1. Introduction
For academic intervention, data mining is a powerful method. Educational Machine Learning is the term for research in the context of education. Educational Data Mining is concerned with the creation of new methods for discovering information from educational databases and applying it to educational decision-making.
It is important to obtain an education. No one should doubt its importance since it is a globally accepted phenomenon. It has an effect on people’s lives. A well- educated person can weather the challenges of life. People’s conduct is improved by education.
COVID-19’s global spread contributed in the cancellation of classes for over 850 million students around the world, undermining schools’ initial teaching plans in these countries and regions.
Soon later, many countries started to offer online teaching to students by:
1) Zoom.
2) Skype.
3) MS teams
4) YouTube
5) Google meeting
6) Teams
7) University LMS
8) Face Time etc.
On 14 March, the Ministry of Education of the People’s Republic of Pakistan an- announced that it will strongly believe in and promote information-based education and teaching, as well as expand the platform’s service ability to support online teaching, in order to encourage online education and restore normal teaching order. As a result of the epidemic’s spread, the online classroom has become a critical tool for staying up to date.
On the other hand, these online education outlets have problems like sys- tem interruptions and the inability to replay live broadcasts. It’s critical to deter- mine whether these network education systems can meet the needs of teachers and students, whether network teaching can effectively complete teaching assignments, and whether network education may become a successful form of special period education, and to make recommendations based on the results to aid the development of network eduction.
Scholars from various countries are currently collaborating to develop an online education platform satisfaction assessment system that employs the analytic hierarchy method (AHP) and incorporates various models and techniques.
Wilbur
| [1] | K. Mouthami, K. N. Devi, V. M. Bhaskaran, Sentiment analysis and classifica- tion based on textual reviews, in: 2013 international conference on Information communication and embedded systems (ICICES), IEEE, 2013, pp. 271-276. |
[1]
, For example, he evaluated the online portion of a textured degree program for practitioners using a formal self-assessment and peer review using an instrument systematically designed according to Moore’s principles of transaction’s distance, and he discovered that tweaking a few course elements improved the layout, conversation, and autonomy of student learning.
A virtual reality tour-guiding network was created, and some students from a Pakistani technical university took part in the research
| [2] | G. Vinodhini, R. Chandrasekaran, Sentiment analysis and opinion mining: a sur- vey, International Journal 2 (6) (2012) 282-292. |
| [3] | V. Dhanalakshmi, D. Bino, A. M. Saravanan, Opinion mining from student feed- back data using supervised learning algorithms, in: 2016 3rd MEC international conference on big data and smart city (ICBDSC), IEEE, 2016, pp. 1-5. |
[2, 3]
. The results revealed their learning efficacy and acceptance of technology in the educational sys- tem.
This graph represents that how much the occurrence of online study in the next years with different platforms, which perform better.
Figure 1. Yearly chart of online platform.
Python is quickly becoming one of the most widely used programming languages in scientific computing. It is an appealing option for algorithmic creation and exploratory data analysis because of its high-level interactive design and maturing ecosystem of science libraries we are doing our research in Python using data mining, and then we will look at various algorithms. These algorithms can also predict which platform is the best and which platform is the most common.
The aim of this project is to present the best platform based on various algorithm researches that specifically describe the adoption of online platforms to solve problems related to acquiring information and completing or continuing their education. The research examines the most recent state-of-the-art research on pro- viding quality education across online platforms.
The classification of online platforms and the problems that these platforms cause are discussed in this paper to assist developers. We also use the algorithms to predict which platform will work best with the others. This classification can be used by developers to choose online platforms based on the setting and context of use. Lastly, research gaps have been identified for future researches.
1.1. Problem Statement
Due to COVID-19 Pandemic in Pakistan every university and collage in- traduces online classes platform, but they don’t know whether it is benefiting or not? were are the students having problems ? they don’t know.
1.2. Research Questions
1) What are the factors that affect the Students during online classes?
2) Which online factors are most convenient for students during online classes?
3) Which online platform is most suitable for online classes?
1.3. Aim and Objective
1) To identify the factors that affect student during online classes.
2) To identify online factors that are suitable for student online classes.
3) To recognize online platform for the better learning of students during online classes.
2. Literature Review
Opinions are a person’s opinions, perceptions, emotions, or feelings, and opinion analysis is a method of exploring and evaluating certain feelings about something (e.g. a product, organization, service, etc.)
| [1] | K. Mouthami, K. N. Devi, V. M. Bhaskaran, Sentiment analysis and classifica- tion based on textual reviews, in: 2013 international conference on Information communication and embedded systems (ICICES), IEEE, 2013, pp. 271-276. |
[1]
.
The study of sentiment/opinion analysis is, by the day, thanks to the increased use of social networking sites such as Facebook and Twitter, which enable people to freely express, and share their opinions with others on any subject. People usually express their opinions in a variety of ways. In certain cases, a term that is considered positive in one situation can be considered negative in another.
| [3] | V. Dhanalakshmi, D. Bino, A. M. Saravanan, Opinion mining from student feed- back data using supervised learning algorithms, in: 2016 3rd MEC international conference on big data and smart city (ICBDSC), IEEE, 2016, pp. 1-5. |
[3]
.
Opinion mining (OM) is one such development in the field of text mining that can be used to assess the polarity of opinion from a large variety of unstructured text data datasets. Researchers have used this approach to identify opinions in a variety of applications, including e-commerce, movie reviews, and product reviews.
| [2] | G. Vinodhini, R. Chandrasekaran, Sentiment analysis and opinion mining: a sur- vey, International Journal 2 (6) (2012) 282-292. |
[2]
.
The researcher
| [4] | K. Vlasenko, S. Volkov, I. Sitak, I. Lovianova, D. Bobyliev, Usability analysis of on-line educational courses on the platform “higher school mathematics teacher”, in: E 3S Web of Conferences, Vol. 166, EDP Sciences, 2020, p. 10012. |
[4]
in this report. The need of investigating the usability of educational software as a means of developing distance education stems from the modern Internet education market’s growth. This generally requires specific criteria for the accessibility of online courses.
We used the Predictive Content Analysis Method to study the current papers on the requirements for usability design. This approach also assisted in defining the educational platform’s actual usability requirements, as well as providing an inference about the need to consider the parameter motivated by the existence and successful use of portable devices.
We created a survey for teachers and students who use online courses to help explain the hypothesis. The survey results were analyzed in two ways: to obtain.
informative statistics of online course users and to investigate the relative value of assessing educational platform usability categories. This method of surveying allowed significant information about the preferences of online course users to be gathered, which should be taken into account during the course’s growth.
As a result of the findings, we believe it is worthwhile to include the criterion Responsiveness, which represents the usability of mobile devices for online education. As a result of the findings of study and surveys, we present the following order of usability requirements in descending order:
1) The quality of the data (IQ),
2) Navigation of the System (SN),
3) Learnability of the System (SL),
4) Adaptability (RS),
5) Assessment of Instruction (IA),
6) Interactivity of the System (SI).
Usability requirements will be the focus of future study. An examination of the educational online portal”Higher Education” Teacher of Mathematics at a Secondary School.
In this article author said
| [5] | N. Tanwani, S. Kumar, A. H. Jalbani, S. Soomro, M. I. Channa, Z. Nizamani, Student opinion mining regarding educational system using facebook group, in: 2017 First International Conference on Latest trends in Electrical Engineering and Computing Technologies (INTELLECT), IEEE, 2017, pp. 1-5. |
[5]
This paper contains 250 student/member comments. Data was collected from the Facebook Academic community using the Facebook graph API and the Opinion Analysis App. to examine students’ optimistic, negative, and neutral attitudes toward an educational system, with the aim of enhancing the educational system’s learning experience. After the comments were extracted, the vocabulary of the comments was extracted. It is identified because Facebook users are not limited to posting. It was limited to English since it was only in one language. The comments were processed and classified using Bayesian classification. A probabilistic model is built based on the opinion/sentiment ratings. A total of 250 comments were collected from 55 MS students. via a Facebook group Text data as well as images can be found in comments. The students/members of the group’s emoticons have It has been examined Based on the sentiment classification ranking, We discovered 56 percent positive, 32 percent neutral, and 12 percent negative. posts demonstrating the educational system’s need improvements to the syllabus, labs, teaching methods, and so on environment effective learning.
This research study can be expanded in the future by classifying/analyzing the text/comments at various levels, such as There are two levels of comparison:
1) document level
2) comparative level
Furthermore, classification can be done on a variety of languages, including Sindhi, Urdu, and English. Urdu, Arabic, and other languages Furthermore, this research study has the potential to be performed in a variety of ways on Facebook groups and pages for other reasons, such as marketing or business.
In this study
| [6] | V. Kagklis, A. Karatrantou, M. Tantoula, C. T. Panagiotakopoulos, V. S. Verykios, A learning analytics methodology for detecting sentiment in student fora: A case study in distance education., European Journal of Open, Distance and E-learning 18 (2) (2015) 74-94. |
[6]
Online conferences have become not only one of the most prominent communication tools in e-learning environments, but also one of the most important aspects of the learning process, especially in distance learning, since they can motivate students to work together to achieve a common goal. The aim of this analysis is to examine data on postgraduate students’ involvement in their course’s online forum at the Hellenic Open University. Text mining techniques are used to analyze the content of the messages written, while social network analysis techniques are used to analyses the network in which the students communicate.
Furthermore, the same dataset is used for sentiment analysis and opinion mining. Our goal is to investigate students’ attitudes toward the course and its features, model their sentiment behavior over time, and determine whether or not this influenced their overall results. The combined knowledge gained from the proposed factors will provide tutors with useful and practical information about the structure and content of students’ data packets, patterns of interaction among them, and the trend of sentiment polarity during the course, allowing them to improve the educational process.
In the age of
| [7] | A. Abdelrazeq, D. Janßen, C. Tummel, S. Jeschke, A. Richert, Sentiment analysis of social media for evaluating universities, in: Automation, Communication and Cybernetics in Science and Engineering 2015/2016, Springer, 2016, pp. 233-251. |
[7]
With the advent of digitalization, a massive amount of sentiment on university-related topics is expressed on a daily basis through social media platforms. Postings from students and instructors, in particular, can be a useful resource for evaluating universities. As one of the most common microblogging sites, Twitter provides a wealth of data for opinion mining. This fact has prompted researchers to look for ways to explore Twitter for knowledge in the sense of universities. This paper examines the use of social media sentiment analysis as a supplement to traditional university evaluation methods. The derived data can be used to back up university rankings that have been criticized for not calculating key indicators. To evaluate opinions posted on Twitter, this paper uses sentiment analysis techniques. Initially, tweets related to selected German universities were gathered for this reason. Second, the tweets were divided into “Positive” and “Not Positive” categories based on their sentiment. Finally, the findings were reviewed to include insight on communicative issues at the university. This paper provides a forecast for future research in the form of an automated study of social media content to aid university assessment.
First, the related data of students’ landing behavior and resources explored behavior are collected and pretreated using the
| [8] | W. Jie, L. Hai-yan, C. Biao, Z. Yuan, Application of educational data mining on analysis of students’ online learning behavior, in: 2017 2nd International Confer- ence on Image, Vision and Computing (ICIVC), IEEE, 2017, pp. 1011-1015. |
[8]
log data produced in the “Engineering Mechanics Experiment” Autonomous Learning Platform built and developed by our institution. The features of the students’ study behavior are then analyzed, and the decision tree algorithm acquires the factors that influence the students’ log in and resource exploration behavior based on the analysis. Ac- cording to the findings, educators may tweak and refine teaching methods, improve the teaching process and curriculum creation, and then organize teaching content and create teaching modes based on student educational outcomes.
The aim of this research project
| [9] | E. P. Yildiz, M. Tezer, H. Uzunboylu, Student opinion scale related to moodle lms in an online learning environment: Validity and reliability study., International Journal of Interactive Mobile Technologies 12 (4) (2018). |
[9]
is to create a Student Opinion Scale for Moodle LMS (Learning Management System) in an Educational Online Learning Environment. A scale form with 34 items was developed for this purpose, and it was then administered to 280 first-year students from different divisions of the Near East University Health Services Vocational High School. The validity of the scale was determined using rotated principal component analysis. The data was analyzed with the SPSS 23.0 software package, and the findings are presented as an integer average and standard deviation. In this sense, it was discovered that the scale’s related elements are organized into six dimensions, which the researchers have labelled as:
1) Proficiency and Motivation
2) Content
3) Feedback
4) Usability
5) Effectiveness
6) Educational Features
Furthermore, Cronbach’s alpha calculates a reliability coefficient of .94 for the developed scale. According to the study’s overall findings, student perceptions of Moodle LMS in an Educational Online Learning Environment are favourable.
In this paper
| [10] | R. Oramas Bustillos, R. Zatarain Cabada, M. L. Barr´on Estrada, Y. Herna´ndez P´erez, Opinion mining and emotion recognition in an intelligent learning environment, Computer Applications in Engineering Education 27 (1) (2019) 90-101. |
[10]
, The creation of an opinion mining module is presented. The module’s implementation included developing an emotion-tagged dataset of opinions; implementing an opinion mining module that processes sentences about computer programming, predicting or recognizing their polarity (positive/negative) and type of emotion (frustrated, bored, excited, engagement, and neutral); and implementing an opinion mining module that processes sentences about computer programming, predicting or recognising their polarity (positive/negative) and type of emotion (frustrated, bored, excited, engagement, and neutral).
We looked at the corpus, text polarity precision, and emotion recognition. The results in terms of polarity are positive (88.26 per), but the results in terms of emotion detection are still poor (60.0 per). A limited (7,777 records) and unbalanced corpus are two factors that are likely to explain these results.
During
| [11] | T. Chen, L. Peng, B. Jing, C. Wu, J. Yang, G. Cong, The impact of the covid-19 pandemic on user experience with online education platforms in china, Sustain- ability 12 (18) (2020) 7329. |
[11]
the COVID-19 pandemic, social education has moved from face-to-face to online in order to prevent massive crowds and stop the virus from spreading. To investigate the effect of the virus on the user interface and to learn more about the needs of the users. By integrating emotional analysis, hot mining technology, and applicable literature, this paper creates a fair assessment index framework based on user feedback of seven major online education platforms before and after the outbreak of COVID-19. Simultaneously, the variance coefficient approach is used to weight each index using the difference in index values. In addition, this paper uses a systematic assessment approach to examine user experience before and after the outbreak of COVID-19, and to determine if users’ concerns about the online education platform have changed. This paper examines the supporting abilities and response levels of online education platforms during COVID-19 in terms of access speed, reliability, timely delivery technology of video content, course management, communication and interaction, and learning and technical support, and proposes corresponding steps to enhance how these platforms work.
When the
| [12] | R. Yang, Machine learning and deep learning for sentiment analysis over students’ reviews: An overview study (2021). |
[12]
As the world continues to be affected by the COVID-19 pandemic, many schools have moved their instruction from physical classrooms to on- line platforms. It is critical for schools and online learning platforms to examine feedback in order to gain useful insights into the online teaching process, so that both platforms and teachers can learn which aspects they can develop in order to improve teaching performance.
2.1. Theoretical Framework
Constructivist theory and teaching paradigms are congruent with the literature analyzed, as well as the issues and potential solutions that address student impressions of online courses.
| [13] | E. Murphy, Constructivism: From philosophy to practice. (1997). |
[13]
Identified 16 constructivism features, three of which are highlighted in this research study:
1) Guides, monitors, mentors, tutors, and facilitators are all roles that teachers play.
2) Student major issue, relatively high reasoning abilities, and deep comprehension are emphasized.
3) Individual contexts, social negotiation, teamwork, and practice all play a role in learning.
In e-learning environments,
| [14] | K. C. Dewi, P. I. Ciptayani, H. D. Surjono, et al., Modeling vocational blended learning based on digital learning now framework., Turkish Online Journal of Educational Technology-TOJET 17 (2) (2018) 89-96. |
[14]
adapted
| [13] | E. Murphy, Constructivism: From philosophy to practice. (1997). |
[13]
characteristics into an advanced model of constructivism. The three components that have been determined are:
Collaboration, teamwork, various interpretations of concepts, and social negotiations are all part of the design of learning activities.
1) Mentoring, acknowledging, giving input, and evaluating student learning are all roles that instructors play.
2) Learning assessment may be done by the mentor, in conjunction with the student alternatively, the student himself.
Educational technologies are frequently utilized to simplify the learning process by creating a conducive learning environment and assessing student progress.
| [15] | L. C. Jackson, S. J. Jones, R. C. Rodriguez, Faculty actions that result in student satisfaction in online courses., Journal of Asynchronous Learning Networks 14 (4). |
[15]
. As a result, a constructivist approach to online education will promote more creativity in the classroom while avoiding the use of technology to measure student success. (For example, presenting knowledge and providing drills and practise). Technology is more useful in theory when students use it to identify a problem and then use the appropriate tools to help them understand it.
Teamwork (amongst teachers and students, as well as between students and other students), online content methods, and educational methodology frameworks are three areas that need particular attention and are discussed in this report are based on these distinctive nature of online courses and e-learning environments, as well as constructivist theory components
| [16] | O. Grundmann, D. Wielbo, I. Tebbett, The implementation and growth of an international online forensic science graduate program at the university of florida., Journal of College Science Teaching 40 (1) (2010). |
[16]
.
2.2. Research and Evaluation
2.2.1. Student Motivations
Several studies have been conducted to better understand the factors that contribute to students’ performance in online/hybrid learning. Students’ expectations of a rigorous curriculum, according to Duncan, Range, and
| [17] | H. E. Duncan, B. Range, D. Hvidston, Exploring student perceptions of rigor online: Toward a definition of rigorous learning, Journal on Excellence in College Teaching 24 (4) (2013) 5-28. |
[17]
, provide inspiration for performance. A comprehensive curriculum is one in which the goals and learning objectives are clearly described. Literature also suggests that an educator should explore approaches other than those already in use used in traditional settings to encourage students to participate in online learning
| [18] | B. R. Brocato, A. Bonanno, S. Ulbig, Student perceptions and instructional eval- uations: A multivariate analysis of online and face-to-face classroom settings, Education and Information Technologies 20 (1) (2015) 37-55. |
[18]
. This study discovered that students are dissatisfied when a course is poorly designed and they spend too much time hunting for information. Students become demotivated and confused about course objectives when expectations are unclear or learning goals are changed regularly during class.
| [17] | H. E. Duncan, B. Range, D. Hvidston, Exploring student perceptions of rigor online: Toward a definition of rigorous learning, Journal on Excellence in College Teaching 24 (4) (2013) 5-28. |
[17]
. The prerequisites for good online learning, according to this study, are course clarification and organization.
According to the research, students’ interest in class improves when they have the ability to interact with their classmates as well as the instructor, and they have access to 19 various points of view.
| [17] | H. E. Duncan, B. Range, D. Hvidston, Exploring student perceptions of rigor online: Toward a definition of rigorous learning, Journal on Excellence in College Teaching 24 (4) (2013) 5-28. |
| [19] | S. Palmer, D. Holt, Students’ perceptions of the value of the elements of an on- line learning environment: Looking back in moving forward, Interactive Learning Environments 18 (2) (2010) 135-151. |
[17, 19]
. Instructors must regularly engage and interact with students, which necessitates being approachable and earning students’ trust and confidence
| [18] | B. R. Brocato, A. Bonanno, S. Ulbig, Student perceptions and instructional eval- uations: A multivariate analysis of online and face-to-face classroom settings, Education and Information Technologies 20 (1) (2015) 37-55. |
[18]
. Students become more inspired when they can develop relative awareness and demonstrate learning. Easy organization, connectivity, and engagement improve students’ motivation for online learning.
2.2.2. Communication
Any educational activity needs effective communication. To succeed in an academic course, students must cooperate with their teacher and peers
| [20] | J. E. Brindley, L. M. Blaschke, C. Walti, Creating effective collaborative learning groups in an online environment, International Review of Research in Open and Distributed Learning 10 (3) (2009). |
| [21] | R. T.-H. Chen, S. Bennett, K. Maton, The adaptation of chinese international students to online flexible learning: Two case studies, Distance Education 29 (3) (2008) 307-323. |
[20, 21].
The absence of a physical environment that would necessarily facilitate inter- action is one of the key and intrinsic characteristics of an online course
| [22] | A. Driscoll, K. Jicha, A. N. Hunt, L. Tichavsky, G. Thompson, Can online courses deliver in-class results? a comparison of student performance and satisfaction in an online versus a face-to-face introductory sociology course, Teaching Sociology 40 (4) (2012) 312-331. |
[22]
.
Students in an online course have a unique opportunity to form groups in which they can ask questions, challenge one another, and learn new things, all of which are critical components of a constructivist approach.
| [23] | J. Bryant, A. J. Bates, Creating a constructivist online instructional environment, TechTrends 59 (2) (2015) 17-22. |
[23]
.
As a result, it is the responsibility of the course designer to choose the interactive resources that can be used in an online course for communication. The problems of deciding and organizing the interact portion of an online course, as well as potential solutions, are discussed in this section of the literature review.
Communication may be asynchronous or synchronous. Synchronous com- medication, as in a traditional face-to-face classroom, refers to communication that takes place in real time. Videoconferencing or virtual sessions would be needed to achieve similar 20 contact in an online course. A software like Blackboard is an ex- ample of one that allows for synchronous communication. There is a temporal gap between information flow and getting responses from others, even if asynchronous contact happens. A relationship between a teacher and his or her students Examples of asynchronous behavior Email and message boards are two popular communicate- ton options. Students will share ideas with other students and with the teacher in both synchronous and asynchronous practices, which is why a greater understanding of the obstacles to communication and cooperation that students face is critical.
Consider communication and collaboration to be the most critical and troublesome aspects of effective online learning
| [20] | J. E. Brindley, L. M. Blaschke, C. Walti, Creating effective collaborative learning groups in an online environment, International Review of Research in Open and Distributed Learning 10 (3) (2009). |
| [21] | R. T.-H. Chen, S. Bennett, K. Maton, The adaptation of chinese international students to online flexible learning: Two case studies, Distance Education 29 (3) (2008) 307-323. |
| [24] | V. A. Durrington, A. Berryhill, J. Swafford, Strategies for enhancing student in- teractivity in an online environment, College teaching 54 (1) (2006) 190-193. |
| [25] | F. Martin, M. Parker, B. Allred, A case study on the adoption and use of syn- chronous virtual classrooms., Electronic Journal of E-learning 11 (2) (2013) 124- 138. |
[20, 21, 24, 25]
. Because of the com- laxities, some researchers focus solely on synchronous tools
| [25] | F. Martin, M. Parker, B. Allred, A case study on the adoption and use of syn- chronous virtual classrooms., Electronic Journal of E-learning 11 (2) (2013) 124- 138. |
[25]
, while others focus solely on asynchronous tools
| [26] | F. Gao, T. Zhang, T. Franklin, Designing asynchronous online discussion environ- ments: Recent progress and possible future directions, British Journal of Educa- tional Technology 44 (3) (2013) 469-483. |
[26]
, and there is a shortage of research on the effects of combining tool.
2.2.3. Synchronous Communication
Voice, videos, messaging app, smart boards, screen sharing, immediate voting, symbols, and conference room, to mention a few capabilities, allow students and instructors to engage in real time in an online environment. However, there are no recommendations for using the full range of educational tool functions to make a virtual class more engaging. When provided access to synchronous communication resources that can be used for a number of purposes. When it comes to functions, students have a proclivity to take advantage of the variety of options available, and when it comes to technical issues, students have a proclivity to When issues arise, they can quickly exceed what a teacher can handle. Problem-solve
| [27] | C. A. Warden, J. O. Stanworth, J. B. Ren, A. R. Warden, Synchronous learning best practices: An action research study, Computers & Education 63 (2013) 197- 207. |
[27]
. In reality, relying on too many technological features can backfire. Causes a teacher to be overworked
| [27] | C. A. Warden, J. O. Stanworth, J. B. Ren, A. R. Warden, Synchronous learning best practices: An action research study, Computers & Education 63 (2013) 197- 207. |
[27]
. Instructors profit by deciding which technological features are most useful to students and their learning, rather than offering too many choices.
In addition to words and other traditional teaching tools like demos, screen sharing, and interactive presentations, students and teachers can communicate via videoconferencing utilizing nonverbally symbols including nonverbal and verbal com- medication cues.
| [28] | M.-j. Wang, Online collaboration and offline interaction between students using asynchronous tools in blended learning, Australasian Journal of Educational Tech- nology 26 (6) (2010). |
[28]
. However, due to the possibility of students being distracted or confused in a simulated setting, caution should be exercised
| [27] | C. A. Warden, J. O. Stanworth, J. B. Ren, A. R. Warden, Synchronous learning best practices: An action research study, Computers & Education 63 (2013) 197- 207. |
[27]
. Students are familiar with on-demand video and immersive video games, but they are not familiar with organized synchronous learning environments
| [27] | C. A. Warden, J. O. Stanworth, J. B. Ren, A. R. Warden, Synchronous learning best practices: An action research study, Computers & Education 63 (2013) 197- 207. |
| [29] | E. R. Cole, Intersectionality and research in psychology., American psychologist 64 (3) (2009) 170. |
[27, 29]
Addresses a variety of problems that occur when students use technology in a passive manner, such as failing to download material and failing to learn how to use apps before class. Instructor intervention is required to provide technical assistance to students or to point them in the correct place for assistance. Knowing what to do if voice or visual isn’t working fine benefits both teachers and students
| [25] | F. Martin, M. Parker, B. Allred, A case study on the adoption and use of syn- chronous virtual classrooms., Electronic Journal of E-learning 11 (2) (2013) 124- 138. |
[25]
.
2.2.4. Asynchronous Communication
A variety of cognitive and social practices have been related to participation in online message boards, wikis, journals, and blogs
| [26] | F. Gao, T. Zhang, T. Franklin, Designing asynchronous online discussion environ- ments: Recent progress and possible future directions, British Journal of Educa- tional Technology 44 (3) (2013) 469-483. |
[26]
. As a result, asynchronous communication tools are used in the majority of online courses. According to
| [26] | F. Gao, T. Zhang, T. Franklin, Designing asynchronous online discussion environ- ments: Recent progress and possible future directions, British Journal of Educa- tional Technology 44 (3) (2013) 469-483. |
| [31] | J. Jun, J. H. Park, Power relations within online discussion context: Based on adult international students’ perspective and their participation in the learning context. (2003). |
[26, 31]
, substantive discussion board involvement necessitates the following four characteristics:
1) Analyze in terms of understanding;
2) Analyze in terms of the methods;
3) Analyze in terms of learning;
4) Analyze in addition to creating.
As a result, students should be able to add a variety of viewpoints and ideas in an online environment and receive feedback. Asynchronous communication is used to encourage social participation and to make sharing and disseminating information and skills within a student group easier. As a result, one goal of online teaching is to create online groups where students can work together to attain common academic objectives and finish school work.
| [30] | A. Januszewski, M. Molenda, Educational technology: A definition with commen- tary, Routledge, 2013. |
[30]
.
With regard to posts, articulation, and overall classroom communication, certain kids require more assistance. “Everyone spoke about their own experiences and feelings, and I didn’t know who to listen to without the teacher’s comments,” a student in
| [21] | R. T.-H. Chen, S. Bennett, K. Maton, The adaptation of chinese international students to online flexible learning: Two case studies, Distance Education 29 (3) (2008) 307-323. |
| [23] | J. Bryant, A. J. Bates, Creating a constructivist online instructional environment, TechTrends 59 (2) (2015) 17-22. |
[21, 23]
analysis said. ” If there is a class discussion, the teacher must participate; otherwise, the platform will appear disorganized. “The immediacy of action and engagement of the teacher in the online learning room informs pedagogical practice, boosting collaboration in multiple ways,” writes the author.
| [32] | M. Toledo, L. Poorter, M. Pen˜a-Claros, A. Alarco´n, J. Balc´azar, C. Lean˜o, J. C. Licona, O. Llanque, V. Vroomans, P. Zuidema, et al., Climate is a stronger driver of tree and forest growth rates than soil and disturbance, Journal of Ecology 99 (1) (2011) 254-264. |
[32]
. In order to inspire and enable students to continue participating in discussion boards, instructors can participate in 24 of them.
To keep students involved and inspired when working with others, a variety of resources are needed. When it comes to calculating collective instrument commitment, one method can be too restrictive
| [26] | F. Gao, T. Zhang, T. Franklin, Designing asynchronous online discussion environ- ments: Recent progress and possible future directions, British Journal of Educa- tional Technology 44 (3) (2013) 469-483. |
[26]
. Instead of shared learning with other students, the majority of students use social networking skills for entertainment or consumption
| [29] | E. R. Cole, Intersectionality and research in psychology., American psychologist 64 (3) (2009) 170. |
[29]
. Instructors can respond by allowing students to practise editing, publishing, and uploading content via any communication platform by offering a space and activities
| [29] | E. R. Cole, Intersectionality and research in psychology., American psychologist 64 (3) (2009) 170. |
[29]
.
In comparison to synchronous communication tools, asynchronous communication tools are far more common in the literature. Regardless of the communication framework, students’ technological and personal restrictions, as well as a basic lack of enthusiasm.
2.2.5. Content Delivery
“Becoming an open dialogue or collaboration forum is not appropriate for every situation”
| [29] | E. R. Cole, Intersectionality and research in psychology., American psychologist 64 (3) (2009) 170. |
[29]
. Instruction, as well as other tools and instructions, can be provided in a variety of ways in virtual worlds. Content can be distributed in a number of ways thanks to online innovations. The degree to which both can be used depends on the medium used, whether it’s a learning management system or a website.
Table 1. Comparison table of different models on Student opinion.
Year | Models | Best | Accuracy | Data set | Ref |
2023 | KNN, RF, Multinomial NB | Multinomial, NB | 79 per | from overall world students opinion | | [39] | M. L. B. Estrada, R. Z. Cabada, R. O. Bustillos, M. Graff, Opinion mining and emotion recognition applied to learning environments, Expert Systems with Ap- plications 150 (2020) 113265. |
[39] |
2023 | SVM. DF, RF, | SVM | 99 per | from different uni-versity opinions | | [40] | A. Souri, M. Y. Ghafour, A. M. Ahmed, F. Safara, A. Yamini, M. Hoseyninezhad, A new machine learning-based healthcare monitoring model for student’s condition diagnosis in internet of things environment, Soft Computing 24 (2020) 17111- 17121. |
[40] |
2021 | SVM, NB | NB | 91 per | Benchmark dataset and Academic domain data | | [41] | I. Sindhu, S. M. Daudpota, K. Badar, M. Bakhtyar, J. Baber, M. Nurunnabi, Aspect-based opinion mining on student’s feedback for faculty teaching perfor- mance evaluation, IEEE Access 7 (2019) 108729-108741. |
[41] |
2019 | NB, RF, Lin-ear SVC | NB | 83 per | student feed backs | | [42] | I. A. Kandhro, M. A. Chhajro, K. Kumar, H. N. Lashari, U. Khan, Student feedback sentiment analysis model using various machine learning schemes: A review, Indian Journal of Science and Technology 12 (14) (2019). |
[42] |
2020 | NB, SVM, RF, LR | NB | 93 per | From students opinion | | [43] | J.-a. P. Lalata, B. Gerardo, R. Medina, A sentiment analysis model for faculty comment evaluation using ensemble machine learning algorithms, in: Proceedings of the 2019 International Conference on Big Data Engineering, 2019, pp. 68-73. |
[43] |
Broadcasting technologies (podcasts and screen casts) seem to be suitable for creating successful learning environments. There is no conclusive answer as to which combination of content delivery strategies benefits pupils the most. Experts advocate for combining technology and teaching strategies to motivate students to understand
| [18] | B. R. Brocato, A. Bonanno, S. Ulbig, Student perceptions and instructional eval- uations: A multivariate analysis of online and face-to-face classroom settings, Education and Information Technologies 20 (1) (2015) 37-55. |
| [33] | A. Brown, C. Brown, B. Fine, K. Luterbach, W. Sugar, D. C. Vinciguerra, Instruc- tional uses of podcasting in online learning environments: A cooperative inquiry study, Journal of Educational Technology Systems 37 (4) (2009) 351-371. |
| [34] | D. Zen, How to be an effective online instructor., Online Submission (2008). |
[18, 33, 34]
advocate for using a variety of content delivery approaches to meet student needs and make it easier to implement various learning strategies.
2.2.6. Role of the Instructor
Warden discovered that when the instructor has full control over the environment, learning is most successful
| [27] | C. A. Warden, J. O. Stanworth, J. B. Ren, A. R. Warden, Synchronous learning best practices: An action research study, Computers & Education 63 (2013) 197- 207. |
[27]
. By offering education and guidance on how to use computer correctly, this control helps students avoid many of the basic problems that come with online courses. Warden also found out that students still have a lot of opportunities to develop their learning in these structured environments, and that as their technical skills grow, they will be able to explore more student-centered control situations. In the end, the most critical element in the success of an online course is the instructor
| [35] | D. Furi´o, M.-C. Juan, I. Segu´ı, R. Viv´o, Mobile learning vs. traditional classroom lessons: a comparative study, Journal of Computer Assisted Learning 31 (3) (2015) 189-201. |
| [36] | C. Zhu, M. Valcke, T. Schellens, A cross-cultural study of online collaborative learning, Multicultural Education & Technology Journal (2009). |
| [37] | M.-H. Cho, G. Rathbun, Implementing teacher-centred online teacher professional development (otpd) programme in higher education: A case study, Innovations in Education and Teaching International 50 (2) (2013) 144-156. |
| [38] | M. Stansfield, E. McLellan, T. Connolly, Enhancing student performance in online learning and traditional face-to-face class delivery, Journal of Information Tech- nology Education: Research 3 (1) (2004) 173-188. |
[35-38]
.
Creating instructions and directions does not inhibit student creativity or learning because it assists students in navigating the course and provides them with important explanations concerning course papers, assignments, and course content goals. Students will have a strong basis on which to construct their own learning and creativity by creating a well-defined learning area. A teacher must first present the boundaries and restrictions in order to better allow student performance.
Table 2. Comparison table of different models on Student opinion.
Year | Models | Best | Accuracy | Data set | Ref |
2018 | SMO, RF, SNM | SMO | 81 per | from different universities students opinion | | [44] | S. C. Harris, V. Kumar, Identifying student difficulty in a digital learning en- vironment, in: 2018 IEEE 18th International Conference on Advanced Learning Technologies (ICALT), IEEE, 2018, pp. 199-201. |
[44] |
2017 | SVM Radial, SVM Linear, SVM poly | SVM Linear | 0.8 per | 1040 students opinion in Span- ish | | [45] | G. Guti´errez, J. Canul-Reich, A. O. Zezzatti, L. Margain, J. Ponce, Mining: Stu- dents comments about teacher performance assessment using machine learning algorithms, International Journal of Combinatorial Optimization Problems and Informatics 9 (3) (2018) 26. |
[45] |
2016 | SVM | SVM | Different accuracy in different laptops | From different re- viewers | | [46] | D. N. Devi, C. K. Kumar, S. Prasad, A feature based approach for sentiment anal- ysis by using support vector machine, in: 2016 IEEE 6th International Conference on Advanced Computing (IACC), IEEE, 2016, pp. 3-8. |
[46] |
2015 | NB, lexicn based, lexi-con pooled | NB | 79 per | From 3 different data set | | [47] | R. Dalal, I. Safhath, R. Piryani, D. R. Kappara, V. K. Singh, A lexicon pooled machine learning classifier for opinion mining from course feedbacks, in: Advances in Intelligent Informatics, Springer, 2015, pp. 419-428. |
[47] |
2015 | POS, AoMR 1 | POS | 72 per | from different uni and colg students opinion | | [48] | T. Chinsha, S. Joseph, A syntactic approach for aspect based opinion mining, in: Proceedings of the 2015 IEEE 9th International Conference on Semantic Comput- ing (IEEE ICSC 2015), IEEE, 2015, pp. 24-31. |
[48] |
4. Results
4.1. Data Analysis
The research question did not differentiate between the four classes because it was designed to measure student barriers to virtual classrooms irrespective of whether they were able or planning to take an online study in the future. The aim of this question will be to identify any potential limitations that almost all students could face when deciding whether to take an online course. A research Question All of the respondents who would not have been able to choose whether to take an online course must be considered. In improving the 34 validity of the data for the first research question, response frequencies were calculated and evaluated.
The research question seeks to determine whether there is a substantial gap between students who choose or plan to take an online course in the future and those who do not. A statistical comparison was made using several Mann-Whitney U Tests. This approach was chosen because it allows for comparisons between groups of various sizes. This approach was used to examine the variations between students who preferred traditional environments and students who chose to take an online class. Data analysis have some steps:
1) Collect data
2) Source (Google form)
3) Prepossessing the data
4) Applying NLP
5) Then splitting the data (testing and training)
6) Applying Models
7) And then predict the values
4.1.1. Original Data
First we visualize the data which we collected. From the source of Google online form. In which students give their opinions about the online study from different platform.
Figure 4. Original data set.
4.1.2. Prepossessing the Data
Figure 5. Prepossessing the data set.
We display all of those things which we doing on my project. First step is conducting the data from students. Then we have to preprocess our data because some students missing the questions and some are not correctly giving authentic answer of the question. That is why we preprocess the data into integer values from text and overcome the missing values.
4.1.3. Opinion Analysis
Now we are going to show the graphical representation of the Students opinions.
Figure 6. Opinion analysis of students for conducting online classes.
Table 3. Opinion Analysis.
Opinion | Comments |
Positive | Good, Appraisal, Initiative, well, Appear, Nice, Enjoyment, Learn |
Negative | Hard, Remove, Go, Cheap, Bycot, |
Neutral | Class, Online, Platform, Student, Mean, Govt, Provide, Packages |
4.1.4. Description of Data Set
Figure 7. Description of data set.
4.1.5. Questions of Data Set
Figure 8. Questions which we taking the answers from the students.
4.2. Graphical Representation
Now we are going to show the graphical representation of some of the basic question which we are taking the answer of our questions.
in question 1 we are show that how much males and females are involve in our survey.
Figure 9. Gender contribution.
4.2.1. First Language of Students
In this question we want to ask that the student belong from which language.
Figure 10. First language of students.
4.2.2. Students Belong from University or College
In this question we want to ask that the student belong from which University.
4.2.3. Platform We Are Used
In this question, we want to ask that the student used which platform for online study.
Figure 11. Students belong from university or college.
Figure 12. Which platform we used.
4.2.4. Facing Electricity Problem
In this question, we want to ask that the student used which kind of problems students facing for taking the online class.
Figure 13. Facing electricity problem.
4.2.5. Facing Internet Problem
In this question, we want to ask that the student used which kind of problems students facing for taking the online class.
Figure 14. Facing internet problem.
4.2.6. Facing Voice Problem at Home
In this question we want to ask that the student facing voice prob for taking the online class.
Figure 15. Facing voice problem at home.
4.2.7. Student Satisfaction to Their Tutor
In this question we want to ask that the student satisfying of their tutor for taking the online class.
Figure 16. Students satisfy of their tutor in online class.
4.2.8. Students Assignment
In this question, we want to ask that the student are able to manage the assignments and listening the lecture carefully in online class.
Figure 17. Students manage their assignment.
4.2.9. Video Quality
In this question, we want to ask that the student are able to watch the slides videos or any kind of video lecture online class.
Figure 18. Video quality.
4.2.10. Online Class Saving Your Time
In a question we want to ask that the student are able to tell, that online class are saving our time or not.
Figure 19. Online class saving your time.
4.2.11. Preferring Online Class
In a question we want to ask that the student are able to tell, that online class are beneficial for them or not.
Figure 20. Preferring online class.
4.3. Models
We are using two models in which we will compare that which models are giving us the better accuracy.
4.3.1. SVM
Multiple classification problems, a support vector machine (SVM) is a supervised machine learning model that uses classification algorithms. SVM models will categories new text after being given sets of labelled training data for each group. Therefore, you are attempting to solve a text classification problem.
Figure 21. Preferring online class.
4.3.2. Random Forest
Random forests, also known as random decision forests, are an ensemble learning method for classification, regression, and other tasks that works by training a large number of decision trees and then outputting the class that is the mode of the classes (classification) or the mean/average prediction (regression) of the individual trees.
Figure 22. Preferring online class.
We are showing the comparison of RF and SVM by this table.
Table 4. Our models predictions.
Sr. no | Models | Accuracy |
1 | SVM | 0.8 |
2 | RF | 0.6 |
Share This Article
Twitter
Linked In
Facebook
Copy
Download
