Description:Complete Specification:
The following specification describes and ascertains the nature of this invention and the manner in which it is to be performed.

Field of the invention:
[0001] The present invention relates to a field of detecting engagement level of a student in an online course and method therefor.

Background of the invention:
[0002] In recent times online classes are becoming very popular where students can subscribe and learn at their own pace. There are also a lot of courses that can be taken online where teaching happen live. That is, the teacher is teaching in real time to a group of students subscribed to these online courses at specific schedules. However, in such online courses, it is very difficult for the teachers to identify if the teachings are effective because there is minimal interaction between the teacher and the students. Hence there is a need to understand the engagement level of a student subscribed for a specific online course.

[0003] According to a prior art US2020135045A, Various systems and methods for engagement dissemination. A face detector detects a face in video data. A context filterer determines a student is on-platform and a section type. An appearance monitor selects an emotional and a behavioral classifiers. Emotional and behavioral components are classified based on the detected face. A context-performance monitor selects an emotional and a behavioral classifiers specific to the section type, Emotional and behavioral components are classified based on the log data. A fuser combines the emotional components into an emotional state of the student based on confidence values of the emotional components. The fuser combines the behavioral components a behavioral state of the student based on confidence values of the behavioral components. The user determine an engagement level of the student based on the emotional state and the behavioral state of the student.

Brief description of the accompanying drawings:
[0004] An embodiment of the disclosure is described with reference to the following accompanying drawings,
[0005] Fig. 1 is a flowchart illustrating a method to detect engagement level of a student in an online course, in accordance with one embodiment of the present invention;
[0006] Fig. 2 is a block diagram of a device (200) to detect engagement level of a student in an online course, in accordance with one embodiment of the present invention; and
[0007] Fig. 3 is a block diagram of a system for detecting engagement level of a student in an online course, in accordance with one embodiment of the present invention.

Detailed description of the embodiments:
[0008] Fig. 1 is a flowchart illustrating a method to detect engagement level of a student in an online course, in accordance with one embodiment of the present invention. At step 105, an emotional aspect of the student is determined by analyzing facial expression of the student. In one example, the facial expression can be captured using a camera of a computing device (200). This computing device (200) may be used by the student to take the online course.

[0009] Various facial expression being captured by the camera is analyzed for determining the emotional aspect of the student. The emotional aspect of the student represents an emotion experienced by the user in real time during the online course. Examples of the emotions include, but are not limited to, happiness, sadness, confused, angry and surprised. By determining such emotional aspect of the student, feelings of the student with regard to that specific online course can be obtained. In other words, the emotional aspect of the student indicates if the student is happy with the online course, or is the student confused with the online course. Based on the emotional aspect of the student, one or more actions can be taken. Examples of the actions can include making alterations in the course material or making alteration in the way the course material is taught to the students.

[0010] The camera being attached to the computing device (200) captures face of the student. Further, the face of the student, captured by the camera, is transmitted to a control unit (210) for analyzing the facial expression of the student. Analysis include assessing head movement, head position, eyebrow position, like eyebrows raised or eyebrows pulled together. Further, the analysis of such facial expressions is performed using machine learning models. In one example, the machine learning model is a Random Forest Model. However, it should be noted that the analysis of facial expression should not be limited to Random Forest Model alone and various other Machine learning models can be used for analyzing the facial expression. For example, if the analysis of the facial expression is such that the eyebrows are pulled together then it is inferred that the emotion experienced by the student is confusion and hence, the emotional aspect of the student is considered to be confused.

[0011] Based on such emotional aspect, a first weightage value is assigned. For example, when the emotional aspect of the student is considered to be confused then accordingly a specific first weightage value is assigned. Similarly, the emotional aspect of the student is considered to be happy, then another specific first weightage value is assigned. Therefore, each emotion will be assigned a specific first weightage value.

[0012] At step 110, a behavioral aspect of the student is determined by monitoring one or more attributes corresponding to the online course. Examples, of the attributes include, but are not limited to, a plurality of assessments, associated with the online course, attempted by the student. The attributes also include a plurality of assignments associated with the online course. The attributes also include time taken, by the student, to complete the online course and number of attempts, made by the student, for completing the online course.

[0013] The behavioral aspect of the student is determined for obtaining the complexity of learning content involved in the online course. For example, if the student is taking multiple attempts to complete the assessment or the assignments associated the online course, then it can be inferred that the student is finding it difficult to comprehend the learning content of the online course. Therefore, monitoring of the one or more attributes are used for understanding if the learning content is comprehensive to the student.

[0014] The monitoring of the one or more attributes is performed by analyzing if the student is taking the assignment or the assessment in one straight go or if there are too many frequent breaks. It is also analyzed, if each online class session in the online course is being completely attended by the student or is the student taking frequent breaks during the online class session. Further, it is also analyzed the number of times pause and play button are being pressed by the student. Various machine learning models are used for such analysis. Examples of the machine learning models include, but are not limited to Random Forest Model.

[0015] Each of the one or more attribute is given a specific score by the machine learning models. Further, an average score of all the attributes is obtained. This average score indicates a second weightage value for the behavioral aspect.

[0016] Some of the example attributes for determining the behavioral aspect include “low time” which means lowest time is taken for completing the online course. “high time” which indicates highest time is taken by the student for completing the online course. “low score” which indicates lowest score is obtained by the student in assessments and assignments. “high score” indicating highest score obtained by the student for assessments and assignments. “Moderate score” indicating moderate score obtained by the student for assessments and assignments. These example attributes can be considered individually or can be taken in combination for determining the second weightage value that indicates the behavioral aspect of the student overall.

[0017] It should be noted that the steps performed in step 105 and step 110 is performed for each student and hence the emotional aspect of each student and the behavioral aspect of each student is calculated and stored in a memory of the control unit (210).

[0018] At step 115, complexity level of the online course is determined. Such determination of the complexity level is performed by calculating an aggregate aspect of the online course. The aggregate aspect of the online course is obtained by calculating the average of the emotional aspect and the behavioral aspect associated with each student.

[0019] Such complexity level is determined to identify if the online course is in a comprehensive manner to the student. For example, if the average of the behavioral aspect associated with each student is “high time” and “moderate score”. Based on the complexity level, one or more actions can be performed. Examples of such actions include, but are not limited to, modification in the learning content of the online course, modification in the flow of the learning content in the online course and teaching method employed for teaching the learning content.

[0020] Examples of the complexity levels include, but are not limited to, “less complex”, “not complex” and “moderately complex”. These complexity level are calculated by aggregating average value of the emotional aspect of all the students and average value of the behavioral aspect of all the students. It should be noted that these are just exemplary denotation and there can be many such complexity levels.

[0021] At step 120, the engagement level of the student based on at least one of the emotional aspect of the student, the behavioral aspect of the student and the complexity level determined in step 115 is calculated. Engagement level denotes the level of engagement of the student on the online course. It provides information on level of interest of the student on the online course and it also provides a hint if the student can complete the online course successfully. Also, by determining how engaged the student is, the teacher can also make necessary modification in the way of teaching so that the engagement level of the student can be increased.

[0022] Such calculation is performed by aggregating the first weightage value and the second weightage value and the complexity level which is an aggregate of the emotional aspect and the behavioral aspect of all the other students taking the online course. Based on such calculation, the engagement level of the student is determined.

[0023] Some example scenarios of such calculation is described in the table below:
First weightage value Second weightage value Aggregate aspect Engagement level Meaning
Student-1 happy Low time Not complex Fully engaged Student-1 is fully engaged in the course. He is able to comprehend learning topics in the online course and complete it well on time.
Student-2 happy High time “moderately complex” Fully engaged Student-2 is also fully engaged in the online course but is taking time to complete the assessment or the assignments or taking time to complete the course. Hence, one inference is that the course topic is difficult to comprehend.
Student-3 confused High time Complex Not engaged Student-3 has a confused emotion and is also taking time to complete the course. Therefore, this situation can be inferred that the course topic is difficult to comprehend.
Student-4 Sad High time Not complex Not engaged Student-4 has a sad emotion and is also taking time to complete the course. Therefore, this situation can be inferred that the student-4 is not interested in the online course.

[0024] Similarly, various such inferences that denote the engagement level can be obtained based on the emotional aspect of the student, the behavioral aspect of the student and the complexity level of the online course. Various Machine learning models are used for obtaining multiple probabilities that are obtained in combination of the emotional aspect of the student and the behavioral aspect of the student. Examples of the machine learning models include, but are not limited to, Fusion by highest confidence (HC), Fusion by Majority Voting (MV), or Fusion by Hybrid Majority Voting (HMV).

[0025] Fig. 2 is a block diagram of a device (200) to detect engagement level of a student in an online course, in accordance with one embodiment of the present invention.

[0026] The device (200) includes an input module (205) for receiving at least one of a facial expression of the student and one or more attributes of the student. Examples of the input module (205) include, but are not limited to, camera, mouse, keyboard and joystick.

[0027] The input module (205), for example, the camera is used for capturing the facial expression of the student in real time. In one embodiment, the camera captures the facial expression continuously. In another embodiment, the camera captures the facial expression at regular intervals. The facial expression is used for determining an emotional aspect of the student. The emotional aspect of the students represents an emotion experienced by the user in real time during the online course. Examples of the emotions include, but are not limited to, happiness, sadness, confused, angry and surprised. By determining such emotional aspect of the student, feelings of the student with regard to that specific online course can be obtained. In other words, the emotional aspect of the student indicates if the student is happy with the online course, or is the student confused with the online course.

[0028] The input module (205) is also used for receiving the attributes of the student. Examples of the attributes include, but are not limited to, various assessments, various assignments attempted by the students. The attributes also include time taken by the student to complete the online course and also the number of attempts taken by the student to complete the online course. The attributes also include number of times pause and play button are pressed by the user for each session of the online course.

[0029] The attributes received by the input module (205) are provided by the user in various forms such as opening a webpage pertaining to the online course, number of clicks using mouse button, time period of being on the webpage pertaining to the online course and carious other forms.

[0030] The behavioral aspect of the student is determined for obtaining the complexity of learning content involved in the online course. For example, if the student is taking multiple attempts to complete the assessment or the assignments associated the online course, then it can be inferred that the student is finding it difficult to comprehend the learning content of the online course. Therefore, monitoring of the one or more attributes are used for understanding if the learning content is comprehensive to the user.

[0031] The emotional aspect and the behavioral aspect received by the input module (205) is provided to a control unit (210) for further processing.

[0032] The input module (205) also includes a control unit (210). The control unit (210) is in communication with the input module (205) and hence is adapted to receive the facial features captured by the input module (205) and the attributes provided by the user.

[0033] The control unit (210), upon receiving the facial features, is adapted to determine the emotional aspect of the student. Analysis include assessing head movement, head position, eyebrow position, like eyebrows raised or eyebrows pulled together. The control unit (210) uses machine learning models for performing such analysis. In one example, the machine learning model is a Random Forest Model. However, it should be noted that the analysis of facial expression should not be limited to Random Forest Model alone and various other Machine learning models can be used for analyzing the facial expression. For example, if the analysis of the facial expression is such that the eyebrows are pulled together then it is inferred that the emotion experienced by the student is confusion and hence, the emotional aspect of the student is considered to be confused.

[0034] Further, the control unit (210) is adapted to assign a first weightage value for the emotional aspect. For example, when the facial features of the student is considered to be confused then, the control unit (210) assigns a specific first weightage value to the emotional aspect that correspond to confusion. Similarly, if the facial feature of the student is considered to be happy, then the control unit (210) assigns another specific first weightage value to the emotional aspect that correspond to happiness. Therefore, each emotion will be assigned a specific first weightage value. Therefore upon analysis of the facial features, the control unit (210) determines the emotional aspect of the student by assigning a first weightage value.

[0035] Further, the control unit (210) is also adapted to determine the behavioral aspect of the student. The control unit (210) is adapted to monitor the attributes corresponding to the online course. Such monitoring involves analyzing if the student is taking the assignment or the assessment in one straight go or if there are too many frequent breaks. It is also analyzed, if each online class session is being completely attended by the student or is the student taking frequent breaks during the online class session. Further, it is also analyzed the number of times pause and play button are being pressed by the student. Various machine learning models are used for such analysis.

[0036] The control unit (210) provides a specific score to each of the attributes and further calculates the aggregate score. This aggregate score indicates a second weightage value for the behavioral aspect.

[0037] Therefore, the control unit (210) provides the first weightage value for the emotional aspect and the second weightage value for the behavioral aspect. It should also be noted that the emotional aspect and the behavioral aspect is calculated for each student enrolled in the online course.

[0038] The control unit (210) further determines the complexity level of the online course. The complexity level is determined by calculating an aggregate aspect of the online course. The aggregate aspect of the online course is obtained by calculating the average of the emotional aspect and the behavioral aspect associated with each student.

[0039] Such complexity level is determined to identify if the online course is in a comprehensive manner to the student. For example, if the average of the behavioral aspect associated with each student is “high time” and “moderate score”. Based on the complexity level, one or more actions can be performed. Examples of such actions include, but are not limited to, modification in the learning content of the online course, modification in the flow of the learning content in the online course and teaching method employed for teaching the learning content.

[0040] Examples of the complexity levels include, but are not limited to, “less complex”, “not complex” and “moderately complex”. These complexity levels are calculated by aggregating average value of the emotional aspect of all the students and average value of the behavioral aspect of all the students. It should be noted that these are just exemplary denotation and there can be many such complexity levels.

[0041] Upon calculating the complexity level, the control unit (210) calculates the engagement level of the student based on at least one of the emotional aspect of the student, the behavioral aspect of the student and the complexity level determined in step 115. Such calculation is performed by aggregating the first weightage value and the second weightage value and the complexity level. Based on such calculation, the engagement level of the student is determined.

[0042] Therefore, the control unit (210) calculates the engagement level of each student based on the complexity level of the course which is obtained as an aggregate of emotional aspect and behavioral aspect of all the students and the emotional aspect and the behavioral aspect of each specific student. Hence, such calculation provides a holistic view on how engaged the student is on the online course.

[0043] Some example scenarios of such calculation is described in the table below:
First weightage value Second weightage value Aggregate aspect Engagement level Meaning
Student-1 happy Low time Not complex Fully engaged Student-1 is fully engaged in the course. He is able to comprehend learning topics in the online course and complete it well on time.
Student-2 happy High time “moderately complex” Fully engaged Student-2 is also fully engaged in the online course but is taking time to complete the assessment or the assignments or taking time to complete the course. Hence, one inference is that the course topic is difficult to comprehend.
Student-3 confused High time Complex Not engaged Student-3 has a confused emotion and is also taking time to complete the course. Therefore, this situation can be inferred that the course topic is difficult to comprehend.
Student-4 Sad High time Not complex Not engaged Student-4 has a sad emotion and is also taking time to complete the course. Therefore, this situation can be inferred that the student-4 is not interested in the online course.

[0044] Similarly, various such inferences that denote the engagement level can be obtained based on the emotional aspect of the student, the behavioral aspect of the student and the aggregate aspect of the student. The control unit (210) uses various Machine learning models are used for obtaining multiple probabilities that are obtained in combination of the emotional aspect of the student and the behavioral aspect of the student. Examples of the machine learning models include, but are not limited to, Fusion by highest confidence (HC), Fusion by Majority Voting (MV), or Fusion by Hybrid Majority Voting (HMV).

[0045] Fig. 3 is a block diagram of a system for detecting engagement level of a student in an online course, in accordance with one embodiment of the present invention.

[0046] The system includes a web application running on a plurality of servers (305). The system also includes a plurality of client devices (200a, 200b and 200c) that executes the web application. Examples of the client devices (200a, 200b and 200c) include, but are not limited to, laptops, smartphones, palmtops, tablets and desktops.

[0047] The client devices (200a, 200b and 200c) are adapted to capture a plurality of facial expressions corresponding to a plurality of students. Further, the client devices (200a, 200b and 200c) are also adapted to capture a plurality of attributes corresponding to the students. Detailed explanation of capturing the facial expression and capturing the attributes are explained elaborately in conjunction with Figure 2.

[0048] Further, the system also includes a control unit (210) for determining emotional aspect of the student based on facial expressions. It should be noted that the control unit (210) can be a part of the server of the web application or it can be a part of the individual client devices (200a, 200b and 200c). The details of determining the emotional aspect based on the facial expression is explained in detail in conjunction with Fig. 1 and Fig. 2. It should be noted that the control unit (210) determines the emotional aspect of each and every student enrolled for the online course.

[0049] Further, the control unit (210) determines the behavioral aspect based on the attributes corresponding to the online course. Examples, of the attributes include, but are not limited to, a plurality of assessments, associated with the online course, attempted by the student. The detailed methodology on determining the behavioral aspect of the student is explained in detail in conjunction with Fig. 1 and Fig. 2. The control unit (210) determines the behavioral aspect of every student enrolled for the online course.

[0050] Furthermore, the control unit (210) determines the complexity level of the online course. Such determination of the complexity level is performed by calculating an aggregate aspect of the online course. The aggregate aspect of the online course is obtained by calculating the average of the emotional aspect and the behavioral aspect associated with each student. The detailed methodology on determining the complexity level of the online course is explained in detail in conjunction with Fig. 1 and Fig. 2.

[0051] Further, upon determining the emotional aspect, the behavioral aspect and the complexity level of the online course, the control unit (210) is configured to determine the engagement level of the student on the online course based on the emotional aspect, the behavioral aspect and the complexity level of the online course. Engagement level denotes the level of engagement of the student on the online course. It provides information on interest of the student on the online course and it also provides a hint if the student can complete the online course successfully. Also, by determining how engaged the student is, the teacher can also make necessary modification in the way of teaching so that the engagement level of the student can be increased.

[0052] A few examples are explained in the following paragraphs for the purpose of clear understanding of the disclosure.

[0053] It is considered that an online course named course X is being enrolled by 10 students. Firstly, facial expression of all the 5 students are captured and an emotional aspect associated with each student is determined. Based on such emotional aspect, a first weightage value is assigned. For example, when the emotional aspect of the student is considered to be confused then accordingly a specific first weightage value is assigned. Similarly, the emotional aspect of the student is considered to be happy, then another specific first weightage value is assigned. Therefore, each emotion will be assigned a specific first weightage value.

[0054] The emotional aspect of every student is captured in the table below:

Student Emotional aspect First weightage value
Student-1 happy 0.7
Student-2 joyful 0.8
Student-3 sad 0.2
Student-4 happy 0.7
Student-5 confused 0.1

[0055] Secondly, the one or more attributes corresponding to the course X is monitored. Examples, of the attributes include, but are not limited to, a plurality of assessments, associated with course X, attempted by the student. The attributes also include a plurality of assignments associated with course X. The attributes also include time taken, by the student, to complete course X and number of attempts, made by the student, for completing course X. The monitoring of the one or more attributes is performed by analyzing if the student is taking the assignment or the assessment in one straight go or if there are too many frequent breaks. It is also analyzed, if each online class session is being completely attended by the student or is the student taking frequent breaks during the online class session. Further, it is also analyzed the number of times pause and play button are being pressed by the student. Various machine learning models are used for such analysis. Each of the one or more attribute is given a specific score by the machine learning models. Further, an average score of all the attributes is obtained. This average score indicates a second weightage value for the behavioral aspect. Some of the example attributes for determining the behavioral aspect include “low time” which means lowest time is taken for completing the online course. “high time” which indicates highest time is taken by the student for completing the online course. “low score” which indicates lowest score is obtained by the student in assessments and assignments. “high score” indicating highest score obtained by the student for assessments and assignments. “Moderate score” indicating moderate score obtained by the student for assessments and assignments. These example attributes can be considered individually or can be taken in combination for determining the second weightage value that indicates the behavioral aspect of the student overall.

[0056] The behavioral aspect of every student is captured in the table below:

Student Behavioral aspect Second weightage value
Student-1 Low time 0.7
Student-2 Low time 0.8
Student-3 High time 0.2
Student-4 Low time 0.7
Student-5 Low time 0.5

[0057] Upon determining the behavioral aspect of every student, the complexity level of the online course is determined. Such determination of the complexity level is performed by calculating an aggregate aspect of the online course. The aggregate aspect of the online course is obtained by calculating the average of the emotional aspect and the behavioral aspect associated with each student. In this example, based on the aggregate aspect of all the students, the complexity level of the Course X is said to be Moderately complex.
[0058] The complexity level of the online course is as below:
Student First weightage value Second weightage value Complexity level
Student-1 0.7 0.7

Moderately complex
Student-2 0.8 0.8
Student-3 0.2 0.2
Student-4 0.7 0.7
Student-5 0.1 0.5

[0059] Further, based on the emotional aspect, behavioral aspect and the complexity level of the online course, the engagement level of the student is determined. Engagement level denotes the level of engagement of the student on the online course. It provides information on interest of the student on the online course and it also provides a hint if the student can complete the online course successfully. Also, by determining how engaged the student is, the teacher can also make necessary modification in the way of teaching so that the engagement level of the student can be increased.
[0060] The engagement level calculated is shown as below:

Student First weightage value Second weightage value Complexity level Engagement level
Student-1 0.7 0.7

Moderately complex Highly engaged
Student-2 0.8 0.8 Highly engaged
Student-3 0.2 0.2 Not engaged
Student-4 0.7 0.7 Highly engaged
Student-5 0.1 0.5 Moderately engaged

[0061] Based on the above example, the emotional aspect of student 1 is happy and the behavioral aspect of student 1 is low time which means student 1 has consumed less time to finish the assessments or the assignments of course X. Also, it is calculated that the complexity level of the course is “moderately complex”. Based on all these three aspects, the engagement level of student 1 is highly engaged. Since the student 1 is highly engaged, certain actions can be performed to enable student 1 to continue being highly engaged. In one example, the action can include, upgrading the assessment or assignments to be a little more challenging so the student 1 can continue being highly engaged rather than getting bored. In another example, the action can include, recommendation of advanced courses to student 1.

[0062] Similarly in the above example, student 3 is not engaged. The emotional aspect of student 3 is sad and the behavioral aspect of student 3 is high time, since the student 3 is taking longer time to complete the assessments. Hence, the engagement level for student 3 is calculated to be not engaged. Here the actions can include, sending reminders to complete the assessments, asking questions to determine why the student is taking longer to complete the assessments or recommend another preliminary course for better understanding of the course X.

[0063] Again in the above example, the emotional aspect of the student 5 is confused and behavioral aspect is low time. Here although student 5 is confused. He is making an attempt to finish the assessments in less time. Hence, it is clear that student 5 is interested in the course. Here, the actions can include empathizing with the emotion of student 5 and accordingly sending motivating messages to motivate student 5 to complete the course X. Another action can also include, altering in teaching methodology to help student 5 alleviate confusion.

[0064] Advantageously, the disclosure enables to determine the engagement level of every student enrolling in an online course. By capturing emotion of the student, motivation can be provided when required and also appropriate recommendations can be provided to the students based on the engagement levels so that the online course can be completed successfully.

[0065] It should be understood that embodiments explained in the description above are only illustrative and do not limit the scope of this invention. Many such embodiments and other modifications and changes in the embodiment explained in the description are envisaged. The scope of the invention is only limited by the scope of the claims.
 
, Claims:Claims

1. A method to detect engagement level of a student taking an online course, said method comprising
determining (105) an emotional aspect of said student by analyzing facial expression of said student, said facial expression being captured using a camera of a computing device;
determining (110) a behavioral aspect of said student by monitoring one or more attributes corresponding to said online course;
determining (115) a complexity level of said online course based on an aggregate aspect of said online course, said aggregate aspect is an average of said emotional aspect and said behavioral aspect associated with each student; and
calculating (120) said engagement level of said student based on at least one of said emotional aspect of said student, said behavioral aspect of said student and said complexity level of said online course.

2. The method as claimed in claim 1, wherein said analysis of said facial expression is performed using machine learning models.

3. The method as claimed in claim 1, wherein said machine learning model is a Random Forest Model.

4. The method as claimed in claim 1, wherein said one or more attributes corresponds to at least one of a plurality of assignments, associated with said online course, attempted by said student, a plurality of assessments, associated with said online course, attempted by said student, time taken, by said student, to complete said online course and number of attempts, made by said student, for completing said online course.

5. The method as claimed in claim 1, wherein said monitoring of said one or more attributes is performed using machine learning models.

6. The method as claimed in claim 1 further comprising:
assigning a first weightage value for said emotional aspect; and
assigning a second weightage value for said behavioral aspect.

7. The method as claimed in claim 1, wherein said calculation of said engagement level is performed based on said first weightage value and second weightage level.

8. A device (200) to detect engagement level of a student in an online course, said device (200) comprising:
an input module (205) for receiving at least one of a facial expression of said student and one or more attributes of said student;
a control unit (210) configured to:
determine an emotional aspect of said student by analyzing said received facial expression;
determine a behavioral aspect of said student by monitoring said received one or more attributes corresponding to said online course;
determine a complexity level of said online course based on an aggregate aspect of said online course, said aggregate aspect is an average of said emotional aspect and said behavioral aspect associated with each student; and
calculate said engagement level of said student based on said emotional aspect, said behavioral aspect and said complexity level of said online course.

9. The device (200) as claimed in claim 9, wherein said control unit (210) is further configured to:
assign a first weightage value for said emotional aspect; and
assign a second weightage value for said behavioral aspect.

10. A system for detecting engagement level of a student in an online course:
a web application running on a plurality of servers (305);
a plurality of client devices (200a, 200b and 200c) that executes said web application, said plurality of client devices (200a, 200b and 200c) being configured to capture a plurality of facial expressions corresponding to a plurality of students and to capture a plurality of attributes corresponding to said plurality of students; and
a control unit (210) of said web application for:
determining an emotional aspect of said plurality of students based on said plurality of facial expressions;
determine a behavioral aspect of said plurality of students based on said plurality of attributes;
determining a complexity level of said online course based on an aggregate aspect of said online course, said aggregate aspect is an average of said emotional aspect and said behavioral aspect associated with each student of said plurality of students; and
calculating said engagement level of said student based on said emotional aspect, said behavioral aspect that correspond to each of said plurality students and said complexity level of said online course