International Journal of Interactive Mobile Technologies (iJIM) – eISSN: 1865-7923 – Vol 17 No 04 (2023) Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred Design https://doi.org/10.3991/ijim.v17i04.37805 A. Helen Victoria1(), P. Supraja1, V. M. Gayathri1, Mohamad Sukeri Khalid2, Nor Hissam B. Sulaiman3 1 Department of Networking and Communications, SRM Institute of Science and Technology, Chennai, India 2 School of Government, College of Law, Government and International Studies, Universiti Utara Malaysia, Malaysia 3 School of Creative Industry Management and Performing Arts (SCIMPA), Universiti Utara Malaysia, Kedah Darul Aman, Malaysia. helenvia@srmist.edu.in Abstract—Many colleges employ online learning as media learning where each piece of multimedia, be it sound, film, or text, can receive feedback from students. The lecture wants to know about the emotions that students experienced when they accessed the media, such as happiness, disappointment, or unhappiness, and instructors want to know how joyful they felt. This study built a tool cell for online media to use in identifying emotions in column comments. This paper focuses on user experience (UX) design, which is based on the Human-Centred Design method that prioritizes empathy for users and focuses on user demands to meet the discovery of user wants and develop high- fidelity prototypes. The results of the testing show that the average accuracy is 1.68%, the average recall is 1.55, the average precision is 1.45, and the average accuracy for the use of this phone website is 80% for emotion recognition based only on a column of comment in the internet media. Keywords—user-centred design, user experience, mobile, emotion detection, visualization, styles 1 Introduction All schools use online learning for at-home instruction during pandemics. For the online lesson, the instructor has produced a multimedia instruction using voice, videos, and papers. With their version, the professors have just invented a medium. They don't care if the media is suitable for influencing students pass or fail. Because the media lacks excitement, is dull, or is emotional, the impacts can occasionally be worse, which kids cannot comprehend. If the teacher gets student input after the pupils have watched or read the media, this problem can be remedied. Online comment sections for columns such as those on YouTube or virtual education platforms like SIPEJAR are where you can find remarks. iJIM ‒ Vol. 17, No. 04, 2023 21 https://doi.org/10.3991/ijim.v17i04.37805 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… On the other hand, effective online learning tools must be developed. Finding out what works and what doesn't in learning medium is one method to make one that works well. For instance, numerous learning platforms leverage YouTube channels, where viewers can comment on each video. Owners of movies can use these remarks as inspiration to produce other videos with quality material. There are other remarks in the online materials for education, as there is a commenting feature. The comments are in text form. It is impossible to edit a page because many users contribute text or images to the comments expressing their feelings or what they want to convey when they view films that internet resources for education. For particular goals, educators or media producers should understand the audience's responses, including their feelings or how much they learned in communication. Artificial intelligence techniques are used to process how text documents are transformed into an emotion based on the comments thread in online learning materials. Text mining is a form of artificial intelligence that is essential to the internet or internet learning and is increasingly used in mobile apps. To assist educators or video producers in raising the calibre of the videos that are produced based on feedback in the form of user emotion, AI must be integrated into the web or mobile platforms [1]. The availability of the monitoring system is increased by the usage of mobile applications. It is crucial to keep the user experience in mind when designing mobile hardware, software, and user interactions. Along with the quick adoption of mobile information systems in daily life, especially in the agriculture industry, the user experience (UX) perspective is receiving more attention from practitioners and scientists. UX is the whole of the consequences that the user experiences as an outcome of the interaction the context, in which they are used, the impact of accessibility, the emotional impact they experience while using the interplay, and the memory they develop after using it. Accessibility and personalization are two factors that influence the physical, artistic, and emotional components of user experience (UX). The process through which users shape their experience of using a product ever since they first met it is a key idea in UX, and it is translated in ISO 9241-210:2019 as the customer perceptions and reactions as a result of following the process, item, or application. Because it presents user abilities and constraints, UX design plays a crucial role in creating and developing apps. A UX design (in Figure 1) for a mobile-based horticulture field monitoring system is suggested in this study. The plan aspires to [2]: • Create applications that are affordable, accessible, and with acceptable usage; • Create software that can reduce negative emotions like irritation and maximise good ones like enjoyment and contentment. • The User-Cantered Design methodology is used in UX design to accommodate the identification of user demands. This methodology stresses empathy for consumers and concentrates on their needs. 22 http://www.i-jim.org Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… Fig. 1. User centred design process The following sections of the essay are arranged as follows. The research on the pertinent earlier work is presented in Section 2. The characteristics of the proposed system, including its proposed system architecture, implementation model, elements of the graph-based technique, and data analysis, are described in Section 3. In Section 4, the system's efficacy is evaluated and the implementation environment is explained. The solution is presented in Section 5. 2 Related works Abuaddous, H. Y., Saleh et.al [3] Heuristic assessment, cognitive walkthrough (CW), field research, and laboratory studies are used in conventional and moderated UX evaluation methodologies. Both desktop computer and mobile device applications can be evaluated using the conventional UX evaluation techniques, which are led by evaluators. However, current evaluation methodologies, particularly for applications on mobile devices, struggle to produce results that are both pertinent and statistically significant. This could be a reference to the fact that it is only possible to accurately simulate real-world uses in an experimental study. As a result, there aren't many different contexts and practical restrictions, and it's challenging to generalize to a fast- changing user situation that probably varies a lot. Bu, L., Chen, C. H., Ng et.al [5] the data layer consists of data gathering, interpretation, transfer, and storing. After gathering information using several data collection techniques, the intelligent VR platform chooses a flexible and secure transmission and engagement mechanism. It is possible to construct a secure, reliable, swift, and flexible data management platform with improved visualization and advanced analytical capabilities. To produce a real pan-simulated clip, the VR system must combine the graphic data. This is because graphic picture generation is a crucial component of VR technology. In light of this, a user-cantered VR software application that uses consumer data can be offered. Define Decide PrototypeEvolve Discover iJIM ‒ Vol. 17, No. 04, 2023 23 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… Santamaria-Granados, L., Mendoza-Moreno et.al [6] to first identify the elements of the emotion-based tourist recommender frameworks, a literature review was conducted. In order to recognize the user’s emotional state as a pertinent contextual aspect in the fulfilment of the suggestion, wearable devices' physical data must be integrated, but this study exposed a gap in that process. The methodologies that were studied mostly focused on sentiment analysis methods to identify emotional responses from comments on social media. Furthermore, these versions did not take into account cheap wearable’s that may be used to identify emotional trends in the user's everyday life. Nguyen, N et.al [7] the goal of the emotional design and construction was to gather user input that focused on their emotions, emotional impressions, and whether or not their emotions were favourable after their encounter. Two users who had previously used the programme and two people who were using it for the first time each participated in the test in March 2019. Users were encouraged to speak out loud while taking part in the test, particularly when they came across situations that captured their interest and caused them to feel certain emotions. Sun, X., Zhang, C., Ding et.al [8] the anomaly detection techniques are mostly dependent on textual, artificial neural, time-series data, analytics, models, and ranks; they call for a sizable amount of annotation but a sizable amount of annotation labor. Additionally, present techniques often categorize and analyze all the data on social media platforms to find breakouts or unusual events from a temporal perspective; however, little study has been done on the identification of unusual emotions for a specific user. Haoxiang, W et.al [9] various techniques including the use of social networking sites are used to analyse the sentiment classification in social media and applications. Every piece of social networking content is accessible in the user's preferred raw format. The necessary data is then transformed into knowledge for the subsequent stage of sentiment classification. The content on social media is examined using the viewpoint approach. Social media is being used by individuals more regularly these days, and they are uploading material in the form of photographs, information, music, and videos. All of them are regarded as raw data. Natural Language Processing (NLP) is used in sentiment classifications to separate that involves a sense from the provided text and determine whether it is favourable or unfavourable. 3 Methods and materials The primary goal of the method suggested in this paper is to identify emotions in inputs. In the suggested method, many processes are carried out on the input to prepare the final determination of the emotion type. Figure 2 depicts the block diagram of these processes sequentially fashioned to emphasize these operations and their impact. To better appreciate how each primary operation in the suggested strategy (Figure 3) contributes, more information is provided in the sections that follow [4]. 24 http://www.i-jim.org Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… Fig. 2. System for detecting emotions in a block diagram 3.1 Processing user-generated data First, wavelet transforms were used to extract the series of complex functions at each window scale. Secondly, each filtering time series' phase data was used to construct a number of connected phase oscillation models. Thirdly, a probability function is created using Bayesian theory. The posterior density, given a prior density, is equal to or greater than 𝑄𝑄𝑥𝑥 � 𝑚𝑚 𝑥𝑥 � = (�𝑚𝑚𝑥𝑥�𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃(𝑁𝑁) (�𝑚𝑚𝑥𝑥�𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃(𝑁𝑁)𝐸𝐸𝑁𝑁 (1) Dataset Gathering And Feature Extraction Standard Dataset Actual Data Control Audio Audio Records Say Report Emotion Detection Feature Selection Classifications Fury Hatred Happy Worry Blues iJIM ‒ Vol. 17, No. 04, 2023 25 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… The expression for the inverse log-likelihood function is: 𝑅𝑅 = 𝑀𝑀 1 𝑖𝑖𝑖𝑖(𝑒𝑒) + 𝑙𝑙 1 ∑(𝑑𝑑𝑑𝑑 𝜎𝜎(𝜃𝜃𝜃𝜃.𝑙𝑙) 𝜎𝜎𝜎𝜎 ) + [∅𝑛𝑛 − 𝑑𝑑𝑙𝑙∅𝑙𝑙(∅𝑚𝑚] (2) Where the relevant index is used in ∅ place of the dot index in and l is the sample step. The following equations can be used to find the e stationary point of iteratively depending on the connection parameters. 𝐸𝐸 = 𝑔𝑔 𝑚𝑚 (∅𝑛𝑛 − 𝑑𝑑𝑙𝑙∅𝑙𝑙(∅. 𝑑𝑑) (3) 𝑠𝑠𝑢𝑢 = �𝐹𝐹𝑝𝑝𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃�𝑑𝑑𝑢𝑢𝑑𝑑𝑢𝑢 + 𝑑𝑑∅(∅. 𝑖𝑖)(𝐸𝐸−1)∅𝑛𝑛 − 𝑙𝑙 1 𝜕𝜕∅(∅.𝑛𝑛) 𝜕𝜕∅ (4) 𝐹𝐹𝑙𝑙𝑢𝑢 = �𝐹𝐹𝑝𝑝𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃� + 𝑑𝑑∅𝑙𝑙(∅. 𝑖𝑖)(𝐸𝐸−1)∅𝑛𝑛 (5) 𝐷𝐷𝑙𝑙 = (𝐹𝐹−1)𝑑𝑑𝑙𝑙𝑠𝑠𝑙𝑙 (6) The connection strength was determined to be the Geometric norm of the inferred oscillation parameters, where D- and -C. To evaluate the efficiency of the 𝑉𝑉𝑗𝑗values, 125 surrogate signals in total were developed. The Transform was used to modify the brightness of each channel [5]. The following is the filter bank approach used in this Virtual system's signal processing: • According to Kirchoff rules, the following is the mathematical expression for input voltage D and output voltage D: 𝑉𝑉𝑗𝑗 = 𝑆𝑆𝐷𝐷 � 𝑒𝑒𝑒𝑒0 𝑒𝑒𝑒𝑒 � + 𝑉𝑉0 (7) • The Lagrange transposition forms 𝐿𝐿(𝑅𝑅)are as follows in light of the preceding fractional derivative: 𝐿𝐿(𝑅𝑅) = 2/(𝑆𝑆𝑆𝑆𝑅𝑅 + 2) (8) • The following is the connection between the Complicated variable M and the N translation operator in a Fourier transposition: 𝑅𝑅 = 2 − 𝑌𝑌 2 𝑀𝑀� (9) • where Q denotes the sampling frequency and JN represents the N-transform function following the M replacement as follows: 𝐿𝐿(𝑌𝑌) = 𝑀𝑀 𝑆𝑆𝑆𝑆(1−𝑌𝑌−2) + 𝑀𝑀) (10) • From the 𝑍𝑍(𝑀𝑀) equation, the following is the differential equation of the output 𝑌𝑌(𝑚𝑚)and the input 𝑋𝑋(𝑚𝑚 − 2): 𝑍𝑍(𝑀𝑀) = 𝑀𝑀 𝑀𝑀+𝑆𝑆𝑆𝑆 𝑌𝑌(𝑚𝑚) + 𝑆𝑆𝑆𝑆 𝑀𝑀+𝑆𝑆𝑆𝑆 𝑋𝑋(𝑚𝑚 − 2) (11) 26 http://www.i-jim.org Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… • The difference equation can be expressed as follows using the assumption that Y = Z WX and I is the sampling interval: 𝑍𝑍(𝑀𝑀) = 𝑄𝑄𝑌𝑌(𝑀𝑀) + (2 − 𝑦𝑦)𝑋𝑋(𝑚𝑚 − 2) (12) Sampling rate is: 𝑔𝑔1 = 𝑌𝑌 3∗𝑞𝑞𝑗𝑗∗𝑚𝑚 (13) 3.2 Emotional state detection The Multi-class imbalance that results from the gathering of user datasets can have an impact on the effectiveness and effectiveness of the prediction models. Due to the participants' varied affective actions in their environments, the aggregated dataset for this study exhibited an unbalanced distribution in the emotion classes. The subjective states of contentment, calmness, exhaustion, and gladness were also predisposed in the subjects. For the Multi-class Imbalanced Classification (MIC) with neural networks, various studies have used opportunistic sampling approaches and oversampling strategies. These sampling methods are based on the subspace of each class's closest neighbour rule. The data balance component sizes the dataset for the aforementioned and modifies the label names according to emotional states or quadrants. To assess the effectiveness of affective detection methods in Figure 4, it also employs class balance techniques [6]. ALGORITHM: Emotional slicing 1. method buildSlices 2. sliceLimit = 25, sliceSize = 35, timeBetweenInstances = 65, and slicesCounter = 105 3. previous = tagHrList[0]; tagHrList = dataLoad(hrData); 4. initSlicee(); 5. addInstToSlice(); 6. getMinimumMaximumByImei(); 7. range(1, len(tagHrList) do for row 8. tagHrList[row 1]; previous 9. tagHrList[row]; current 10. If past[0] = present[0], then 11. If timeBetweenInstances = int(current[0]) - int(previous[0]) then 12. When current[4] = 0 movie0 13. If prior[3], subsequent[3], then 14. closeeSlice(); 15. close if 16. addInstanceToSlicee(); 17. new 18. addInstanceToSlicee(); iJIM ‒ Vol. 17, No. 04, 2023 27 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… 19. close if 20. new 21. closeeSlice(); 22. addInstanceToSlicee(); 23. close if 24. new 25. closeSlicee(); 26. addInstanceToSlicee(); 27. close if 28. close for 29. close process Fig. 3. Example and surroundings 3.3 User-centered design The standard placed a strong emphasis on the users' participation and degree of co- creation during the system design. Understanding the users' backgrounds, environments, issues, requirements, emotions, and expectations for the effectiveness of a proposed interactive system is the first rule. The criteria for the software architecture will only be explicitly obvious when developers and the project leader interact and have a thorough understanding of people. For instance, several users of various ages and gender identities will use the system to assist them for various purposes, and only that particular group of users can contribute information regarding the system's design specifications. To ensure that the system offers users genuine worth and meaning, the second counsel’s organisations to involve users in the co- Use Person Emotion Recogntion Affective Dtection Based On Data Recommend Scheme Engenders On Emotion Datas Positions Recommend List Grow Operator Outline And Statistics Contexual 28 http://www.i-jim.org Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… creation process of the development and design. The third element, which is about architecture and is led and determined by usability tests, resulted from this action [7]. In a conclusion, people for users assess and enhance the program's architecture, quality, and usability. The idea of performing iterative design rounds, beginning with the initial conception, and continuing to enhance, modifies, and innovate it throughout the design process, is described in the founding principles. It is critical for the designers and developers' team to present people with graphical concepts and solicit input from them regarding these design elements in order to be user centric. The ones that consumers select based on their preferences and usefulness for their duties will then go to the next phase of the design process. The essential value of the system is user satisfaction, according to the fundamental criterion. The design team must consider issues such as how to make people utilize the online system as a part of their lives seem simple, pleasant, great, happy, enjoyable, useful, efficient, and supportive. For instance, the design team can constantly imagine a situation or user journey and the interaction system in a connection to determine how to create a successful and pleasurable connection. 3.4 Design of a visualisation model The elements of the visualization model are displayed in Figure 4. Along with another server-side codebase, it operates on the server. It is made up of an interfaces definer, a contextual manager, an OWL user model that has already been developed, and a native RDF store. The ontology-driven user model is managed by the interfaces definer, which also retrieves the HMI specification for the distribution module. Fig. 4. Visualization for emotion Detection The service keeps a copy of the OWL user model that was first presented in section 3.4 as templates. A particular user model will be created for the new user at the moment the different user accounts are entered into the system by the HMI definer, which loads the OWL file. The server's native RDF database has this model, and TDB is used to save it there. The distinction is that something that built on a Database server, allowing the user model to be stored on a distant database server. Both the two storages were evaluated during the design process [10]. The results indicated that implies that organizations offer faster speeds for storing the user model and retrieving the model than does the Downloads folder in the configuration. The contextual Participation Descriptions Expression Detection Feature Abstraction Production Imageries Emotion Recognition Appearance Analysis iJIM ‒ Vol. 17, No. 04, 2023 29 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… manager is in charge of changing the user model during runtime by the contextual data. The user model already has all the contextual information, but it has to be modified to reflect the current scenario. Create the appropriate Display description for the circumstance. The UCD is a part of the collaborative innovation (PD) approaches, as suggested in this paper, and it emphasises on users' active participation in the design process and ongoing observations while designing. UCD makes ensuring that users are included in the designing process. Customers have the chance to directly influence the design of a product based on their requirements. The report must also define web accessibility from the very beginning [12]. This strategy was based on a series of previous initiatives to clarify crucial aspects of a user-centered strategy, such as the work on assuring usability by Cutler and others [13]. Fig. 5. Visualization mode's operational diagram 4 Experimental and implementation results The five different types of micro blogs that users create each month can be thought of as five-dimensional matrices, where the data for each dimension is examined and confirmed to see if it is susceptible to Gaussian kernel or not. Var1-Var4 represents the four variables "calm, pleased, startled, upset, and furious" according to the K-S test, as illustrated in Tables 2 and 3 The P-value is what is known as Symposia. (Bilateral), and it is often set to a threshold level of 0.04. The K-S test was independently performed for both solo and corporate users, and the results are displayed in the following tables [8]: Deployment Of Device Interfaces And Communication OWL Database Server Contexual manager Database Attendant Visualization Method 30 http://www.i-jim.org Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… Table 1. Evaluation of the four techniques' accuracy in comparison TECHNIQUE ACCURATENESS Matrix factorization that is not negative 52.00% Real-time observation 74.32% Identifying social spammers in micro blogging 86.10% Gaussian multivariate model 88.85% The K-S tests for a specific user are shown in Table 2. As can be observed, the "calm, pleased, and upset" kinds have P-values greater than 0.05, supporting the previous assumption. However, the "surprised" and "mad" emotions do not follow a normal distribution because they are characterized by outbursts. Table 3 displays the K-S test results for each form of personal blog for the sample (3000 users). Table 2. K-S test for a specific user N VAR-1 VAR-2 VAR-3 VAR-4 Uncaring 1.2 2.12 3.1 1.12 Most Complete 1.4 2.10 3.2 1.13 Very Positive 1.6 2.8 3.6 2.5 Distinction Negative -1.8 -2.6 -3.8 -3.9 Kolmogorov-SmirnovZ 1.10 2.4 3.10 1.25 Asymp.Sign(bilateral) 1.12 2.2 3.12 2.16 Since the P-value is less than 0.04, it is clear that the group's emotional state on the micro blog is not governed by a normally distributed. The group's attitude on micro blogs tends to satisfy the simple linear distribution, a common social phenomenon also known as the long-tail dispersion, which is exponentially distributed. Table 3. Group K-S test N VAR-1 VAR-2 VAR-3 VAR-4 Uncaring 3.2 3.5 2.2 3.2 Most Complete 1.2 1.6 2.3 3.6 Very Positive 5.6 5.3 2.6 3.4 Distinction Negative -2.1 -2.3 -1.6 -3.9 Kolmogorov-SmirnovZ 3.6 1.2 5.6 2.3 Asymp.Sign(bilateral) 0.00 0.00 0.00 0.00 The data can be considered to be regularly distributed if the P-value is greater than 0.04; otherwise, the original hypothesis (that the data are subject to normal distribution) is denied. In this study, we undertake an experiment to examine this conclusion, using x as the abscissa, which is the total number of micro blogs created by all users in a month, and ordinate, which is the rate of Y. Applications like Mat lab can be used to fit these data. iJIM ‒ Vol. 17, No. 04, 2023 31 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… A collection of the data's initial distribution is shown in Figure 6. The form of the long-tailed distributions is reflected in the raw data. The log data, which roughly resembles a straight line after the row data has been logarithm zed, is produced. This is demonstrated in Figure 7. To determine whether the distribution is accurate, the residue sum is utilized; the smaller the residual sum, the more likely it is that the data set's simple linear distribution may be taken into account. Fig. 6. Original Data Distribution These datasets include subjective user opinions that might offer detailed information about quality, such as how the users feel about the services that have been implemented, which can be used to look into customer satisfaction and inclinations. However, user feedback also poses challenges because unorganized texts can't be used directly for quality acquisition without textual data [11]. Fig. 7. Logarithmic data distribution 0.7, 2.7 1.8, 3.2 2.6, 0.8 0 0.5 1 1.5 2 2.5 3 3.5 0 0.5 1 1.5 2 2.5 3 RAW… 0.7, 2.7 1.8, 3.2 2.6, 0.8 0 0.5 1 1.5 2 2.5 3 3.5 0 0.5 1 1.5 2 2.5 3 LOG… 32 http://www.i-jim.org Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… The outstanding sum of the instance in Figure 8, as can be observed from Figure 7, indicates that the circumstance roughly follows the power law distribution. This experiment tests the monthly distributions of all the users'[9] micro blogs and demonstrates that the power law distribution governs the distribution of a group's emotions on micro blogs. Fig. 8. Sum remaining in the case This study investigated a model for user emotion and abnormality identification on social media under the presumption that the user's multidimensional emotion follows the regular structure. Through experiments and the examination of actual data, the following conclusions are also reached: • Micro blogs' "negative, joyful, and sad" feelings are based on normally distributed. • In micro blogs, strong emotions like "shocked" and "mad" don't follow the usual distribution. • On micro blogs, group emotions try to satisfy a different index distribution called the simple linear distribution rather than the normally distributed. 5 Conclusion Many academics have been working on emotion detection in various disciplines, including social networks, in recent years. This paper covered the significance of social network emotion detection. This study uses a set of guidelines for a message CO-EFFICIENT 55 57 57 58 58 INTERCEPT 11 12 13 11 35 RESIDUAL SUM 32 35 34 35 43 0 10 20 30 40 50 60 70 O U T ST A N D IN G CIRCUMSTANCE NUMBER RESIDUAL CASE ORDER PLOT iJIM ‒ Vol. 17, No. 04, 2023 33 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… emotion class type and k-NN to determine each student's feelings based solely on user comments and online learning resources. A proposed prototype for the user experience design for the visualisation of a mobile-based land monitoring system. Based on the UCD, a prototype interface design is created. The results of the experiment indicate were given an application with a very high level of usefulness. All of the prototype design's UX elements scored well above average in terms of user experience evaluation. Our findings indicate that a person's level of depression may be significantly predicted by using sentiment analysis, emotions, and negative phrases in a statement. Terms that convey negative emotions, such as "regretfully," "afraid," and "horrific," as well as words that denote a desire to die, such as "start dying," "life,” and “self-harm,” are frequently employed. Using optimization models inspired by nature, this research could be refined further to increase system accuracy. 6 References [1] Wahyono, I. D., Saryono, D., Putranto, H., Asfani, K., Rosyid, H. A., Sunarti, M. M. M., & Shih, J. S. (2022). Emotion Detection based on Column Comments in Material of Online Learning using Artificial Intelligence. iJIM, 16(03), 83. https://doi.org/10.3991/ ijim.v16i03.28963 [2] Setiyawati, N., Purnomo, H. D., & Mailoa, E. (2022). User Experience Design on Visualization of Mobile-Based Land Monitoring System Using a User-Centered Design Approach. International Journal of Interactive Mobile Technologies, 16(3). https://doi.org/ 10.3991/ijim.v16i03.28499 [3] Abuaddous, H. Y., Saleh, A. M., Enaizan, O., Ghabban, F., & Al-Badareen, A. B. (2022). Automated User Experience (UX) Testing for Mobile Application: Strengths and Limitations. International Journal of Interactive Mobile Technologies, 16(4). https://doi. org/10.3991/ijim.v16i04.26471 [4] Al-Saedi, L. M., & Al-Jobouri, L. (2018, September). Proposed Approach of Detecting Facial Emotion using Neural Network and Representational of HOG Features. In 2018 10th Computer Science and Electronic Engineering (CEEC) (pp. 276-281). IEEE. https://doi.org/10.1109/CEEC.2018.8674197 [5] Bu, L., Chen, C. H., Ng, K. K., Zheng, P., Dong, G., & Liu, H. (2021). A user-centric design approach for smart product-service systems using virtual reality: A case study. Journal of Cleaner Production, 280, 124413. https://doi.org/10.1016/j.jclepro.2020. 124413 [6] Santamaria-Granados, L., Mendoza-Moreno, J. F., Chantre-Astaiza, A., Munoz-Organero, M., & Ramirez-Gonzalez, G. (2021). Tourist Experiences Recommender System Based on Emotion Recognition with Wearable Data. Sensors, 21(23), 7854. https://doi.org/10.3390/ s21237854 [7] Nguyen, N. (2019). User-centred Mobile Augmented Reality application’s concept evaluation. Case: Freshman App. [8] Sun, X., Zhang, C., Ding, S., & Quan, C. (2018). Detecting anomalous emotion through big data from social networks based on a deep learning method. Multimedia Tools Appl, 5, 1-22. [9] Haoxiang, W. (2020). Emotional analysis of bogus statistics in social media. Journal of Ubiquitous Computing and Communication Technologies (UCCT), 2(03), 178-186. https://doi.org/10.36548/jucct.2020.3.006 34 http://www.i-jim.org https://doi.org/10.3991/ijim.v16i03.28963 https://doi.org/10.3991/ijim.v16i03.28963 https://doi.org/10.3991/ijim.v16i03.28499 https://doi.org/10.3991/ijim.v16i03.28499 https://doi.org/10.3991/ijim.v16i04.26471 https://doi.org/10.3991/ijim.v16i04.26471 https://doi.org/10.1109/CEEC.2018.8674197 https://doi.org/10.1016/j.jclepro.2020.124413 https://doi.org/10.1016/j.jclepro.2020.124413 https://doi.org/10.3390/s21237854 https://doi.org/10.3390/s21237854 https://doi.org/10.36548/jucct.2020.3.006 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… [10] Xu, X. (2014). An ontology-driven visualization model for production monitoring systems (Master's thesis). [11] Hu, K., Gui, Z., Cheng, X., Wu, H., & McClure, S. C. (2019). The concept and technologies of quality of geographic information service: improving user experience of GIServices in a distributed computing environment. ISPRS International Journal of Geo- Information, 8(3), 118. https://doi.org/10.3390/ijgi8030118 [12] Miah, S. J. (2004, June). Accessibility improvement of multicultural educational web interface by using the User Centred Design (UCD) approach. In Proceedings of the 2004 Informing Science and IT Education Joint Conference (pp. 25-28). Informing Science Institute. [13] Marti, P., & Bannon, L. J. (2009). Exploring user-centred design in practice: Some caveats. Knowledge, technology & policy, 22(1), 7-15. https://doi.org/10.1007/s12130- 009-9062-3 7 Authors Dr. A. Helen Victoria is an Assistant Professor in SRM Institute of Science and Technology. Her Research Interests are Deep Learning, Machine Learning, Radar Signal Applications, Optimization Techniques. She has published papers in reputed journals and conferences (email: helenvia@srmist.edu.in). Dr. P. Supraja is Currently working as Associate Professor, Department of Networking and Communications in SRM Institute of Science and Technology Kattankulathur, India. She was a recipient of AICTE Visvesvaraya Best Teacher Award 2020. Previously she completed the Indo-US WISTEMM Research fellowship at University of Southern California, Los Angeles, USA funded by IUSSTF and DST Govt of India and she served as a Post-Doctoral Research Associate at Northumbria University, Newcastle, UK and completed her PhD from Anna University in 2017. She has published more than 50 research papers in reputed national and international level journals/conferences. She received her university-level Best Research Paper Award in 2022 & 2019 also she has received funding from AICTE for conducting STTP. Her research interests include Cognitive Computing, Optimization algorithms, Machine learning, Deep Learning, Wireless Communication, and IoT. She is a reviewer in IEEE, Inderscience, Elsevier and Springer Journals. She is also a member of several national and international professional bodies including IEEE, ACM, ISTE, etc. In addition, she has received the young women in Engineering award and Distinguished Young Researcher award from various International Organizations (email: suprajap@srmist.edu.in). Dr. V. M. Gayathri is currently working as an Assistant Professor in SRM Institute of science and technology in the department of Networking and Communications from June 2017. She completed her Ph.D., in Saveetha Institute of Medical and Technical Sciences in the year 2019 on Ad-hoc Network security. She received her Master’s Degree in the field of Computer Science and Engineering from B.S.Abdur Rahman Crescent University. She earned her Bachelor’s Degree in the field of Computer Science from Anna University. Her Areas of Interests are Wireless iJIM ‒ Vol. 17, No. 04, 2023 35 https://doi.org/10.3390/ijgi8030118 https://doi.org/10.1007/s12130-009-9062-3 https://doi.org/10.1007/s12130-009-9062-3 Paper—Visualization for Emotion Detection in Mobile-Based Land Monitoring Using User-Centred… Communication, Network Security, Deep Learning and Artificial Intelligence (email: gayathrm@srmist.edu.in). Mohamad Sukeri Khalid obtained his doctoral degree from the Heriot-Watt University, Edinburgh UK. Currently, he is a Senior Lecturer at the School of Government, College of Law, Government and International Studies, Universiti Utara Malaysia, Sintok Kedah, Malaysia. His research interest is directed more on property economics, land development, property development, strata management community and construction management. (email: sukeri@uum.edu.my) Nor Hissam B. Sulaiman is a Lecturer at School of Creative Industry Management and Performing Arts (SCIMPA), College of Arts and Sciences (CAS), Universiti Utara Malaysia since 2003, and currently the professional members of Malaysian Board of Technologist (MBOT), specialist in Telecommunication, Mass Media, and Broadcasting. He is also an Associate Member of the Film Directors Association of Malaysia (FDAM). His major research interest focusing on the area such as audience identification, media production, information communication technology (ICT), media law and ethic, and self-regulation of media content. (email: norhissam@uum.edu.my). Article submitted 2022-11-02. Resubmitted 2023-01-05. Final acceptance 2023-01-07. Final version published as submitted by the authors. 36 http://www.i-jim.org