4.2.4. Analysis of Eye Movement Hot Zone

The teaching contents of the system were divided into four sessions for analysis and comparison of hot zones to understand the hot zones of learners' attention to the contents of the textbook in the course of study. Figures 5 and 6 are the first session of the course contents of the control group and the experiment group. Figures 7 and 8 are the hot zones for the course contents of the control group and the experiment group in the second session. Figures 9 and 10 are the course contents of the control group and the experiment group in the third session of the fixation hot zone. Figures 11 and 12 show the course contents in the fourth session of the control group and the experiment group. The experiment shows that both the control group and the experiment group were completely fixated on the course contents in the course content fixation hot zone. *Sustainability* **2022**, *14*, x FOR PEER REVIEW 13 of 18

**Figure 5.** Control group (session 1). **Figure 5.** Control group (session 1).

**Figure 6.** Experiment group (session 1).

**Figure 7.** Control group (session 2).

**Figure 6.** Experiment group (session 1). **Figure 6.** Experiment group (session 1). **Figure 6.** Experiment group (session 1).

**Figure 5.** Control group (session 1).

**Figure 5.** Control group (session 1).

*Sustainability* **2022**, *14*, x FOR PEER REVIEW 13 of 18

**Figure 7.** Control group (session 2). **Figure 7. Figure 7.** Control group (session 2). Control group (session 2).

**Figure 8.** Experiment group (session 2). **Figure 8.** Experiment group (session 2).

**Figure 9.** Control group (session 3).

**Figure 10.** Experiment group (session 3).

**Figure 9.** Control group (session 3). **Figure 9.** Control group (session 3). **Figure 9.** Control group (session 3).

**Figure 8.** Experiment group (session 2).

**Figure 8.** Experiment group (session 2).

*Sustainability* **2022**, *14*, x FOR PEER REVIEW 14 of 18

**Figure 10.** Experiment group (session 3). **Figure 10.** Experiment group (session 3). **Figure 10.** Experiment group (session 3).

In this study, the affective computing was built on the intelligent tutoring system, which not only enhanced the learning interest of the users, but also provided a deeper learning experience. The real-time interaction between the users and the emotional agents provided emotional feedback and guidance that makes the users turn negative emotions into positive ones during the learning process and improve the learning interest of the users. SUS was used to understand the usability of the system, and eye movement statistics were analyzed to explore the fixation duration at the system. The analysis shows that the average score of the users for the usability of the system is 81.5, and the overall satisfaction is very high. As shown in the descriptive statistics, the users found it easy to use the system and the learning process was attractive, which increased the willingness to learn. Therefore, the statistics show that the design of the system can be adopted, and its usability is quite good. In the analysis of the eye movement statistics, the use of affective tutoring system increased the learning duration of the course. The eye movement statistics show that the fixation duration of the users from sessions one to four of the learning and the length and position of the fixation duration could represent the distribution and preference of personal attention. In this study, the fixation duration of the experiment group was longer than that of the control group, and the eye movement statistics analysis divided the experiment group into the two ROI blocks, the agent block and the course block,

**Figure 11.** Control group (session 4). **Figure 11.** Control group (session 4).

**Figure 12.** Experiment group (session 4).

**5. Discussion and Conclusions** 

**Figure 12.** Experiment group (session 4). **Figure 12.** Experiment group (session 4).

**Figure 11.** Control group (session 4).

### **5. Discussion and Conclusions**

**5. Discussion and Conclusions**  In this study, the affective computing was built on the intelligent tutoring system, which not only enhanced the learning interest of the users, but also provided a deeper learning experience. The real-time interaction between the users and the emotional agents provided emotional feedback and guidance that makes the users turn negative emotions into positive ones during the learning process and improve the learning interest of the users. SUS was used to understand the usability of the system, and eye movement statistics were analyzed to explore the fixation duration at the system. The analysis shows that the average score of the users for the usability of the system is 81.5, and the overall satisfaction is very high. As shown in the descriptive statistics, the users found it easy to use the system and the learning process was attractive, which increased the willingness to learn. Therefore, the statistics show that the design of the system can be adopted, and its usability is quite good. In the analysis of the eye movement statistics, the use of affective tutoring system increased the learning duration of the course. The eye movement statistics show that the fixation duration of the users from sessions one to four of the learning and the length and position of the fixation duration could represent the distribution and preference of personal attention. In this study, the fixation duration of the experiment group was longer than that of the control group, and the eye movement statistics analysis divided the experiment group into the two ROI blocks, the agent block and the course block, In this study, the affective computing was built on the intelligent tutoring system, which not only enhanced the learning interest of the users, but also provided a deeper learning experience. The real-time interaction between the users and the emotional agents provided emotional feedback and guidance that makes the users turn negative emotions into positive ones during the learning process and improve the learning interest of the users. SUS was used to understand the usability of the system, and eye movement statistics were analyzed to explore the fixation duration at the system. The analysis shows that the average score of the users for the usability of the system is 81.5, and the overall satisfaction is very high. As shown in the descriptive statistics, the users found it easy to use the system and the learning process was attractive, which increased the willingness to learn. Therefore, the statistics show that the design of the system can be adopted, and its usability is quite good. In the analysis of the eye movement statistics, the use of affective tutoring system increased the learning duration of the course. The eye movement statistics show that the fixation duration of the users from sessions one to four of the learning and the length and position of the fixation duration could represent the distribution and preference of personal attention. In this study, the fixation duration of the experiment group was longer than that of the control group, and the eye movement statistics analysis divided the experiment group into the two ROI blocks, the agent block and the course block, while there was only the course block in the ROI block of the control group. To further confirm whether the subjects are interested in the course contents, remove the fixation duration of the agent block in the experiment group and intercept the course block of both the experiment group and the control group to compare the course learning duration. The statistics show that the total learning duration of the course contents of the experiment group was better than that of the control group. The research proved that the system can increase the learning duration of users' courses. Through our experiments, the following items were found: (1) For the semantic emotion recognition module, the system was unable to recognize some fashionable terms used by young people according to the users' feedback. It was expected to collect some popular social platform, advertising works and even some emojis that can express emotions. (2) For the sentiment analysis and judgment of words and sentences, more considerations are needed because, for some popular words used by young people, the meaning is usually not the meaning of emotions, and there is some irony or implied meaning. It is easy to be confused with the original words. It is expected that, in the future, the system will be more fashionable to match social trends and be closer to the users. (3) For the technology of the agent module, it is expected to perform deep learning to train the machine to become a ChatBot, which can interact with users more smoothly and naturally. For this part, we have planned them into the work for the future.

### **6. Future Prospects**

In recent years, the intelligent tutoring system has integrated users' emotional state into the intelligent tutoring system, which has led to the transformation of the intelligent tutoring system to the affective tutoring system. Therefore, affective computing is undoubtedly important in the education environment. In the future, multi-media contents can be applied in the design and development of the affective tutoring system. In addition, we expect to apply deep learning to the affective tutoring system and make the system feature a ChatBot message function. Message function has become an indispensable part of contemporary daily life. Under this premise, it is an inevitable trend to be equipped with a good ChatBot. Chatbots are no longer seen as mere assistants, for they interact in a way that brings them closer to the users as friendly companions [42]. Based on this, we will explore the user experience and evaluate the learning effectiveness in the subsequent study.

**Author Contributions:** Conceptualization, H.-C.K.L. and Y.-C.L.; methodology, H.-C.K.L. and Y.-C.L.; software, H.-C.K.L.; formal analysis, Y.-C.L. and H.-T.W.; writing—original draft preparation, Y.-C.L. and H.-T.W.; writing—review and editing, H.-C.K.L., Y.-C.L. and H.-T.W.; data collection, Y.-C.L. and H.-T.W. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

### **References**

