in low efficiency of intelligent recommendation of online courses.

[0007] To achieve the above object, the present invention provides an intelligent recommendation method for online courses based on big data, including:

[0008] Obtain the behavioral tendency characteristics of the user side before the behavior of purchasing online courses and the changes in sub-characteristics among several behavioral tendency characteristics;

[0009] Based on the changes in sub-characteristics among the behavioral tendency characteristics of the user side, analyze the sensitive sub-characteristics of each user side, construct a set of sensitive sub-characteristics, and analyze the sensitive tendency representation value based on the set of sensitive sub-characteristics corresponding to the user side;

[0010] Based on the sensitive tendency representation value, determine the sensitive tendency category of the user side, analyze the real-time data, and push the courses, including:

[0011] Conduct placement verification push, obtain the response of each sub-characteristic after the placement verification push, analyze whether to adjust the push method, and push the courses in the corresponding push method;

[0012] Or, collect the page closing rate of the user side, analyze whether the user side has aversion behavior based on the page closing rate, and re-conduct verification after reducing the push frequency;

[0013] Among them, the behavioral tendency characteristics include the course trial view duration, trial view completion rate, and click-through rate of the user side; the placement verification push includes pushing the courses at a predetermined push frequency.

[0014] Further, the process of analyzing the sensitive sub-characteristics of each user side based on the changes in sub-characteristics among the behavioral tendency characteristics of the user side includes:

[0015] Calculate the mean difference ratio between the data of the sub-characteristics in the behavioral tendency characteristics of the user side and the preset threshold of the corresponding sub-characteristics, and determine the difference ratio mean as the characteristic factor;

[0016] Compare the characteristic factors of each sub-characteristic;

[0017] Determine the two sub-characteristics with larger numerical values in the compared characteristic factors as the sensitive sub-characteristics.

[0018] Further, the process of analyzing the sensitive tendency representation value based on the set of sensitive sub-characteristics includes:

[0019] Calculate the ratio of the first sensitive sub-feature of the client to the corresponding threshold of the first sensitive sub-feature to obtain the first recommendation influencing factor;

[0020] Calculate the ratio of the second sensitive sub-feature of the client to the corresponding threshold of the second sensitive sub-feature to obtain the second recommendation influencing factor;

[0021] Determine the sum of the first recommendation influencing factor and the second recommendation influencing factor as the sensitive tendency characterization value.

[0022] Furthermore, determine the sensitive tendency category of the client based on the sensitive tendency characterization value, including,

[0023] If the sensitive tendency characterization value is greater than the behavior tendency feature threshold, it is determined as the first sensitive tendency category;

[0024] If the sensitive tendency characterization value is less than or equal to the behavior tendency feature threshold, it is determined as the second sensitive tendency category.

[0025] Furthermore, analyze the real-time data and push courses, including,

[0026] If the classification result is the first sensitive tendency category, conduct a delivery verification push, obtain the response of each sub-feature after the delivery verification push, and analyze whether to adjust the push method to push the course in the corresponding push method;

[0027] If the classification result is the second sensitive tendency category, collect the page closing rate of the client, and analyze whether the client has an aversion behavior based on the page closing rate, and then re-verify after reducing the push frequency.

[0028] Furthermore, the process of obtaining the response of each sub-feature after the delivery verification push and analyzing whether to adjust the push method includes,

[0029] Extract the reduction amount of each sensitive sub-feature within the time period;

[0030] If the reduction amount of any sensitive sub-feature is greater than or equal to the predetermined change threshold, it is determined to adjust the push method and push the course in the corresponding push method.

[0031] Furthermore, adjusting the push method to push the course in the corresponding push method includes,

[0032] Determine the keywords in each of the online course pages;

[0033] Determine the online course category to which each keyword belongs;

[0034] Among them, the attribution relationship between each keyword and the online course category is preset;

[0035] Determine the remaining unvisited online course categories based on the visited online course categories;

[0036] Determine the online course push frequencies that need to be adjusted for each of the unvisited online course categories;

[0037] Determine to increase the online course push frequencies.

[0038] Further, analyze whether there is an aversion behavior at the user side based on the behavior characteristics, including,

[0039] Extract the page closing rate before the user side purchases an online course;

[0040] If the page closing rate is greater than or equal to a predetermined page closing threshold, it is determined that there is an aversion behavior at the user side, and the push frequency is reduced and then verified again.

[0041] Further, the process of reducing the push frequency includes,

[0042] Solve the average reduction amount of each sensitive sub-feature;

[0043] Determine that the reduction amount of the online course push frequency is positively correlated with the average reduction amount.

[0044] Further, the present invention provides an intelligent online course recommendation system based on big data, including,

[0045] A data collector for obtaining the behavior tendency characteristics before the user side purchases an online course and the change amounts of the sub-features of the behavior tendency characteristics;

[0046] A feature analyzer connected to the data collector for analyzing each sensitive sub-feature of each user side based on the change situation of the sub-features in the user side behavior tendency characteristics, and for analyzing the sensitive tendency representation value based on the set of sensitive sub-features;

[0047] A course recommendation component respectively connected to the data collector and the feature analyzer, including a clustering unit and a control unit, where the clustering unit is used to distinguish sensitive tendency categories according to the sensitive tendency representation value, including;

[0048] The control unit determines the sensitive tendency category of the user side based on the sensitive tendency representation value, analyzes the real-time data, and pushes courses, including,

[0049] Perform a delivery verification push, obtain the response situations of the sub-features after the delivery verification push, analyze whether to adjust the push method, and push courses in the corresponding push method;

[0050] Alternatively, collect the page closing rate of the user terminal, analyze whether the user terminal has aversion behavior based on the behavioral characteristics, reduce the push frequency and then re-verify;

[0051] Among them, the behavioral tendency characteristics include the trial viewing duration, trial viewing completion rate, and click-through rate of the user terminal; the delivery verification push includes pushing courses at a predetermined push frequency.

[0052] Compared with the prior art, the present invention can analyze the sensitive sub-characteristics for each user terminal based on the changes in the sub-characteristics of the behavioral tendency characteristics of the user terminal before purchasing an online course and the change amounts of the sub-characteristics of the behavioral tendency characteristics, and the course subscription characteristics, improving the efficiency of intelligent recommendation of online courses. By analyzing the sensitive tendency of the course to distinguish sensitive tendency categories, it customizes the push frequency of online courses for each user terminal. By analyzing the page closing rate, it reduces the occurrence of the situation where the user terminal is averse to online pushes. By determining the online course pages already visited by the user terminal, determining the categories of the online courses already visited based on each of the online course pages, and being able to effectively adjust the course push frequency for each online course category based on each of the categories of the online courses already visited, thereby further improving the efficiency of the online course push frequency.

[0053] In particular, the present invention provides an intelligent online course recommendation system based on big data, including a data collector, a feature analyzer, and a course recommendation component. The present invention can obtain the behavioral tendency characteristics of the user terminal before the purchase behavior and the change amounts of the sub-characteristics of the behavioral tendency characteristics of the user terminal through the data collector, improve the efficiency of the course push frequency through the behavioral tendency characteristics and the user-sensitive tendency characterization value, can distinguish the first sensitive tendency category and the second sensitive tendency category through the course recommendation component, and can accurately analyze whether to adjust the push method through the change range of each sensitive sub-characteristic of the user terminal. Therefore, the present invention considers the online course pages already visited by the user terminal, considers the online course categories with potential tendencies that the user terminal has not browsed, mines the potential tendencies of the user terminal, determines the online course pages already visited by the user terminal, determines the categories of the online courses already visited based on each of the online course pages, and can effectively adjust the course push frequency for each online course category based on each of the categories of the online courses already visited, further improving the efficiency of the course push frequency and improving the accuracy of online course recommendation.

[0054] In particular, for the second sensitive tendency category of the course sensitive tendency, by collecting the page closing rate of the user terminal and analyzing the user terminal based on the page closing rate, it can analyze whether aversion behavior occurs, reduce the push frequency and then re-verify, reducing the occurrence of abnormal aversion of the user terminal and improving the online course push efficiency. Description of the Drawings

[0055] Figure 1 The flowchart of the steps of the online course intelligent recommendation method based on big data according to the embodiment of the present invention;

[0056] Figure 2 The flowchart of the steps for analyzing the sensitive sub-features of each client based on the changes of each sub-feature of the client behavior tendency characteristics according to the embodiment of the present invention;

[0057] Figure 3 The flowchart of the steps for analyzing the sensitive tendency characterization value based on the sensitive sub-feature set according to the embodiment of the present invention;

[0058] Figure 4 The block diagram of the structure of the online course intelligent recommendation system based on big data according to the embodiment of the present invention. Detailed implementation manners

[0059] In order to make the purpose and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0060] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0061] It should be noted that in the description of the present invention, unless otherwise clearly defined and limited, the term "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0062] Please refer to Figure 1 As shown, it is the flowchart of the steps of the online course intelligent recommendation method based on big data according to the embodiment of the present invention. The present invention provides an online course intelligent recommendation method based on big data, including:

[0063] Step S1, obtaining the behavior tendency characteristics of the client before purchasing an online course and the changes of sub-features among several behavior tendency characteristics;

[0064] Step S2, analyzing the sensitive sub-features of each client based on the changes of sub-features in the client behavior tendency characteristics, constructing a sensitive sub-feature set, and analyzing the sensitive tendency characterization value based on the sensitive sub-feature set corresponding to the client;

[0065] Step S3: Determine the sensitive tendency category of the client based on the sensitive tendency characterization value, analyze the real-time data, and push courses, including:

[0066] Conduct delivery verification push, obtain the response of each sub-feature after the delivery verification push, and analyze whether to adjust the push method to push the course in the corresponding push method;

[0067] Or, collect the page closing rate of the client, analyze whether the client has aversion behavior based on the page closing rate, and re-conduct verification after reducing the push frequency;

[0068] Among them, the behavior tendency features include the course preview duration, preview completion rate, and click-through rate of the client; the delivery verification push includes pushing courses at a predetermined push frequency.

[0069] Specifically, for the online course, it can be a pre-recorded college entrance examination tutoring video course, or other forms of pre-recorded video courses, which will not be elaborated here.

[0070] Specifically, there is no limitation on the classification of online course categories. In implementation, the online course categories are mathematics, physics, chemistry, biology, Chinese, and English, and other forms can also be adopted, which will not be elaborated here.

[0071] Specifically, there are three sub-features of the client behavior tendency features, all of which are data obtained within a predetermined time before the purchase behavior is generated. The predetermined time can be selected within [5min, 25min].

[0072] Specifically, each sub-feature is the preview duration of a single course, the preview completion rate of a single course, and the click-through rate corresponding to a single course.

[0073] Specifically, there is no limitation on the method of obtaining the client behavior tendency features. It can be obtained by using relevant tracking software, such as one of ClickMagick and Trackier. Of course, other forms can also be adopted as long as the behavior tendency features can be obtained.

[0074] Specifically, the present invention can analyze the sensitive sub - features for each client based on the changes in the sub - features of the client's behavior tendency characteristics before the behavior of purchasing online courses, and the change amounts of each sub - feature of the behavior tendency characteristics, improving the efficiency of intelligent recommendation of online courses. By analyzing the sensitive tendency of the courses to distinguish the sensitive tendency categories, it customizes the push frequency of online courses for each client. By analyzing whether to adjust the push method, it determines the accessed online course categories based on each of the online course pages, and can effectively adjust the course push frequency for each online course category based on each accessed online course category. By using the page closing rate of the client to reduce the push frequency, it reduces the occurrence of the situation where the client is disgusted with online pushes, thereby improving the efficiency of the online course push frequency.

[0075] Please refer to Figure 2 as shown, which is a flowchart of the steps for analyzing the sensitive sub - features for each client based on the changes in the sub - features of the client's behavior tendency characteristics in the embodiment of the present invention. The process of analyzing the sensitive sub - features for each client based on the changes in the sub - features in the client's behavior tendency characteristics includes:

[0076] Step S11, calculate the mean of the difference ratios between the data of the sub - features in the client's behavior tendency characteristics and the preset thresholds of the corresponding sub - features, and determine the mean of the difference ratios as the feature factor;

[0077] Step S12, compare the feature factors of each sub - feature;

[0078] Step S13, determine the two sub - features with larger numerical values among the compared feature factors as the sensitive sub - features.

[0079] In implementation, there are several behavior tendency characteristics corresponding to the client, which are obtained based on several purchase behavior records of the client, and this will not be elaborated here.

[0080] In implementation, among the behavior tendency characteristics, the preset threshold corresponding to the sub - feature of the trial - view completion rate is determined in advance. Specifically, within the historical period, the mean click frequency of several clients is calculated, and the preset threshold is set as the product of the mean of the trial - view completion rate and the precision coefficient. The precision coefficient is selected within the interval [1.1, 1.2], and the historical period is the same as the predetermined duration.

[0081] Similarly, the preset thresholds of each sub - feature all come from the product of the behavior tendency characteristics of the corresponding sub - feature in the historical period of big data and the precision coefficient, and this will not be elaborated here.

[0082] The difference ratio is the ratio of the difference between two values to the mean of the two values, and this will not be elaborated here.

[0083] In implementation, before the client has a purchase behavior, the feature factors corresponding to the sensitive sub - features of each sub - feature are respectively:

[0084] The corresponding feature factor of the single-course trial view duration sub-feature is 0.62;

[0085] The corresponding feature factor of the single trial completion rate sub-feature is 0.58;

[0086] The corresponding feature factor of the click-through rate sub-feature corresponding to a single course is 0.60;

[0087] Since 0.62 > 0.60 > 0.58, the user-side sensitive sub-features of this embodiment are the sub-feature of the course trial view duration and the sub-feature of the click-through rate corresponding to the course, indicating that the user side has strong data representativeness in the course trial view duration and the sub-feature of the click-through rate corresponding to the course before the purchase behavior occurs, and the numerical change is relatively obvious. Therefore, the online course intelligent recommendation system described in the present invention will subsequently adaptively analyze the sensitive sub-features held by the user side. Furthermore, the first sensitive sub-feature is the course trial view duration, and the second sensitive sub-feature is the click-through rate. The customized recommendation of the present invention improves the efficiency of online course recommendation.

[0088] Please refer to Figure 3 as shown, which is a step flow chart for analyzing the sensitive tendency characterization value based on the sensitive sub-feature set. The process of analyzing the sensitive tendency characterization value based on the sensitive sub-feature set includes

[0089] Step S21: Calculate the ratio of the first sensitive sub-feature of the user side to the corresponding first sensitive sub-feature threshold to obtain the first recommendation influencing factor;

[0090] Step S22: Calculate the ratio of the second sensitive sub-feature of the user side to the corresponding second sensitive sub-feature threshold to obtain the second recommendation influencing factor;

[0091] Step S23: Determine the sum of the first recommendation influencing factor and the second recommendation influencing factor as the sensitive tendency characterization value.

[0092] Specifically, the first sensitive sub-feature threshold is 1.15 times the average value of the corresponding first sensitive sub-feature value of the user side in the historical period in the system; similarly, the second sensitive sub-feature threshold is 1.25 times the average value of the user's second sensitive sub-feature in the three months before the system recommendation behavior in the historical period

[0093] Specifically, the sensitive sub-feature thresholds corresponding to different sub-features are determined in advance,

[0094] For the single-course trial view duration sub-feature, it is determined as the average value of the single-course trial view duration of the user side in each historical period;

[0095] For the single trial completion rate sub-feature, it is determined as the average value of the single trial completion rate of the user side in each historical period;

[0096] For the click-through rate sub-feature corresponding to a single course, it is determined as the average value of the click-through rates corresponding to the single course on the user side in each historical period.

[0097] Specifically, determining the sensitive tendency category of the user side based on the sensitive tendency characterization value includes,

[0098] If the sensitive tendency characterization value is greater than the behavior tendency feature threshold, it is determined as the first sensitive tendency category;

[0099] If the sensitive tendency characterization value is less than or equal to the behavior tendency feature threshold, it is determined as the second sensitive tendency category.

[0100] Specifically, since the data reference value of the sensitive sub-feature is high, therefore, it is determined in the form of taking the ratio of the corresponding sensitive sub-feature threshold, the change of the actual behavior tendency feature relative to the normal situation, and then the user side with an obvious behavior sensitive tendency is identified.

[0101] Specifically, analyzing the real-time data and pushing courses, including,

[0102] If the classification result is the first sensitive tendency category, then conduct a delivery verification push, obtain the response of each sub-feature after the delivery verification push, analyze whether to adjust the push method, and push the course in the corresponding push method;

[0103] If the classification result is the second sensitive tendency category, then collect the page closing rate of the user side, analyze whether the user side has an aversion behavior based on the page closing rate, and re-conduct the verification after reducing the push frequency.

[0104] In implementation, in response to the classification result of the sensitive tendency category, determine the types of online courses recommended for the user side, determine the online course page corresponding to the behavior tendency feature generated by the user side, determine the online course category to which each keyword belongs from the keywords of the online course page, recommend the courses of the corresponding online course category, and then analyze the change range of each sensitive sub-feature of the user side after the online course recommendation to determine whether to adjust the push method to reduce the online course push frequency.

[0105] Specifically, the process of obtaining the response of each sub-feature after the delivery verification push and analyzing whether to adjust the push method includes,

[0106] Extract the reduction amount of each sensitive sub-feature within the time period;

[0107] If there is any reduction amount of a sensitive sub-feature that is greater than or equal to the predetermined change threshold, it is determined to adjust the push method and push the course in the corresponding push method.

[0108] In implementation, the change threshold for sensitive sub-features is determined based on the sensitive sub-feature thresholds of the corresponding sub-features, and is set to be between 1.1 times and 1.2 times the sensitive sub-feature thresholds.

[0109] Specifically, adjust the push method and push the courses in the corresponding push method, including,

[0110] Determine the keywords in each of the online course pages;

[0111] Determine the online course categories to which each of the keywords belong;

[0112] Among them, the attribution relationship between each keyword and the online course category is preset;

[0113] Based on the accessed online course categories, determine the remaining unaccessed online course categories;

[0114] Determine the online course push frequency that needs to be adjusted for each of the unaccessed online course categories;

[0115] Determine to increase the online course push frequency.

[0116] Specifically, increase the online course push frequency to be between 1.5 times and 1.8 times the initial online course push frequency.

[0117] Specifically, for the second sensitive tendency category of the course sensitive tendency, its browsing behavior data has poor representativeness, and the behavior sensitive tendency or tendency is not obvious. Therefore, the present invention considers the online course categories with potential tendencies that have not been browsed by the user terminal based on the online course pages accessed by the user terminal, and mines the potential tendencies of the user terminal to improve the accuracy of online course recommendation.

[0118] Specifically, based on the behavioral characteristics, analyze whether the user terminal has an aversion behavior, including,

[0119] Extract the page closing rate of the user terminal before the behavior of purchasing an online course;

[0120] If the page closing rate is greater than or equal to the predetermined page closing threshold, it is determined that the user terminal has an aversion behavior, and the push frequency is reduced and then verified again.

[0121] Specifically, the process of reducing the push frequency includes,

[0122] Solve the mean value of the reduction amounts of each sensitive sub-feature;

[0123] Determine that the reduction amount of the online course push frequency is positively correlated with the mean value of the reduction amounts.

[0124] In implementation, the ratio of the reduction amount mean to the Euclidean change threshold is solved, and the obtained ratio is multiplied by the initial online course push frequency to obtain the online course push frequency that needs to be reduced.

[0125] Please refer to Figure 4 shown, which is a structural block diagram of an intelligent online course recommendation system based on big data according to an embodiment of the present invention. The present invention also provides an intelligent online course recommendation system based on big data, including,

[0126] A data collector, which is used to obtain the behavior tendency characteristics before the user terminal purchases an online course and the change amounts of the sub-characteristics of the behavior tendency characteristics;

[0127] A feature analyzer, which is connected to the data collector, and is used to analyze the sensitive sub-characteristics of each user terminal based on the change situation of the sub-characteristics in the user terminal behavior tendency characteristics, and is used to analyze the sensitive tendency representation value based on the set of sensitive sub-characteristics;

[0128] A course recommendation component, which is respectively connected to the data collector and the feature analyzer, and includes a clustering unit and a control unit. The clustering unit is used to distinguish sensitive tendency categories according to the sensitive tendency representation value, including;

[0129] The control unit determines the sensitive tendency category of the user terminal based on the sensitive tendency representation value, analyzes the real-time data, and pushes courses, including,

[0130] Performing delivery verification push, obtaining the response situation of each sub-characteristic after the delivery verification push, and analyzing whether to adjust the push method to push courses in the corresponding push method;

[0131] Or, collecting the page closing rate of the user terminal, analyzing whether the user terminal has aversion behavior based on the behavior characteristics, and re-verifying after reducing the push frequency;

[0132] Among them, the behavior tendency characteristics include the course preview duration, preview completion rate, and click-through rate of the user terminal; the delivery verification push includes pushing courses at a predetermined push frequency.

[0133] Specifically, the specific structures of the data collector, the feature analyzer, and the course recommendation component are not limited. It itself or each unit therein can be composed of logic components, and the logic components include field programmable components, computers, or microprocessors in a computer.

[0134] The present invention provides an intelligent recommendation system for online courses based on big data, including a data collector, a feature analyzer, and a course recommendation component. Through the data collector, the present invention can obtain the behavioral tendency characteristics before the purchase behavior of the user terminal, the change amount of each sub-feature of the behavioral tendency characteristics of the user terminal, and the course ordering characteristics. The efficiency of the course push frequency is improved through the behavioral tendency characteristics and the sensitive tendency characterization value of the user terminal. The course recommendation component can distinguish the first sensitive tendency category and the second sensitive tendency category. Whether to adjust the push method can be accurately analyzed through the change range of each sensitive sub-feature of the user terminal. Based on each of the online course pages, the accessed online course categories are determined. Based on each of the accessed online course categories, the course push frequency for each online course category can be effectively adjusted, improving the efficiency of the course push frequency. At the same time, through the analysis of the page closing rate of the user terminal, the occurrence of abnormal aversion on the user terminal is reduced.

[0135] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the protection scope of the present invention.

Claims

1. An intelligent recommendation method for online courses based on big data, characterized in that, including: Obtaining the behavioral tendency characteristics of the client before purchasing an online course and the changes in sub-characteristics among several behavioral tendency characteristics; Analyzing the sensitive sub-characteristics of each client based on the changes in sub-characteristics among the client's behavioral tendency characteristics, constructing a set of sensitive sub-characteristics, and analyzing the sensitive tendency representation value based on the set of sensitive sub-characteristics corresponding to the client; Determining the sensitive tendency category of the client based on the sensitive tendency representation value, analyzing the real-time data, and pushing courses; including, Conducting delivery verification push, obtaining the response of each sub-characteristic after the delivery verification push, analyzing whether to adjust the push method, and pushing the course in the corresponding push method; or, collecting the page closing rate of the client, analyzing whether the client shows aversion behavior based on the page closing rate, and re-verifying after reducing the push frequency; wherein, the behavioral tendency characteristics include the course trial duration, trial completion rate, and click-through rate of the client; the delivery verification push includes pushing courses at a predetermined push frequency.

2. The intelligent online course recommendation method based on big data according to claim 1, wherein The process of analyzing the sensitive sub-characteristics of each client based on the changes in sub-characteristics among the client's behavioral tendency characteristics includes, Calculating the mean difference ratio between the data of the sub-characteristics among the client's behavioral tendency characteristics and the preset threshold of the corresponding sub-characteristics, and determining the difference ratio mean as the characteristic factor; Comparing the characteristic factors of each sub-characteristic; Determining the two sub-characteristics with larger numerical values among the compared characteristic factors as the sensitive sub-characteristics.

3. The intelligent online course recommendation method based on big data according to claim 1, wherein The process of analyzing the sensitive tendency representation value based on the set of sensitive sub-characteristics includes, Calculating the ratio of the first sensitive sub-characteristic of the client to the corresponding first sensitive sub-characteristic threshold to obtain the first recommendation influencing factor; Calculating the ratio of the second sensitive sub-characteristic of the client to the corresponding second sensitive sub-characteristic threshold to obtain the second recommendation influencing factor; Determining the sum of the first recommendation influencing factor and the second recommendation influencing factor as the sensitive tendency representation value.

4. The intelligent online course recommendation method based on big data according to claim 1, wherein, Determining the sensitive tendency category of the client based on the sensitive tendency representation value includes, If the sensitive tendency representation value is greater than the behavioral tendency characteristic threshold, it is determined as the first sensitive tendency category; If the sensitive tendency representation value is less than or equal to the behavioral tendency characteristic threshold, it is determined as the second sensitive tendency category.

5. The intelligent online course recommendation method based on big data according to claim 4, wherein, Analyzing the real-time data and pushing courses includes, If the classification result is the first sensitive tendency category, conducting a delivery verification push, obtaining the response of each sub-characteristic after the delivery verification push, analyzing whether to adjust the push method, and pushing the course in the corresponding push method; If the classification result is the second sensitive tendency category, collecting the page closing rate of the client, analyzing whether the client shows aversion behavior based on the page closing rate, and re-verifying after reducing the push frequency.

6. The intelligent online course recommendation method based on big data according to claim 1, wherein, The process of obtaining the response of each sub-characteristic after the delivery verification push and analyzing whether to adjust the push method includes, Extracting the reduction amount of each sensitive sub-characteristic within the time period; If there is any reduction amount of a sensitive sub-characteristic that is greater than or equal to the predetermined change threshold, it is determined to adjust the push method and push the course in the corresponding push method.

7. The online course intelligent recommendation method based on big data according to claim 1, characterized in that Adjusting the push method and pushing the course in the corresponding push method, including, Determining the keywords in each of the online course pages; Determining the online course category to which each of the keywords belongs; Among them, the attribution relationship between each keyword and the online course category is preset; Based on the accessed online course categories, determine the remaining unaccessed online course categories; Determine the need to adjust the online course push frequency for each of the unaccessed online course categories; Determine to increase the online course push frequency.

8. The online course intelligent recommendation method based on big data according to claim 1, characterized in that, Based on the behavioral characteristics, analyze whether there is an aversion behavior at the user side, including Extract the page closing rate before the user side purchases an online course; If the page closing rate is greater than or equal to a predetermined page closing threshold, it is determined that there is an aversion behavior at the user side, and the push frequency is reduced and verified again.

9. The online course intelligent recommendation method based on big data according to claim 1, characterized in that, The process of reducing the push frequency includes Solve the average reduction amount of each sensitive sub-feature; Determine that the reduction amount of the online course push frequency is positively correlated with the average reduction amount.

10. A system using the online course intelligent recommendation method based on big data according to any one of claims 1 to 9, characterized in that, Including A data collector for obtaining the behavioral tendency characteristics before the user side purchases an online course and the change amounts of the sub-characteristics of the behavioral tendency characteristics; A feature analyzer connected to the data collector for analyzing each sensitive sub-feature of each user side based on the change of the sub-feature in the user side behavioral tendency characteristics, and for analyzing the sensitive tendency representation value based on the set of sensitive sub-features; A course recommendation component respectively connected to the data collector and the feature analyzer, including a clustering unit and a control unit, where the clustering unit is used to distinguish sensitive tendency categories according to the sensitive tendency representation value, including; The control unit determines the sensitive tendency category of the user side based on the sensitive tendency representation value, analyzes the real-time data, and pushes the courses Including Conduct delivery verification push, obtain the response of each sub-feature after the delivery verification push, analyze whether to adjust the push method, and push the courses in the corresponding push method; Or, collect the page closing rate of the user side, analyze whether there is an aversion behavior at the user side based on the behavioral characteristics, reduce the push frequency and verify again; Among them, the behavioral tendency characteristics include the course trial duration, trial completion rate and click-through rate of the user side; the delivery verification push includes pushing the courses at a predetermined push frequency.

Images