ChapterPDF Available

Student Sentiment Analysis Using Gamification for Education Context


Abstract and Figures

Internet users are expressing their sentiments (opinions) online using blogs and social media. Sentiment Analysis is a new technology that is used to improve the quality of the institutions including Higher education institution (HEI). Egypt educational institutions face a difficulties based on student motivation and learning engagement. Gamification provides a great help for educational institutions to motivate student and increase their learning ability however it depends on the teacher skills to use the Gamification tools. The paper reviews work in sentiment analysis related to education field, Gamification in learning. The paper will propose a Sentiment Analysis Classifier that will analyze the sentiments of students while using Gamification tools in an educational course.
Content may be subject to copyright.
Student Sentiment Analysis Using Gamication
for Education Context
Lamiaa Mostafa
Business Information System Department, Arab Academy for Science
and Technology and Maritime Transport, Alexandria, Egypt,
Abstract. Internet users are expressing their sentiments (opinions) online using
blogs and social media. Sentiment Analysis is a new technology that is used to
improve the quality of the institutions including Higher education institution
(HEI). Egypt educational institutions face a difculties based on student moti-
vation and learning engagement. Gamication provides a great help for edu-
cational institutions to motivate student and increase their learning ability
however it depends on the teacher skills to use the Gamication tools. The paper
reviews work in sentiment analysis related to education eld, Gamication in
learning. The paper will propose a Sentiment Analysis Classier that will
analyze the sentiments of students while using Gamication tools in an edu-
cational course.
Keywords: Sentiment Analysis Gamication WordNet Learning
Slang Arabic
1 Introduction
Education systems depend on four key elements which are teacher, the student, the
university, and the curriculum. Student motivation affects the student performance in
the educational institution. Gamication is dened as the use of game design elements
in non-game contexts [1]. Gamication can be used to solve actual problems in dif-
ferent elds.
Internet helped people to express their thoughts and feelings [2]. Blog post and
online forums allow users to write their reviews. The people are connecting with each
other with the help of the internet through the blog post, online conversation forums,
and many more online user-generated reviews which is also called Sentiments also
known as opinion mining [3] can be used in different ways: used in decision making
process of consumers, sentiments provide efcient low cost feedback channel, improve
the quality of companies products and services which will also enhance their repetition
[4]. Authors in [2] proposed a tool which decides the quality of text based on scientic
papers annotations.
The rest of this paper is organized as follows: Sentiment Analysis, Gamication in
the second and third Section, sentiment analysis classier will be described in Sect. 4;
results will be analyzed in Sect. 5; Sect. 6include the paper conclusion and future
©Springer Nature Switzerland AG 2020
A. E. Hassanien et al. (Eds.): AISI 2019, AISC 1058, pp. 329339, 2020.
2 Sentiment Analysis
Sentiments and opinions can be recognized in different ways. Overall sentiment which
is sentiment as a whole piece of text can be analyzed using classication tools. Clas-
siers divide sentiments into positive or negative sentiment [2]. Different researchers
classify overall sentiment analysis [5].
Another type of sentiment classication is Aspect-based sentiment analysis [6,7].
Aspect sentiment include two major tasks, the rst is detect hidden semantic aspect
from given texts, the second is identify ne-grained sentiments [2].
Data has two perspectives: objective and subjective. Objective based on emotions
and personal feelings and subjective is based on facts. Sentiment analysis aim is to
analyze subjective perspective of data [8]. Text mining techniques used to analyze
sentiments of students [9]. Sentiment analysis recognize student emotions [10].
Researchers are working on sentiment analysis in different patterns: Twitter sentiment
analysis [11] and cross lingual portability [12] and others. The following subsections
will discuss Sentiment processing steps and previous research in sentiment analysis
used in learning.
2.1 Sentiment Processing
Authors in [13] divide the sentiment process into 4 stages: data acquisition, data
preparation, review analysis, and sentiment classication. Sentiment process consists of
many steps such as stop word removal in which irrelevant terms are removed such as
theor and)[14]. Remove html tags; remove numbering and punctuation conver-
sion to lower case and stemming words (removing sufxes and prexes to identify the
stem [15]. Term-frequency-inverse document frequency (TF-IDF) should be used to
count the frequency of keywords on the document [14]. Clustering algorithms should
be used to dene correlation between the terms and the topic of the document.
Classify sentiments depends on two techniques: machine learning divided into
supervised and unsupervised or lexicon-based divided into dictionary based and corpus
based [16]. Machine learning uses traditional mining algorithms such as Naive Bayes
(NB), Support Vector Machines (SVM) and Neural Networks (NN). Naïve Bayes is the
simplest and most used classier [17], Support Vector Machine apply learning models
to analyze data. Neural Networks requires a large corpus with three classes: positive,
negative and neutral opinions to learn and classify new sentiment into one of the three
classes [18].
Lexical based approach depends on dictionaries of terms such as WordNet. The
lexicon-based approach classies a text according to the positive, negative and neutral
words and it does need training phase. Lexicon-based approach divided into dictionary-
based and corpus-based approach. Dictionary-based approach uses synonyms, anto-
nyms, and hierarchies found in lexical databases, Corpus-based approach nd word
patterns occurrence to determine the polarity of a text [13]. After understanding the
processing of sentiment analysis, the paper will explore some of the researches that
implement the sentiment analysis method in leaning eld.
330 L. Mostafa
2.2 Sentiment in Learning
Opinions written by online users have been studied using SA [19]. Opinions text
should be cleaned since it is full of spelling mistakes and implicit meaning. Online text
processing is a very important process before extracting sentiments [20]. Educational
Data Mining (EDM) task is to extract knowledge to enhance studentslearning pro-
cesses [21]. Li [22] proposed a sentiment-enhanced learning model for the online
language learning, the experimental results show that sentiment learning is effective
tool online language learning. Permana et al. [23] proposed a sentiment model that
evaluates student satisfaction in learning using NB in classication which resulted in
enhanced of 16.49% over the existing system.
3 Gamication
The previous section discussed using sentiment analysis in learning, the aim of the
paper is to understand student sentiments on using Gamication in learning, and this
section will discuss Gamication concept and previous researchers that uses Gami-
cation in different sectors. Game design techniques help motivate people to nish the
required duties [24]. Games can be used in different services and increase the
involvement of people in non-game services [25]. The Gamication has 2 main ben-
ets: context role in game and the user qualities [26,27]. Gamication can be used in
different contexts tourism [28] and job recommendation in which a framework pro-
posed by [29] to understand the level of student in each working track and enhance his
learning path through Gamication tools. When person chooses to play a game over
responsibility, this is due to pressure and reduce stress [30]. Using Gamication in
higher learning assignment enhance the academic progress while reducing stress,
providing learning milestones and improving personal satisfaction [31]. It is important
to understand the previous researchers that use Gamication in learning so the next
subsection will discuss the previous work in Gamication.
Educational games pretend to be games, as the fun factor is an additive, and not
the goal of those creations, however it is a specic learning outcome goal[32].
Gamication in education aims to engage and motivate students [33], Gamication
develops the students skills such as collaboration, self-regulation and creativity. There
are many Gamication tools such as ClassDojo, Classcraft, Moodle plugin, Open
Badges [32] Games development has 2 main types: Game based learning and serious
games, game-based learning is using games to support learning however specic games
are games developed for the aim of teaching [1].
Different researchers use Gamication techniques in learning [3436]. Authors in
[34] implemented a Gamication tool (Funprog) for teaching Programming funda-
mentals for 80 students of rst-year students from University of Ecuador. The result of
the survey shows an acceptance level from the student and the teacher. Authors in [1]
provides a strategy to use e-learning platform and Gamication techniques in infor-
mation system courses.
Five interactivity story-based games where added by Arizona State University to
scientic curriculum to enhance student engagement [37]. Authors in [36] tested
Student Sentiment Analysis Using Gamication 331
Gamication on gender, age, and type of institution. They concluded that no differ-
ences in use of gamication by age, gender or type of institution. [35] applied Gam-
ication on software teaching in a specic class and the results had shown that
Gamication motivate students and enhance the learning tools used by the teacher. [38]
Focuses on the social gamication of e-learning and learner performance.
Authors in [25] proposed a Gamication model to be applied in Human Resource
Management course for Masters in Engineering. Authors in [39] designed a Gami-
cation tool that include levels based on storyline to enhance the teaching productivity
and learning path of the student concept for teaching in universities, they presented
Gamication for EPUB using Linked Data (GEL) is a framework combines Gami-
cation concepts and digital textbook they proposed Gamication Ontology (GO) that
share knowledge of gamied books between applications.
Briers in [40] applied gamied PMBOK for project management course targeting
junior project managers in Europe. PMBOK involves testing the following skills:
evaluating the reactions, cognitive learning, and the behavior changes.
Sobocinski in [32] proposed properties of Gamication tool used in education as
follows: narrative, points-badges, no grades, missions, small tasks, skills-badges,
individual and team tasks, boss ght, window of opportunity, optionality, schedule,
realoutcome. Narrative is to explain the course components, points and badges moti-
vate the students to reach different levels, grades are associated with difculty so it is
replaced by points and badges, mission which is tasks enhance the student ability in
learning, small missions represent milestones, skills- badges reect the development of
student level, team tasks create collaboration scenario between the student and his
colleagues, boss ght remove the fear of the nal exams associated works in the
thinking of the students that is replaced by badges and points, window of opportunity
in which milestones and tests can be retaken, optionality of attendance gives the
students the freedom to stay home and learn, schedule exibility enhance the student
ability to strength his skills, real outcomes represented in social experiment outside
class which give the student ability to present his progress to friends and family [32].
Learning motivations affect the student performance in class [41,42]. Authors in
[41] use sentiment analysis to improve the learning process in an e-learning environ-
ment. Sentiment analysis used to analyze the opinions of the students for better
understanding their opinion. After exploring the literature review in Sentiment Analysis
and Gamication, the Sentiment Analysis Classier is designed and implemented.
4 Sentiment Analysis Classier (SAR)
Sentiment Analysis classier is used to analyze the studentsopinion in using Gami-
cation tool in learning, sentiment analysis classier passes by different stages
including text processing, feature selection and classication. The section is divided
into 4 subsections including Text Processing, Features selection, Machine Learning
Classiers and Sentiment Dataset. Figure 1represents the Sentiment Analysis Classi-
er steps.
332 L. Mostafa
4.1 Text Processing
The phase of text processing passes by two steps: the rst step is the preprocessing step
and the second, is the feature selection or the representation of a document. The steps
are executed sequentially in which the output of the preprocessing step is the input of
the feature selection step [43].
Preprocessing is the step before the text is fed into a classier. Preprocessing
includes many steps. Classier preprocessing steps include parsing, removing stop
word and stemming [43]. Since the text is written in Arabic Chat Alphabet (ACA) also
known as (Slang Arabic),is a writing system for Arabic in which English letters are
written instead of Arabic ones[20], for example: gama3aha feh haad ye3rf el
gamication deh eh, the translation is people, anyone knows about gamication?.
ACA dictionary was created to be used in the Sentiment analysisClassier. Student
Sentiment is reviewed by two experts to conrm the accuracy of the ACA dictionary.
Documents are preprocessed before classication. Preprocessing include stop word
removal and stemming. One of the stemmers that usually used in classication is Porter
Stemmer [14]. Parsing process converts the HTML le into stream of terms and remove
the HTML tags, stop word removal is the process of removing words that does not
affect the meaning of document sentences e.g. (and, a, or, the, on) and stemming is the
process of reducing a word to its stem or root form. Thus, stems represent the key terms
of a query or document rather than by the original words, e.g. computationmight be
stemmed to compute[43].
Fig. 1. Sentiment analysis classier steps.
Student Sentiment Analysis Using Gamication 333
Word Net
is a large lexical database of English. George A. Miller is the main
director. WordNet works as follows: it groups networks called sunsets, these sunsets
include nouns, verbs, adjectives and adverbs, and each sunsets aim is to express a
distinct concept. Synsets include conceptual-semantic and lexical relations. WordNet
can be downloaded freely. WorldNets structure which is the form of networks, makes
it a useful tool for text mining techniques as classication and natural language pro-
cessing. Different members develop WordNet as George A. Miller, Christiane Fell-
baum, Randee Tengi, and Helen Langone. WordNet is maintained at the Cognitive
Science Laboratory of Princeton University under the direction of psychology pro-
fessor George A. Miller. WordNets goal is to develop a system that can acquire
knowledge extracted from human thinkings. One of these examples is the ability of a
human being in classifying different items into groups.
4.2 Classication Features
Any Webpage or document is represented using a combination of features. The basic
idea of feature selection algorithm is Searching through all possible combination of
features in the data to nd which subset of features work best for prediction. The
selection is done by reducing the number of features of the feature vectors, keeping the
most meaningful discriminating ones, and removing the irrelevant or redundant ones
[45]. For the extraction of these features there are many known methods, including
Document Frequency (DF), Information Gain (IG) [43], Mutual Information
(MI) which is biased to rare terms opposite to IG [44]. Sentiment Analysis Classier
uses DF technique and features are expanded using synonyms feature extracted from
4.3 Machine Learning Classiers
After dening the features, sentiments can be classied. Statistical methods and
Machine Learning classiers are usually used in sentiment classication including
Multivariate Regression Models, Bayes Probabilistic Approaches, Decision Trees,
Neural Networks, Support Vector Machine (SVM) [46], Symbolic Rule Learning,
Concept Vector Space Model (CVSM) [46], and Naïve Bayes (NB) is the most used
classier in sentiment analysis [19]. Sentiment analysis classier will use SVM and
is used in preprocessing and keyword extraction and classication process.
4.4 Sentiment Dataset
Sentiments were collected from 1000 students in Arab Academy for Science and Tech-
nology and Maritime Transport University (AAST) University in a business course.
700 students agreed to examine Gamication in learning while 300 students were not
interested. Before the process of using machine learning classier, two datasets must be
334 L. Mostafa
dened: the rst dataset is agree sentiment (700) and the second dataset is disagree
sentiment (300). ACA dictionary was created and mapped to every word in sentiments.
Two professors in College of linguistics in the AAST revised the dictionary to ensure the
accuracy level. After mapping the words of ACA to equivalent English words, the text
processed by the regular text processing steps including stop word removal, stemming,
WordNet is used for the enriching the keywords selected that represent agree and disagree
set. Synonyms relationship was selected such as agree = good idea = agree = happy,
all there words represent an agree sentiment. Two machine learning classiers were used:
SVM and NB, both are trained and tested in the 1000 sentiments with the percent 80% for
training and 20% for testing (Lewis and Catlett 1994).
After collecting the sentiments of the students, a written exam was executed on the
two groups: agree and disagree, to test the performance of the student when using
gamication and without using gamication and to conrm whether using gamication
enhance the learning curve of the students in the learning process or not. The following
section will discuss the results from using the proposed Sentiment Analysis Classier
(Table 1).
5 Sentiment Analysis Classier Results
Sentiment analysis classier tests the machine learning classiers based on 1000
sentiments, sentiments are classied into agree sentiment and disagree sentiment. Naive
Bayes, SVM and Decision Tree were used for the classication of the sentiments.
Sentiments are converted in to English language using ACA dictionary created by the
author and revised by two linguistic experts. Table 2shows the results of the sentiment
Table 1. Sample for ACA dictionary
ACA English
Ana mesh fahem I cannot understand
3agbny awy fekrt el game I do like the game idea
Brnamg bared Application is not good
Mewafee gedan I do strongly agree
Table 2. Sentiment analysis classier results
Classier Accuracy Precision Recall
NB 83% 0.76 0.67
SVM 79% 0.61 0.83
Decision Tree 76% 0.51 0.8
Student Sentiment Analysis Using Gamication 335
The results of sentiment analysis classier agreed with the conclusion of [1,34]
proving that using Gamication can enhance the efciency of student learning.
Sentiment analysis techniques is a very good reector of the student emotions
towards learning, this is equivalent to the conclusion of [21,22]. NB had shown the
highest accuracy results which is agreed with [23] who uses NB to evaluate the sen-
timent option of students in learning.
Gamication affects the performance of student in learning. The two groups were
tested to compare the performance when using Gamication and the results had shown
that performance is affected by Gamication in which the highest grade reached by the
Gamication group whose sentiment is agree reached 98 which is equivalent to A + in
the credit hour system. Table 3summarize the test results.
In conclusion, sentiment analysis classier compared to the other previous
researchers focuses on ACA which is used by the Arabic internet users; sentiment
analysis classier uses an ACA dictionary that can be used on the following researchers
as a base for converting ACA to English.
6 Conclusion and Future Work
In this paper we proposed a sentiment analysis classier, the sentiment classier data
set was 1000 student sentiments that are divided into 700 agree sentiment and 300
disagree sentiments. The sentiment analysis classier passes by text processing, feature
selection and machine learning classication, three classiers were used NB, SVM and
decision tree. The results had shown that the best classier accuracy results is the NB,
also when executing a test on the 1000 students, the agree group of using Gamication
in learning shown a better results comparing to the disagree group, this proves that
Gamication will enhance the student performance in learning. The limitation of the
Sentiment Analysis classier is listed as follows: Sentiment analysis classier should
be tested in a larger dataset and should also be applied in different students categories
such as: different learning year, different type of courses and even different universities
and the author should consider using different Machine learning classiers can be used
in classication such as Neural Networks and more semantic relationships can be
extracted from the sentiment not only the synonyms WordNet relationship. The author
will focus on future plans to expand the results of the sentiment analysis classier to be
tested on more students in different majors.
Table 3. Summary of test results.
Traditional (Disagree) group Gamication (Agree) group
Total number of students 300 700
Minimum grade 47 78
Maximum grade 91 98
336 L. Mostafa
1. Elabnody, M., Fouad, M., Maghraby, F., Hegazy, A.: Framework for gamication based E-
learning systems for higher education in Egypt. Int. J. Intell. Comput. Inf. Sci. IJICIS 17(4),
59 (2017)
2. Raut, S., Mune, A.: Review paper on sentiments analyzer by using a supervised joint topic
modeling approach. IJESC 8(1) (2018)
3. Vinodhini, G., Chandrasekaran, R.: Sentiment analysis and opinion mining: a survey. Int.
J. Adv. Res. Comput. Sci. Softw. Eng. 2, 282292 (2012)
4. Liu, B.: Sentiment analysis and opinion mining. In: Synthesis Lectures on Human Language
Technologies, vol. 5, no. 1, pp. 1167 (2012)
5. Yang, B., Cardie, C.: Context-aware learning for sentence-level sentiment analysis with
posterior regularization. In: Proceedings of the 52nd Annual Meeting of the Association for
Computational Linguistics, ACL 2014, 2227 June 2014, Baltimore, MD, USA, vol. 1,
Long Papers, pp. 325335 (2014)
6. Yang, Z., Kotov, A., Mohan, A., Lu, S.: Parametric and nonparametric user-aware sentiment
topic models. In: Proceedings of the 38th International ACM SIGIR Conference on Research
and Development in Information Retrieval, ser. SIGIR 2015, pp. 413422. ACM, New York
7. Rahman, M., Wang, H.: Hidden topic sentiment model. In: Proceedings of the 25th
International Conference on World Wide Web, ser. WWW 16. Republic and Canton of
Geneva, Switzerland: International World Wide Web Conferences Steering Committee,
pp. 155165 (2018)
8. Mandinach, E., Cline, H.: Classroom Dynamics: Implementing a Technology-Based
Learning Environment. Taylor & Francis, New York (2013)
9. Schouten, K., Frasincar, F.: Survey on aspect-level sentiment analysis. IEEE Trans. Knowl.
Data Eng. 28(3), 813830 (2016)
10. Yu, L.C., Lee, L.H., Hao, S., Wang, J., He, Y., Hu, J., Zhang, X.: Building Chinese
affective resources in valencearousal dimensions. Paper presented at the Proceedings of
11. Rosenthal, S., Nakov, P., Kiritchenko, S., Mohammad, S.M., Ritter, A., Stoyanov, V.:
Sentiment analysis in Twitter. Paper presented at the Proceedings of the 9th International
Workshop on Semantic Evaluation (2015)
12. Xu, R., Gui, L., Xu, J., Lu, Q., Wong, K.-F.: Cross lingual opinion holder extraction based
on multi-kernel SVMs and transfer learning. World Wide Web 18(2), 299316 (2015)
13. Baidal, K., Vera, C., Avilés, E., Espinoza, A., Zambrano, J., Tapia, E.: Sentiment Analysis in
Education Domain: A Systematic Literature Review, CCIS 883, pp. 158173. Springer
14. Mostafa, L., Farouk, M., Fakhry, M.: An automated approach for webpage classication. In:
ICCTA09 Proceedings of 11th International Conference on Computer Theory and
Applications, Alexandria, Egypt (2009)
15. Lenco, D., Meo, R.: Towards the automatic construction of conceptual taxonomies, vol.
5182. Springer (2008)
16. Anitha, N., Anitha, B.: Sentiment classication approaches a review. Int. J. Innov. Eng.
Technol. 3,2231 (2013)
17. Zhang, H.: The optimality of Naive Bayes. Am. Assoc. Artif, Intell (2004)
18. Tang, D., Qin, B., Liu, T.: Document modeling with gated recurrent neural network for
sentiment classication. In: Proceedings of the 2015 Conference on Empirical Methods in
Natural Language Processing, pp. 14221432 (2015)
Student Sentiment Analysis Using Gamication 337
19. Mostafa, L., Abd Elghany, M.: Investigating game developersguilt emotions using
sentiment analysis. Int. J. Softw. Eng. Appl. (IJSEA) (2018). ISSN: 09759018
20. Mostafa, L.: A Survey of Automated Tools for ARAB CHAT ALPHABET. Lab LAMBERT
Academic Publishing (2016). ISBN:978-3-330-01845-7
21. Romero, C., Ventura, C., Garcıa, E.: Data mining in course management systems: moodle
case study and tutorial. Comput. Educ. 5(1), 368384 (2008)
22. Li, L.: Sentiment-enhanced learning model for online language learning system (2018)
23. Permana, F., Rosmansyah, Y., Abdullah, A.: Naive Bayes as opinion classier to evaluate
students satisfaction based on student sentiment in Twitter Social Media. In: The Asian
Mathematical Conference 2016 (AMC 2016) (2016)
24. Werbach, K., Hunter, D.: For the Win: How Game Thinking Can Revolutionize Your
Business. Published by Wharton Digital Press, The Wharton School, University of
Pennsylvania, 3620 Locust Walk - 2000 Steinberg Hall-Dietrich Hall, Philadelphia, PA
19104 (2012)
25. Martins, H., Freire, J.: Games people play creating a framework for the gamication of a
masters course in a Portuguese University. In: Proceedings of ICERI 2015 Conference 16th
18th November 2015, Seville, Spain, pp. 28212827 (2015). ISBN: 978-84-608-2657-6
26. Surendeleg, G., Murwa, V., Yun-Kyung, H., Kim, Y.: The role of gamication in education
a literature review. Contemp. Eng. Sci. 7(29), 16091616 (2014)
27. Hamari, J., Koivisto, J., Sarsa, H.: Does gamication work?A literature review of
empirical studies on gamication. In: 47th Hawaii International Conference on System
Science, pp. 30253034, 978-1-4799-2504-9/14 (2014)
28. Alčaković, S.: Millennials and gamication a model proposal for gamication application
in Članci/papers millennials and gamication a model proposal for gamication
application in tourism destination (2018)
29. Mostafa, L., Elbarawy, A.: Enhance job candidate learning path using gamication. In:
ICCTA18 Proceedings of 28th International Conference on Computer Theory and
Applications, Alexandria, Egypt (2018)
30. McGonigal, J.: Reality is Broken: Why Games Make Us Better and How They Can Change
The World. Penguin Group, New York (2011)
31. Urh, M., Vukovic, G., Jereb, E., Pintar, R.: The modeleducation. Procedia Soc. Behav. Sci.
197, 388397 (2015).
32. Sobocinski, M.: Necessary denitions for understanding gamication in education a short
guide for teachers and educators (2018)
33. Tsai, M.-J., Huang, L.-J., Hou, H.-T., Hsu, C.-Y., Chiou, G.-L.: Visual behavior, ow and
achievement in game-based learning. Comput. Educ. 98, 115129 (2016)
34. Tejada-Castro, M., Aguirre-Munizaga, M., Yerovi-Ricaurte, E., Ortega-Ponce, L.,
Contreras-Gorotiza, O., Mantilla-Saltos, G.: Funprog: a gamication-based platform for
higher education, CCIS 883, pp. 255268. Springer (2018)
35. Şahin, M., Arsla, N.: Gamication and effects on studentsscience lesson achievement. Int.
J. New Trends Educ. Their Implications 7(1), 4147 (2016). Article: 04, ISSN 1309-6249
36. Martí-Parreñoa, J., Seguí-Masa, D., Seguí-Masb, E.: Teachersattitude towards and actual
use of gamication. Procedia Soc. Behav. Sci. 228(18770428), 682688 (2016)
37. ASU Online Pilots Games for Environmental Science.
newsroom/asu-online-pilots-games-environmental-science. Accessed 21 Sept 2018
38. De-Marcos, L., García-Lopez, E., García-Cabot, A.: Social network analysis of a gamied e-
learning course: small-world phenomenon and network metrics as predictors of academic
performance. Comput. Hum. Behav. 60, 312321 (2016)
338 L. Mostafa
39. Müllera, B., Reise, C., Seligera, G.: Gamication in factory management education a case
study with Lego Mindstorms. In: 12th Global Conference on Sustainable Manufacturing,
Procedia CIRP 26, pp. 121126 (2015)
40. Briers, B.: The gamication of project management. In: Proceedings of PMI Global
Congress, New Orleans, Louisiana, New Orleans (2013)
41. Kechaou, Z., Alimi, A.: Improving e-learning with sentiment analysis of usersopinions. In:
IEEE Global Engineering Education Conference Learning Environment Ecosystem for
Engineering Education, pp. 10321038 (2011)
42. Yu, L., Lee, C., Pan, H., Chou, C., Chao, P., Chen, Z., Tseng, S., Chan, C., Lai, K.:
Improving early prediction of academic failure using sentiment analysis on self-evaluated
commen. J. Assess. Learn. (2018)
43. Lin, Z., Deng, K., Hong, Y.: Research of web pages categorization. In: IEEE International
Conference on Granular Computing, GRC, GRC 2007, p. 691 (2007)
44. Wang, S., Jing, F., He, J., Du, Q., Zhang, L.: IGroup: presenting web image search results in
semantic clusters. In: ACM CHI, pp. 377384 (2007)
45. Zhang, J., Qin, J., Yan, Q.: The role of URLs in objectionable web content categorization.
In: IEEE/WIC/ACM International Conference on Web Intelligence, 1822 December 2006
46. Gyongyi, Z., Molina, H., Pedersen, J.: Web content categorization using link information,
Technical report, Stanford University (2006)
Student Sentiment Analysis Using Gamication 339
... In addition, it provides a high level of citizen engagement and enjoyment. Gamification strategies are used in different fields including education [2,21], business context [3,4], job recruitment [20,22] and government [11,15]. ...
... Gamification is used in fields like education, job recruitment, human computer interaction and business [2,3,19,21,22]. In the field of human computer interaction, gamification a well-established technique. ...
... In education, a sentiment analysis classifier was designed to analyze the sentiments of students while using gamification tools in an educational course [21]. The author concluded that students used gamification tools had highest grade that is equivalent to A+ in the credit hour system when compared to other group that study without gamification tool. ...
Full-text available
Over the last decade, governments manage their services in a digital form. Government works to enhance communication with citizens. Furthermore, gamification tools spread quickly and affect many fields such as education, professional work and life activities. It is essential to design a gamification model to help governments provide their services to citizens in a digital fun procedure. This paper presents a new E-government gamification model that helps citizens manage the government series digitally and in the form of a game. The proposed approach has four basic modules: create citizen account, citizen pay bills, calculate points and transfer points to citizen. This paper pioneers governmental gamification models since the pre-existing literature is limited.KeywordsGamificationEgyptE-governmentDigitalization
... The emotions of students are categorized in [16]. Students' opinions are selected when using gamification techniques in the learning process. ...
... Pre-processing is the process of cleaning and removing unneeded words [16]. Porter Stemmer is used in University Selection Model. ...
... Precision and recall levels are higher than negative data set precision and recall that agreed with the results in [18,21]. The classifiers' accuracy is defined as "To what extent the classifier classifies the new document correctly [16]. Table 4 represents the classifiers' accuracy level, which shows that the NB classifiers level is higher than SVM, which is agreed with [17][18][19]. ...
Full-text available
College admission is a decision that affects the student’s career life. Students have to think about all available options and draw their career path before selecting the college. High-school graduates must commit to several years until graduation before starting careers. University Selection decision depends on student interests, student academic results, student family standard of living, student education language. Students usually consult their friends, school teachers, and family members to select the university; however, sometimes, students after graduation make a career shift and start from the beginning if they make a wrong university selection decision. The research aims to define the factors that affect students’ decision to choose the university and predict student decisions based on testing cases using machine learning techniques. One thousand two hundred applicants were questionnaire. An expert model uses Support Vector Machine (SVM) and Naive Bayes (NB) classifications algorithms’s. Results had shown that students with high school programs (British-IG) use an Individual ecological system. Students with National high school program decisions are affected by their exosystem, while American high school program students’ decisions are affected by their parents and relatives, including their Microsystems. NB had shown better accuracy, recall, and precision values compared to SVM.
... Sentiment analysis is the process of understanding and classifying the texts written by people to represent their opinions and emotions and sentiments related to entities [12]. Sentiment analysis is used in different sectors such as tourism [12], the education sector [13][14][15], and game developers sentiment analysis [16]. ...
... Social Media allows all users to write their sentiments to represent their emotions and opinions. The sentiment analysis process consists of three processes: sentiment extraction, keyword preparation, review analysis, and classification [13][14][15][16]. The researcher in [12] proposes a Traveler Review Sentiment Classifier that provides a classification for traveler's reviews on five Aswan -Egyptian Hotels. ...
... Students' sentiments are classified in [14] to understand their opinion using Gamification techniques in the education process. Two hundred student sentiments were analyzed using term frequency selection techniques, and three machine learning techniques, including Naive Bayes (NB), SVM, and decision tree. ...
Full-text available
Due to Covid19 Pandemic, all processes turn into a digital form that leads to online recruitment growth. Job portals are job recruiting systems. Human resources employees suffer from many job applications and the interviewing process and select the best candidate for the job vacancy. This paper presents the Job Candidate Rankmodel that analyzes the interviewers’ sentiment and calculates a rank for each job candidate to help Human resources employees make the hiring decision. The proposed approach analyses the interviewers’ sentiment and create a list of ranked keywords and then classified them using machine learning techniques. The results show that the Job Candidate Rank approach provides an accuracy of 93%.
... Several studies relate sentiment polarity with students' motivation: learners with a positive attitude are more confident and motivated to learn (Mostafa, 2020;Weston et al., 2020in Okoye et al., 2020. However, we argue for a more grounded interpretation of these SA results. ...
... SA techniques are mostly used to assess the emotional climate of students in higher education, as in Spatiotis et al. (2018), or related to one particular topic, such as the use of Mobile phones in learning, as in Abdulsalami et al. (2017). Following this approach, in some studies the emotional climate assessment is carried out to (2015) and Mostafa (2020), the implementation of practical and hands-on activities in Suwal and Singh (2018), or the comparison between online or hybrid learning in Camacho and Goel (2018). These studies evidence that, although SA needs to overcome important limitations mentioned in previous sub-sections, SA can help higher education teachers and researchers to have quick results of the implementation of educational innovations complementary to traditional students' gradings, releasing research teams from the burden of analyzing a big amount of data in a first screening phase. ...
Full-text available
Sentiment Analysis (SA), a technique based on applying artificial intelligence to analyze textual data in natural language, can help to characterize interactions between students and teachers and improve learning through timely, personalized feedback, but its use in education is still scarce. This systematic literature review explores how SA has been applied for learning assessment in online and hybrid learning contexts in higher education. Findings from this review show that there is a growing field of research on SA, although most of the papers are written from a technical perspective and published in journals related to digital technologies. Even though there are solutions involving different SA techniques that can help predicting learning performance, enhancing feedback and giving teachers visual tools, its educational applications and usability are still limited. The analysis evidence that the inclusion of variables that can affect participants’ different sentiment expression, such as gender or cultural context, remains understudied and should need to be considered in future developments.
... Authors used the TF-IDF Algorithm to measure the similarity and calculate sentiment for each word and decide if the result is positive posts, negative posts, or neutral posts. Another contribution that used sentiment analysis techniques is presented in (Mostafa, 2020) where authors proposed a Sentiment Analysis Classifier that can determine the students' sentiments while using gamification tools in an educational course. ...
... Text mining techniques can also be used to analyze the feedbacks and messages posted by students to measure their levels of knowledge construction and cognitive skills (Roseli & Umar, 2015a) and predict students' emotions while learning (Giang et al., 2020), and while using gamification tools in an educational course (Mostafa, 2020) with the goal to provide guidance for tutors to help them with interventions (X. Wang, Zhao, Huang, Zhu, & Tang, 2019a). ...
The increasing interest in recent years in Educational Data Mining (EDM) has promoted the enhancement of educational settings by developing novel approaches. The wide number of research that has been done in this context has focused mainly on improving the courses’ recommendation based on students’ needs by building online-based educational systems that rely on personalization. Such an adaptive system requires the construction of an efficient student model that represents the students’ characteristics and preferences. Student modeling is considered as one of the most important EDM applications, this task aims to predict both the students’ performance and the various students’ features based on which the adaptation can be done. In this thesis, we were interested in proposing new automatic approaches to predict the student’s performance and the learning style (LS). LS remains one of the major student’s characteristics that affect learning achievement since it helps systems to personalize the learning process according to learners’ preferences. Learning style refers to the preferred way in which an individual learns better. The traditional method to detect learning styles depends on asking learners to self-evaluate their own attitudes and behaviors through surveys and questionnaires. This approach presents several weaknesses including the lack of self-awareness of learners of their own preferences. Furthermore, the vast majority of learners experience boredom when they are asked to fill out the corresponding questionnaire. Besides that, the traditional approach assumes that learning styles are fixed, and cannot change over time. In this thesis, we have proposed a generic approach for detecting learning styles automatically according to a given learning style model. In fact, our approach does not depend on a specific learning style model (LSM). This approach consists of two major steps. The first step aims to extract learning sequences from log files using web usage mining techniques, and then the extracted learners’ sequences were grouped according to a specific learning style model using clustering algorithms. The second step aims to build a predictive model by applying classification techniques on the results obtained from the clustering step since those results can be considered as a training data set of labeled sequences. To perform our approach we used Felder-Silverman Model as LSM and Fuzzy C-Means, K-means and K-modes as clustering algorithms, and Naïve Bayes, ID3, CART, and C4.5 as classification techniques. We have conducted an experimental study using a real-world dataset. The obtained results show that our approach outperforms the traditional approach and provides promising results. Besides the learning style, we were also interested in dealing with the student’s performance prediction task, in this context we have proposed a methodology to build a student’s performance prediction model using the multiple linear regression (MLR) analysis method. Our methodology consists of three major steps, the first step aims to analyze and preprocess the students’ attributes/variables using a set of statistical analysis methods, and then the second step consists in selecting the most important variables using different methods. The third step aims to construct different MLR models based on the selected variables and compare their performance using the k-fold cross-validation technique.
... For example, the authors of [27] investigated student satisfaction with massive open online courses (MOOCs) by employing supervised machine learning models to identify the course features, where the capability evaluation indicated an F-score of 88.32% for student satisfaction for learning via video instruction. In comparison, the authors of [28] focused on using sentiment classification to enhance higher education standards by adopting machine learning as the classifier to isolate students' comments, achieving an accuracy of 83% for classification performance. ...
... Education [27,28] -These collected datasets came from student comments after registering for the course. The proposed technique can conduct sentiment multi-classification based on the directed weighted ability to identify the product review with good results. ...
Full-text available
Three-dimensional (3D) technology has attracted users’ attention because it creates objects that can interact with a given product in a system. Nowadays, Thailand’s government encourages sustainability projects through advertising, trade shows and information systems for small rural entrepreneurship. However, the government’s systems do not include virtual products with a 3D display. The objective of this study was four-fold: (1) develop a prototype of 3D handicraft product application for smartphones; (2) create an online questionnaire to collect user usage assessment data in terms of five sentiment levels—strongly negative, negative, neutral, positive and strongly positive—in response to the usage of the proposed 3D application; (3) evaluate users’ sentiment level in 3D handicraft product application usage; and (4) investigate attracting users’ attention to handicraft products after using the proposed 3D handicraft product application. The results indicate that 78.87% of participants’ sentiment was positive and strongly positive under accept using 3D handicraft product application, and evaluations in terms of assessing attention paid by participants to the handicraft products revealed that positive and strongly positive sentiment was described by 79.61% of participants. The participants’ evaluation results in this study prove that our proposed 3D handicraft product application affected users by attracting their attention towards handicraft products.
... To target the problems of students' evasion, disengagement, and lack of motivation in educational environments, recent research has been using gamification 1 ) along with its activities (Battistella & von Wangenheim, 2016;Legaki et al., 2020;Oliveira et al., 2022). Gamification in education usually aim to improve students' concentration, engagement, performance, and/or decrease students' frustration and demotivation in educational systems (Cózar-Gutiérrez & Sáez-López, 2016;Shi & Cristea, 2016;Mostafa, 2019;Lopes et al., 2019;Metwally et al., 2020). Overall, these studies are implementing and evaluating the use of gamification in educational environments (Oliveira & Bittencourt 2019;Toda et al., 2019aToda et al., , 2019b. ...
Full-text available
Gamification has been widely used to design better educational systems aiming to increase students’ concentration, motivation, engagement, flow experience, and others positive experiences. With advances in research on gamification in education, over the past few years, many studies have highlighted the need to tailor the gamification design properties to match individual students’ needs, characteristics and preferences. Thus, different studies have been conducted to personalize the gamification in education. However, the results are still contradictory and need to be better understood to advance this field. To provide a complete understanding of this research domain, we conducted a systematic literature review to summarize the results and discussions on studies that cover the field of tailored gamified education. Following a systematic process, we analysed 2108 studies and identified 19 studies to answer our research questions. The results indicate that most of the studies only consider students’ gamer types to tailor the systems, and most of the experiments do not provide sufficient statistical evidence, especially regarding learning performance using tailored gamified systems. Based on the results, we also provided an agenda with different challenges, opportunities, and research directions to improve the literature on tailored gamification in education. Our study contributes to the field of gamification design in education.
... In higher education, sentiment analysis has been used to analyse lecture and peer assessment in Massive Open Online Courses (MOOCs), and, how students react to online educational videos see [39], [40]. Sentiment analysis has also been used in evaluating students' perceptions on learning strategies, learning experiences and learning outcomes; as well as the learning experience gained from using a chatbot to support learning a language in an informal learning space See [41], [42]- [44]. ...
Full-text available
This article presents the results of a study aimed at understanding the value of lecture recordings to student learning. We analysed transcripts of discussions on social media (Facebook) that students generated on the value of lecture recordings. Students discussed whether recording lectures and making them available should be compulsory. While the efficacy of lecture recording has been studied using conventional methods (e.g. questionnaires and interviews) on highly structured data, we employed social network and sentiment analysis techniques to examine individual messages posted on the Student Union’s Facebook page. We chose to employ social network and sentiment analysis because these methods are useful in examining semi-structured and unstructured social media data. Overall findings suggest students generally view lecture recordings as resources for supplementing live lectures rather than replacing them. Students stated that lecture recordings could facilitate the creation of an inclusive learning environment and inculcate a positive learning experience. Work presented in this article adds to the growing debate on the institutional deployment of lecture recordings and their impact on students’ engagement and learning. It also demonstrated how educational researchers could utilise social network and sentiment analysis to examine critical issues in education.
With the fast growth of e-commerce, a large number of products are sold online, and a lot more people are purchasing products online. People also give feedback of product purchased in the form of reviews. The user-generated reviews for products and services are largely available on the Internet. To extract the valuable understanding, classification of reviews is required from a huge set of feedbacks which have been converted into positive and negative sentiments. The process of sentiment analysis (SA) has mined the attitude, opinions and emotions spontaneously from text, speech and database via natural language process (NLP). It contains feedback review about product, product features or some sentiment emotional views on the product given by the customer. In this research work, feedback from the customer which is associated with smart phones is taken from in order to predict the rating of the product given by the user feedback using SA. Feedback review of the customers has been collected from, and this research work had nearly 4000 customer feedback reviews based on related categories, namely, ID of the product, name of the product, name of the brand, rating, review of the product and vote based on review. This kind of analysis will be helpful for the customers to identify the better product with quick analysis and find the unspoken commodity. Perhaps the e-commerce sector will increase revenue by giving discounts on unique tacit goods.KeywordsCustomer reviewFeedbackSentiment analysisSmart phoneProduct
Full-text available
In the last decade, sentiment analysis has been widely applied in many domains, including business, social networks and education. Particularly in the education domain, where dealing with and processing students’ opinions is a complicated task due to the nature of the language used by students and the large volume of information, the application of sentiment analysis is growing yet remains challenging. Several literature reviews reveal the state of the application of sentiment analysis in this domain from different perspectives and contexts. However, the body of literature is lacking a review that systematically classifies the research and results of the application of natural language processing (NLP), deep learning (DL), and machine learning (ML) solutions for sentiment analysis in the education domain. In this article, we present the results of a systematic mapping study to structure the published information available. We used a stepwise PRISMA framework to guide the search process and searched for studies conducted between 2015 and 2020 in the electronic research databases of the scientific literature. We identified 92 relevant studies out of 612 that were initially found on the sentiment analysis of students’ feedback in learning platform environments. The mapping results showed that, despite the identified challenges, the field is rapidly growing, especially regarding the application of DL, which is the most recent trend. We identified various aspects that need to be considered in order to contribute to the maturity of research and development in the field. Among these aspects, we highlighted the need of having structured datasets, standardized solutions and increased focus on emotional expression and detection.
Full-text available
there is a gap between the graduates learning path and job market requirements. Different recruitments WebPages focus on the match between the candidate and the job without focusing on defining the actual level of the candidate in the career and the whether this candidate can enhance his learning path to match more job vacancies or even match different job field. Gamification is used nowadays to motivate the learner to study and even it is used in working daily activities. In this paper a job matching model is proposed (matching the candidate personal and educational details with Information and Communication Technology - National Competency framework (ICT-NCF) with the ability to enhance the candidate learning path that will develop also his career path using Gamification tools.
Full-text available
Gamification is an approach that uses game design elements in nongame contexts. The gamification approach has been successfully applied to a variety of different contexts such as tourism, architecture, and education. In Colombia and Ecuador, there are several works that have generated great contributions to the gamification domain. However, in South America, and specifically in Ecuador, there are few higher education applications based on the gamification approach. In this sense, this work presents Funprog, a gamification-based platform for higher education that aims to generate an emotional and social impact on students. Funprog defines a set of game levels where students face new challenges that allow them to obtain more knowledge and improve their skills. Funprog was used by first-year students from the Agrarian University of Ecuador. Specifically, this application was focused on the teaching of the Programming Fundamentals subject. Finally, a set of surveys were conducted to know the level of acceptance of Funprog among students and teachers. The surveys’ results denote a clear acceptance of this application.
Full-text available
Abstract: A wide range of scientific and professional literature deals with the topic of gamification in tourism, defining it as a new trend. In this paper a model for applying the concept of gamification at the level of the tourism destination is proposed. As a basis, and an introduction to the analysis in this paper, millennials are selected as a market segment with high dynamic growth on the tourism market. The role and importance of millennials for tourism and destinations are especially reflected in their traveling habits, lifestyle and dedication to the use of new technologies in everyday life, and in particular playing games. The commitment of this tourism demand segment to gaming activities, as well as the fact that millennials will make a significant part of the tourism market in the future, points to the need to establish a link between millennials, gamification and its application which may provide multiple benefits for both tourists and tourism destinations. Those benefits are reflected in the improvement of the overall tourist experience as a fundamental product in tourism, on the one hand, and in the improvement of all tourism destination elements.
Full-text available
With the rapid development of the Internet, online learning is gradually taking the place of part of the offline learning. At the same time, a wide range of methods have been adopted to improve the e-learning system, such as semantic technologies, sentiment computing, forgetting curve. There are still some defects in the existing e-learning models when used in the online language learning. (1) The resource organization module is inefficient when facing a large amount of various language learning resources. (2) The sentiment module is too complex to apply to those simple online learning systems. To solve the above-mentioned problems, we propose a sentiment-enhanced learning model for the online language learning system, which consists of a knowledge-flow-based learning resource organization module, a time-decayed user profile module, and a simple and effective sentiment analysis module. With the support of these modules, the online learning system can organize the language learning resources, model the learners’ characters, and intervene in the language learning process effectively. Finally, the experimental results show that the proposed method provides more effective means for bettering the efficiency of online language learning.
Full-text available
Students activity on social media can provide implicit knowledge and new perspectives for an educational system. Sentiment analysis is a part of text mining that can help to analyze and classify the opinion data. This research uses text mining and naive Bayes method as opinion classifier, to be used as an alternative methods in the process of evaluating studentss satisfaction for educational institution. Based on test results, this system can determine the opinion classification in Bahasa Indonesia using naive Bayes as opinion classifier with accuracy level of 84% correct, and the comparison between the existing system and the proposed system to evaluate students satisfaction in learning process, there is only a difference of 16.49%.
This study presents a model for the early identification of students who are likely to fail in an academic course. To enhance predictive accuracy, sentiment analysis is used to identify affective information from text‐based self‐evaluated comments written by students. Experimental results demonstrated that adding extracted sentiment information from student self‐evaluations yields a significant improvement in early‐stage prediction quality. The results also indicate the limited early‐stage predictive value of structured data, such as homework completion, attendance, and exam grades, due to data sparseness at the beginning of the course. Thus, applying sentiment analysis to unstructured data (e.g., self‐evaluation comments) can play an important role in improving the accuracy of early‐stage predictions. The findings present educators with an opportunity to provide students with real‐time feedback and support to help students become self‐regulated learners. Using the exploring results for improvement in teaching and learning initiatives is important to maintain students' performances and the effectiveness of the learning process.