ArticlePDF Available

A Sentiment Analysis System to Improve Teaching and Learning

Authors:

Abstract and Figures

Natural language processing and machine learning can be applied to student feedback to help university administrators and teachers address problematic areas in teaching and learning. The proposed system analyzes student comments from both course surveys and online sources to identify sentiment polarity, the emotions expressed, and satisfaction versus dissatisfaction. A comparison with direct-assessment results demonstrates the system's reliability.
Content may be subject to copyright.
36 COMPUTER PUBL ISHED B Y THE IEEE C OMPUT ER SOCIE TY 0018-9162/17/$33.00 © 2017 IEEE
COVER FEATURE ADVANCES IN LEARNING TECHNOLOGIES
Sujata Rani and Parteek Kumar, Thapar University
Natural language processing and machine learning can be
applied to student feedback to help university administrators
and teachers address problematic areas in teaching
and learning. The proposed system analyzes student
comments from both course surveys and online sources
to identify sentiment polarity, the emotions expressed, and
satisfaction versus dissatisfaction. A comparison with direct-
assessment results demonstrates the system’s reliability.
Sentiment analysis (SA) is the process of iden-
tifying and classifying users’ opinions from
a piece of text into dierent sentiments—for
example, positive, negative, or neutral—or
emotions such as happy, sad, angry, or disgusted to
determine the user’s attitude toward a particular sub-
ject or entity. SA plays an important role in many elds
includ ing education , where student fe edback is essen tial
to assess the eectiveness of learning technologies.
Many universities obtain such feedback via a student
response system (SRS) during or at the end of a course to
analyze the teacher’s performance.1 Student feedback
about teacher performance, the learning experience,
and ot her course att ributes can a lso be gathere d through
social media. In recent years, online learning portals
like Coursera (www.coursera.org) have attracted many
students by providing free courses from a growing num-
ber of selected institutions.2 Millions of students join
these massive open online courses each year and share
their opinions about the course content and quality of
teaching on the course’s discussion forum. Students
also comment about their educational experiences in
blogs, online forums such as College Condential (www
.collegecondential.com), and teacher review sites
such as Rate My Professors (www.ratemyprofessors
.com).3 This feedback not only yields useful insights
A Sentiment Analysis
System to Improve
Teaching and Learning
MAY 2 01 7 37
for university administrators and
instructors but also plays a key role
in inuencing student decisions
on which universities to attend or
courses to take.4
SENTIMENT ANALYSIS
Course outcomes can be assessed
directly or indirectly. Direct assess-
ment considers samples of actual stu-
dent work including exams, assign-
ments, quizzes, and project reports.
Indirect assessment is based upon
student observations of the learn-
ing experience and teaching qual-
ity. SA of student feedback is a form
of indirect assessment that analyzes
text written by students—whether
in formal course surveys or infor-
mal comments from online platforms—to determine stu-
dents’ interest in a class and to identify areas that could be
improved through corrective actions.
SA raises many technical challenges. First, word mean-
ing varies across dierent doma ins. For example, in an edu-
cation context the word “early” connotes a negative sen-
timent in the sentence “The lecture is too early!” but in a
consumer context it connotes a positive one in the sentence
“The courier arrived early.” Second, performing SA on text
in dierent languages can be dicult. In India, for exam-
ple, people often express their opinions using a transliter-
ated form of Hindi; thus, they might write
,
which translates into English as “He teaches very well in
the class,” as “Wo class mein achha padhate hain.” These
types of challenges motivate the need to develop a context-
sensitive, multilingual SA system.
Most SA st udies have foc used on user-review cor pora—for
example, product, movie, and hotel reviews—with research-
ers generally classifying the reviews into positive, negative,
and sometimes neutral. SA has not been extensively applied
to education, though work in this area has grown recently as
described in the “Related Research” sidebar. However, most
of these approaches limit the classication of sentiments to
the two or three categories indicated above, without consid-
ering the wide range of emot ions that can also aect st udent
feedback. Moreover, they do not process multilingual data.
Finally, previous researchers have not attempted to validate
their systems by comparing the results of their analysis with
those of traditional direct- assessment methods.
PROPOSED SA SYSTEM
Our proposed SA system helps to improve teaching and
learning by performing temporal sentiment and emotion
ana lysis of mult ilin gual st udent feedb ack in term s of teacher
performance and course satisfaction. The system classies
sentiments into two categories, positive and negative, and
emotions into Robert Plutchik’s eight categoriesanger,
antic ipation, disg ust, fear, joy, sad ness, surpr ise, and tr ust—
from which it computes satisfaction or dissat isfaction.
Figure 1 shows the system architecture, which has ve
main components: data collection, data preprocessing,
sentiment and emotion identication, satisfaction and dis-
satisfaction computation, and data visualization. The sys-
tem uses t he open source R la nguage (ww w.r-project.org) to
perform data preprocessing and sentiment classication.
Data collection
Our initial data corpus consists of student feedback about
a Coursera course as well as data obtained from a univer-
sity SRS. The Coursera dataset includes approximately
4,000 student comments made during the course, which
ran from August 2015 to August 2016, and 1,700 student
comments made after completion of the course. The SRS
Feedback
data
Tokenization Sentiment and
emotion identication
Token matching
Vector creation
Emotions Sentiment
Data visualization
NRC
Lexicon
Lowercasing
Normalization
Stemming
Removal of
irrelevant content
Transliteration
Data preprocessing
Satisfaction and
dissatisfaction computation
Feedback
data
Data collection
FIGURE . Proposed sentiment analysis (SA) system architecture. After preprocessing
input data—student feedback obtained from both formal sources such as course surveys
and informal sources such as blogs and forums—the system uses natural language pro-
cessing in conjunction with the NRC Emotion Lexicon to classify sentiments and emotions.
Sentiments are classified into two categories, positive and negative, and emotions are
classified into one of eight categories—anger, anticipation, disgust, fear, joy, sadness,
surprise, and trust—from which the system computes satisfaction or dissatisfaction. The
SA system can process multilingual content and includes a data-visualization component
to facilitate analysis.
38 COMPUTER WWW.COMPUTER.ORG/COMPUTER
ADVANCES IN LEARNING TECHNOLOGIES
dataset includes about 500 student comments and ratings
for lecture and lab sessions after midterm and nal semes-
ter examinations for a course taught by one teacher over
the past 10 years. It also includes student surveys and com-
ments for 25 courses taught by dierent teachers at the uni-
versity over the past 2 years, which we used in conjunction
with d irect assessments of s tudent perfor mance to eva luate
the system’s reliability.
Data preprocessing
Dur ing this ph ase, the SA sys tem prepares col lected dat a for
fur ther processing. This involves six steps.
Tokenization. Students’ comments are split into words, or
tokens, using the tokenize function in R.
Lowercasing. Characters are converted to lower case to
ease the process of matching words in student comments to
words in the NRC Emotion Lexicon.5 This step is performed
using the tm_map function in R’s tm package.
Normalization. Abbreviated content is normalized by
using a dictionary to map the content to frequently used
Internet slang words. For example, “gud” and “awsm” are
mapped to “good” and “awesome,” respectively.
Stemming. To fu rther fac ilitate wor d matchin g, words in st u-
dent comme nts are conver ted to their roo t word using the tm_
map function in R’s SnowballC package. For example, “mov-
ing,” “moved,” and “movement” a re all converted to “move.”
Removal of irrelevant content. Punctuation and stop
words, which are irrelevant for SA, are removed to improve
system response time and eect iveness.
Transliteration. To address the issue of use of mixed
RELATED RESEARCH
In recent years, researchers have begun to apply
sentiment analysis (SA) to the education field
using various machine learning and natural lan-
guage processing techniques.
In 2011, Zied Kechaou, Mohamed Ben Ammar,
and Adel M. Alimi performed sentiment classi-
fication of e-learning blogs and forums using a
supervised hybrid technique that combined hidden
Markov models with support vector machines
(SVMs). They performed experiments using three
feature-selection methods—mutual information,
information gain, and chi statistics—and determined
that the chi-statistics method outperformed the
other two.1
Two years later, Myriam Munezero and her
colleagues performed emotion analysis of
student learning diaries and classified them into
Robert Plutchik’s eight emotion categories. They
also computed frustration and anxiety from
these eight emotions.2
In 2014, Nabeela Altrabsheh, Mihaela Cocea,
and Sanaz Fallahkhair performed SA of student
feedback using naive Bayes (NB), complement
NB (CNB), SVM, and maximum-entropy classifi-
ers using unigrams as features. They concluded
that an SVM with a radial basis function kernel
and the CNB technique achieved good results for
real-time feedback analysis. They also observed
better performance without including the neutral
class.3
The following year, Trisha Patel, Jaimin
Undavia, and Atul Patel analyzed feedback from
meetings of students’ parents using the General
Architecture for Text Engineering (GATE) tool and
its ANNIE application to classify comments as
positive or negative.4
Several studies were published in 2016.
Francis F. Balahadia, Ma. Corazon G. Fer-
nando, and Irish C. Juanatas proposed an SA sys-
tem to evaluate teacher performance in courses
from student responses in both English and
Filipino. They calculated sentiment scores from
qualitative and quantitative response ratings us-
ing an NB algorithm and graphically represented
the percentage of positive and negative senti-
ments to help university administrators be aware
of students’ concerns.5
V. Dhanalakshmi, Dhivya Bino, and A.M.
Saravanan performed SA on feedback from a
student evaluation survey of Middle East College in
Oman. They used the RapidMiner tool to clas-
sify the comments into positive and negative on
the basis of features like teacher, exam, module
content, and resources. The researchers compared
MAY 2 01 7 39
language in student comments, the text is transliterated
using the Google Transliterate API.
Sentiment and emotion identification
Dur ing thi s phase, the SA s ystem an alyzes t he preprocessed
data to identif y insta nces of sentiment and emot ion. It uses
the NRC Emotion Lex icon, also known as EmoLex, to a sso-
ciate words with positive or negative sentiment and the
eight basic emotions. The lexicon supports  languages
including several Indian ones like Hindi, Tamil, Gujarati,
Marathi, and Urdu. It includes annotations for , uni-
gram words for English and , for Hindi.
Each word in the lexicon has an emotion vector (
E
)
containing a Boolean value (b) for each sentiment (s) and
emotion (e):

=+
∈∈∀∈
EE
E
EbbEbb b
,
wh
ere{,, }and {,}, {0,1}.
es
esi
07 89
If a word in a student comment matches a word in the
lexicon , the corres ponding emot ion vector is re turned ; if the
word matc hes more than one word i n the lexicon, t he sum of
the corresponding emotion vectors is returned. In this way,
an emot ion vector is crea ted for each commen t representin g
the dierent emotions and sentiments contained within.
For example, for the sentence “Sir, you are great!” the SA
system would ret urn the following emotion vector:
Anger
Anticipation
Disgust
Fear
Joy
Sadness
Surprise
Trust
Negative
Positive
000000010 1
This equates to the positive sentiment, as trust and positive
parameters have a b value equal to .
the performance of their approach using NB,
SVM, k-nearest neighbors, and neural-network
classifiers.6
Brojo Kishore Mishra and Abhaya Kumar Sahoo
used CUDA C programming with a GPU architec-
ture to evaluate faculty performance. They cate-
gorized faculty members as excellent, very good,
good, average, or poor on the basis of average
marks given by students in feedback form. The
researchers favorably compared their approach in
terms of time execution to a similar performance
evaluation using a CPU architecture.7
Guadalupe Gutiérrez Esparza and her col-
leagues proposed a model for SA of student
tweets about teacher performance in Spanish.
They used an SVM algorithm to classify the
tweets into positive, negative, and neutral; they
also proposed a syntactic pat tern model to com-
pare results using SVMs and syntactic pat terns.8
References
1. Z. Kechaou, M.B. Ammar, and A.M. Alimi, “Improving E-learning
with Sentiment Analysis of Users’ Opinions,Proc. IEEE Global
Eng. Education Conf. (EDUCON 11), 2011, pp. 1032–1038.
2. M. Munezero et al., “Exploiting Sentiment Analysis to Track
Emotions in Students’ Learning Diaries,” Proc. 13th Koli Calling
Int’l Conf. Computing Education Research, 2013, pp. 145 –152.
3. N. Altrabsheh, M. Cocea, and S. Fallahkhair, “Learning
Sentiment from Students’ Feedback for Real-time Interven-
tions in Classrooms,” Adaptive and Intelligent Systems, A.
Bouchachia, ed., LNCS 8779, Springer, 2014, pp. 40– 49.
4. T. Patela, J. Undavia, and A. Patela, “Sentiment Analysis
of Parents Feedback for Educational Institutes,” Int’l J.
Innovative and Emerging Research in Eng., vol. 2, no. 3,
2015, pp. 75–78.
5. F.F. Balahadia, M.C.G. Fernando, and I.C. Juanatas,
“Teacher’s Performance Evaluation Tool Using Opinion
Mining with Sentiment Analysis,” Proc. IEEE Region 10
Symp. (TENSYMP 16), 2016, pp. 95–98.
6. V. Dhanalakshmi, D. Bino, and A.M. Saravanan, “Opinion
Mining from Student Feedback Data Using Supervised
Learning Algorithms,” Proc. 3rd MEC Int’l Conf. Big Data
and Smart City (ICBDSC 16), 2016; doi:10.1109/ICBDSC
.2016.74 60390.
7. B.K. Mishra and A.K. Sahoo, “Evaluation of Faculty Perfor-
mance in Education System Using Classification Technique
in Opinion Mining Based on GPU,” Computational Intelli-
gence in Data Mining, vol. 2, H. Behera and D. Mohapatra,
eds., AISC 411, Springer, 2016, pp. 109–119.
8. G.G. Esparza et al., “Proposal of a Sentiment Analysis
Model in Tweet s for Improvement of the Teaching-
Learning Process in the Classroom Using a Corpus of
Subjec tivit y,” Int’l J. Combinatorial Optimization Problems
and Informatics, vol. 7, no. 2, 2016, pp. 22–34.
40 COMPUTER WWW.COMPUTER.ORG/COMPUTER
ADVANCES IN LEARNING TECHNOLOGIES
To enable tempora l analys is of sentime nts and emot ions,
the SA sy stem generates a m ean emotion vec tor (
Ej
) for each
month and year:
E
nEE NN
1,where .
ji
n
j
p
ji ji
11
=∑∑∀
∈≥
=
=

0
00
Here, n represents the number of comments in each month
and year and p represents the emotion and sentiment para-
meters such that p {anger, anticipation, disgust, fear, joy, sad-
ness, surprise, trust, negative, positive}. This vector is created to
avoid any anomalies that might result from an increase in
the value of a particular emotion in that month or yea r.
Satisfaction and dissatisfaction computation
Satisfaction and dissatisfaction are crucial parameters
in education. The SA system derives these from six of the
eight emotion parameters—namely, joy, trust, anticipation,
anger, disgust, and sadness. Anticipation and trust clearly
connote satisfaction, but in some circumstances joy could
have a negative connation—for example, a student could
feel joy at skipping a boring class. Therefore, in computing
student satisfaction, we multiply the sum of anticipation
and trust by a constant (α = .) to give these parameters
more weight . We employ the same mec hanis m in computi ng
student dissatisfaction to give more weight to anger and
disgust than to sadness.
The calculations are as follows:
Satisfaction = [α(TA) + ( – α)(J)]/n
Dissatisfaction = [α(AD) + (– α)(S)]/n,
where TA = trust + anticipation, J = joy, AD = anger + disgust,
S= sadness, and n = max(TA or AD, J or S).
Consider two examples. For the sentence “He is good at
teaching,” the SA system returns the following emotion
vector from the NRC lexicon:
Anger
Anticipation
Disgust
Fear
Joy
Sadness
Surprise
Trust
Negative
Positive
0100101 101
Here, TA = , J = , and n = max(TA , J) = . Satisfaction is thus
calculated a s [.() + .()]/ = ./ = .. For the sentence
“He is bad at teaching and every student has doubts about
the class,” the sy stem returns the fol lowing emotion vector:
Anger
Anticipation
Disgust
Fear
Joy
Sadness
Surprise
Trust
Negative
Positive
1 0 1 2 0 2 0 1 2 0
Lectures
Labs
99
97
95
93
91
89
87
85
2006
Overall rating (%)
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Year(a)
100
90
80
70
60
50
40
30
20
10
0
Comments and rating (%)
Year(b)
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Positive
comments
Negative
comments
Positive
rating
Negative
rating
FIGURE 2. Temporal sentiment analysis. (a) Student ratings of a teacher’s performance in lectures and lab sessions of one course over
a 10-year period. Students rated the performance in lectures slightly higher, and average overall performance exceeded 90 percent
during the last six years. (b) Percentage of positive and negative student comments about and ratings of the same teacher; on average,
85 percent of comments were positive and 15 percent were negative.
MAY 2 01 7 41
In this case, AD = 2, S = 2, and n = max(AD, S) = 2. Dis satis-
fact ion is therefore ca lculated a s [0.6(2) + 0.4(2)]/2 = 2.0/2 = 1.
Data visualization
To facilitate analysis of student feedback about course
satisfaction and teacher performance, our SA system has a
data-visualization component that creates sentiment and
emotion word clouds as well as line graphs of changes in
sentiments and emotions over time.
Sentiment and emotion word clouds. Students use a
variety of words to convey their sentiments or emotions
while g iving feedback. Visualizing f requently used positive
words (“great,” “excellent,” interesting,” and so on) and
negative words (“dull,” “confusing,” “terrible,” and so on) in
the form of word clouds can help identify student learning
behavior—for example, whether or not they are taking an
interest in lectures and lab sessions.
Temporal sentiment and emotion analysis. As indicated
earlier, our SA system groups together positive and
negative comments and ratings in student feedback by
month and year. This makes it possible to track teacher
performance and course satisfaction over time. Figure 2a
plots overall st udent ratings (ranging from 0 to 100 percent)
of one teacher’s performance in lectures and lab sessions of
a university course from 2006 to 2016; the graph shows that
stude nts rated th e teacher’s perf ormance in le ctures s lightly
higher than that in lab session s and that the average overall
rating was more than 90 percent during the last six years.
Figure 2b plots the percentage of positive and negative
student comments about and ratings of the teacher over the
same period; the graph reveals that, on average, 85 percent
of comments were positive and 15 percent were negative.
Sentiment polarity can also be tracked across dierent
teachers and courses over time to analyze overall teaching
quality at a given institution.
Our SA system also groups together emotions identied
in comments about courses and teachers by month and
year, providing more granular insight. Figure 3a plots the
percentage of emotions extracted from student feedback
on a one-year Coursera course by month; the graph shows
Anger Anticipation Disgust Fear Joy
Sadness Surprise Trust Satisfaction
Aug. 2016
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Month
(a) Year
(b)
70
75
65
60
55
50
45
40
35
30
25
15
20
10
5
0
Emotions (%)
56
52
48
44
40
36
32
28
24
20
12
16
8
4
0
Aug. 2015
Sep. 2015
Nov. 2015
Emotions (%)
Oct. 2015
Dec. 2015
Feb. 2016
Jan. 2016
Mar. 2016
May 2016
Apr. 2016
Jun. 2016
Jul. 2016
FIGURE 3. Temporal emotion analysis. (a) Percentage of emotions extracted from student feedback on a one-year Coursera course.
Students expressed positive emotions more than negative ones, signaling satisfaction with the experience. (b) Percentage of emotions
extracted from student comments about the teacher in Figure 2. Trust in the instructor gradually increased, and each year the percentage
of positive emotions exceeded that of negative emotions.
42 COMPUTER WWW.COMPUTER.ORG/COMPUTER
ADVANCES IN LEARNING TECHNOLOGIES
that students expressed the positive emotions of trust, joy,
anticipation, a nd surprise more than the negative emot ions
of sadness, fear, disgust, and anger. Figure 3b plots the
percentage of emotions extracted from student comments
about the teacher from Figure 2; it shows that students’
trust in the instructor gradually increased over the decade
and that each year the percentage of positive emotions
exceeded that of negative emotions. In both datasets, about
55 percent of students were satised with the teacher.
SYSTEM EVALUATION
In education, there is a general consensus that direct and
indirect assessments of teaching quality and learning
behavior should agree. Students who perform well in a
course, for example, would be expected to give the teacher
high ratings and favorable comments; conversely, those
who perform poorly are likely to be dissatised.
To validate our SA system, we analyzed student surveys
and comments obtained from a university SRS system for
25 dierent courses over a two-year period and compared
the percentage of positive sentiments students had about
each course with the average course grade on a 0–100 scale.
As Figure 4 shows, the results generally agreed, with less
than 20 percent absolute dierence between the methods.
In those courses where student performance exceeded
satisfaction, there could be a number of explanations: the
exams were relatively easy, the course had a particularly
bright or hard-working group of students, or students did
not like the teacher for personal reasons or felt they did
not gain much value from the class. In those courses where
student satisfaction exceeded performance, perhaps the
exams were exceptionally challenging or students failed to
adequa tely prepare. In e ither case, t he discrepa ncy in resu lts
obtai ned from both approaches inv ites continued analysis.
Our proposed SA system has great potential to
improve teaching and learning in universities by
analyzing sentiment, emotion, and satisfaction
parameters in student feedback to help administrators and
teachers understand problematic areas and take corrective
actions. The large volume of information contributed by
students to course surveys, discussion forums, blogs, and
other sources is a largely underutilized resource that can be
eec tivel y leveraged with the application of machi ne learning
techn iques, which a re continua lly improv ing. A compari son of
our proposed system’s results with direct assessments of class
perfor mance demonstrates it s reliabil ity.
Despite its promise, the system has some limitations. It
is only as good as the data it analyzes, so care must be taken
in collecting feedback from students. SRSs must be well
designed to ensure that they are engaging, and instructors
must ma ke a concert ed eort to en sure th at as ma ny student s
as possible provide complete and acc urate feedback.
In future work, we plan to adapt the SA system API
to integrate with SRSs and online learning portals to
enable real-time analysis of student feedback. We will
also add other Indian languages to extend the system’s
multilingual capabilities.
REFERENCES
1. N. Altrabsheh, M.M. Gaber, and M. Cocea, “SA-E: Sentiment
Analysis for Education,” Intelligent Decision Technologies,
R. Neves-Si lva et al., eds., FAI A 255, IOS Press, 2013,
pp.353–362.
2. M. Wen, D. Yang, and C.P. Rosé, “Sentiment Analysis in
MOOC Discussion Forums: What Does It Tell Us?,” Proc. 7th
Int’l Con f. Educat ional Data Mining (EDM 14), 2014; www
.cs.cmu.edu/~mwen/papers/edm2014-camera-ready.pdf.
100
95
90
85
80
75
70
65
60
55
Course
C1
C3
C5
C7
C9
C17
C19
C21
C23
C25
C11
C13
C15
Percentage
(a)
100
95
90
85
80
75
70
65
60
Course
C1
C3
C5
C7
C9
C17
C19
C21
C23
C25
C11
C13
C15
Percentage
(b)
Class performance
Course survey
Class performance
SRS comments
FIGURE 4. Comparison of student performance, quantified as
average class grade on a 0–100 scale, with the percentage of
positive sentiments in (a) surveys and (b) comments obtained
from a university student response system (SRS) for 25 different
courses over a two-year period. The results of the two methods
generally agreed.
MAY 2 01 7 43
3. B.K. M ishra and A.K. Sahoo, “Evalu ation of Facult y
Perfor mance in Education System Using Classication
Technique in Opinion Mining Based on GPU,” Computational
Intelligence in D ata Mining, vol. 2, H. Behera and D.
Mohapat ra, eds., AISC 411, Springer, 2016, pp. 109–119.
4. A. Abdelrazeq et al., “Senti ment Analysis of Social Media
for Evaluating Univer sities,” Proc. 2n d Int’l Conf. Digital
Information Pro cessing, Data Mining , and Wireless Comm.
(DIPDMWC 15), 2015, pp. 49–62.
5. S. M. Mohammad and P.D. Turney, “Crowdsourcing a
Word–Emotion Association Lexicon ,”Computational
Intelligence, vol.29, no. 3, 2013, pp. 436–465.
ABOUT THE AUTHORS
SUJATA RANI is a research scholar in the Department
of Computer Science and Engineering at Thapar Uni-
versity. Her research interests include natural language
processing (NLP) and machine learning. Rani received
an ME in computer science and engineering from
Thapar University. She is a member of ACM. Contact her
at sujata.singla@thapar.edu.
PARTEEK KUMAR is an associate professor in the
Department of Computer Science and Engineering at
Thapar University. His research interests include NLP,
databases, and machine learning. Kumar received a
PhD in NLP from Thapar University. He is a member of
ACM. Contact him at parteek.bhatia@thapar.edu.
NOMI NATE A COLLE AGUE FOR THI S AWAR D!
DUE: 15 OCTOBER 2 017
CALL FOR STANDARDS AWARD NOMINATIONS
IEEE COMPUTER SOCIETY HANS K ARLSSON
STANDARDS AWARD
Submit your nomination elect ronically: awards.computer.org | Questions: awards@computer.org
• Requires 3 endorsements.
• Self-nomi nations are not accepted.
• Do not need I EEE or IEEE Computer Society membership to apply.
A plaque and $2,000 honorarium is presented in recognition of
outstanding skills and dedication to diplomacy, team facilitation, and
joint achievement in the development or promotion of standards in the
computer industry where individual aspirations, corporate competition,
and organizational rivalry could otherwise be counter to the bene t
of society.
Read your subscriptions through
the myC S publications portal at
http://mycs.computer.org
... Sentiment analysis is increasingly used in natural language processing (NLP) research and is an important component of data mining (Liu, 2012). Generally, SA has five stages: data collection, data processing, sentiment and emotion identification, satisfaction/dissatisfaction computation and data visualization (Rani and Kumar, 2017). Detecting the subjectivity or polarity of documents, this method retains the subjective nature of opinions and emotions through a computational systematic application of textual analysis and classification. ...
... Originally widely used in consumer behavior research, SA is now applied in educational settings to generate a rich understanding of learner attitudes (Peng et al., 2020). As a general practice, this approach classifies text comments into positive, negative and neutral categories, comparing ratios or extracting meaning (Rani and Kumar, 2017;Toço glu and Onan, 2020). Yoo et al. (2012) employ such an approach to extract emotional features from students' queries relating to performance. ...
Article
Purpose Student motivation underpins the challenge of learning, made more complex by the move to online education. While emotions are integral to students' motivation, research has, to date, overlooked the dualistic nature of emotions that can cause stress. Using approach-avoidance conflict theory, the authors explore this issue in the context of novel online students' responses to a fully online class. Design/methodology/approach Using a combination of critical incident technique and laddering, the authors implemented the big data method of sentiment analysis (SA) which results in approach tables with 1,318 tokens and avoid tables with 1,090 tokens. Using lexicon-based SA, the authors identify tokens relating to approach, avoid and mixed emotions. Findings The authors implemented the big data method of SA which results in approach tables with 1,318 tokens and avoid tables with 1,090 tokens. Using lexicon-based SA, the authors identify tokens relating to approach, avoid and mixed emotions. These ambivalent emotions provide an opportunity for teachers to rapidly diagnose and address issues of student engagement in an online learning class. Originality/value Results demonstrate the practical application of SA to unpack the role of emotions in online learner motivation.
... Sentiment analysis later identifies and classifies learners' opinion into for example, positive, negative, or neutral states that showing learner's attitude toward a particular course. Meanwhile, thematic analysis can describe the "pattern" and "theme" of the learners' opinions (Rani & Kumar, 2017). The discovery of the underlying pattern of thematic analysis may help teacher to prioritize the corrective actions in course alignment. ...
... Sentiment analysis, also known as opinion mining, is a fundamental task in natural language processing (NLP) that involves the process of analyzing, processing, generalizing and reasoning network public opinions. Sentiment analysis has a wide range of application in domains such as the financial [32], e-commerce [33,34] and assessment teaching quality [35] domains, among others. The main methods of sentiment analysis are based on a sentiment lexicon, which in turn is based on deep learning, a subset of machine learning. ...
Article
Full-text available
In recent years, the price of small agricultural products has both plummeted and skyrocketed, which has a great impact on people’s lives. Studying the factors affecting the price fluctuation of small agricultural products is of great significance for stabilizing their price. With the development and application of social media, farmers and consumers are more greatly influenced by online public opinion, resulting in irrational planting behavior or purchasing behavior, which has a complex impact on the price of small agricultural products. Taking garlic as an example, we crawled through network public opinions about garlic price from January 2015 to December 2020 using web crawler technology. Then, the network public opinions were quantified using a natural language processing and time-varying parameter vector autoregression (NLP-TVP-VAR) model to empirically analyze their dynamic influence on garlic price fluctuation. It was found that both public attitude and public attention have a short-term influence on garlic price fluctuation, and the influences of each differ according to direction, intensity and timing. The influence of public attitude on garlic price fluctuation is positive, while the influence of public attention on garlic price fluctuation is largely negative. The influence intensity of public attitude is stronger than of public attention on garlic price fluctuation. The influence of public attitude on garlic price fluctuation shows a trend of intensifying, while that of public attention has been weaker than in previous years. In addition, based on the results of our study, we present some recommendations for improving the comprehensive information platform and price fluctuation early warning system for the whole industry chain of small agricultural products.
... We chose a Distance Education course, as it is a growing application area for Sentiment Analysis (Altrabsheh et al., 2014;Bóbó et al., 2019;Rani & Kumar, 2017;Shapiro et al., 2017). Some studies show that a student's Emotional State is essential, as it directly impacts the possibility of school dropout (Feng et al., 2016;Lei et al., 2015;Pong-Inwong & Rungworawut, 2014). ...
Article
Dropping out of school comes from a long-term disengagement process with social and economic consequences. Being able to predict students' behavior earlier can minimize their failures and disengagement. This article presents the SASys architecture, based on a lexical approach and a polarized frame network. Its main goal is to define the author's sentiment in texts and increase the assertiveness of detecting the sentence's emotional state by adding authors' information and preferences. The author's emotional state begins with the phrase extraction from Virtual Learning Environments; then, pre-processing techniques are applied in the text, which is submitted to the complex frame network to identify words with polarity and the author's text sentiment. The flow ends with the identification of the author's emotional state. The proposal was evaluated by a case study, applying the Sentiment Analysis approach to the students' school dropout problem. The results point to the feasibility of the proposal for asserting the student's emotional state and detection of students' risks of dropout.
... But due to the large number of lectures and students, it is often impossible to analyze each of the comments manually. Thus, many research papers focus on how to automate this process in order to extract meaningful information from students' feedback (e.g., (Rani and Kumar, 2017;Kandhro et al., 2019;Sindhu et al., 2019;Rakhmanov, 2020a)). SA in education generally analyzes such sentiments with machine learning techniques and lexicon-based approaches. ...
Conference Paper
Full-text available
We describe our work on sentiment analysis for Hausa, where we investigated monolingual and cross-lingual approaches to classify student comments in course evaluations. Furthermore, we propose a novel stemming algorithm to improve accuracy. For studies in this area, we collected a corpus of more than 40,000 comments-the Hausa-English Sentiment Analysis Corpus For Educational Environments (HESAC). Our results demonstrate that the monolingual approaches for Hausa sentiment analysis slightly outperform the cross-lingual systems. Using our stemming algorithm in the pre-processing even improved the best model resulting in 97.4% accuracy on HESAC.
... At the same time, there are not many excellent Chinese flute teachers, let alone very systematic training and teaching. is makes most Chinese flute learners have poor finger flexibility, inaccurate pronunciation, incorrect breathing methods, and many other basic skills [2,3]. Kneading is one of the most important skills and contents in flute playing. ...
Article
Full-text available
Influenced by cultural background, economic development, social system, education system, and other factors, there is still a big gap between Chinese institutions and developed countries in flute teaching, even with our neighbors, South Korea and Japan. Under the influence of cultural background, economic development, social system, and educational system, there is still a very big gap between Chinese colleges and universities and developed countries in flute teaching, even with our neighbors, South Korea and Japan. Because of its local perception and weight-sharing structure, the convolutional neural network is closer to the biological neural network in the real world. The weight-sharing structure reduces the complexity of the neural network, which can avoid the complexity of feature extraction and classification process in data reconstruction. This paper studies the analysis and optimization of flute playing and teaching system based on a convolutional neural network. By applying local perception field and parameter sharing in a convolutional neural network at the same time and adding multiple filters, it can not only effectively reduce the number of parameters but also extract features layer by layer. In the process of convolution, the parameters of the characteristic map obtained by each layer decrease layer by layer, but the number increases gradually. Based on the analysis of the problems faced by the flute performance teaching, this paper puts forward the corresponding solutions in order to promote the flute performance teaching in China to achieve better results.
... • If negation words were not handled efficiently, then it would definitely show its impact on sentiment classification. • Identifying the exact context of the word is difficult as the words take different meanings at different points of the sentences as observed by [26]. • In some machine learning-based approach, it fails to handle new data as its availability as labeled data is required [15]. ...
Article
Full-text available
Text generation is one of the complex tasks associated with natural language processing. For efficient text generation, syntax and semantics of the language have to be considered to assign context to key phrases. The main objective of the proposed work is to perform text generation specifically for movie scripts. The training data consist of a self-annotated corpus of movie scripts depicting scenes, specific to certain genre where the annotation mainly focuses on a specific director’s movie scripts. The scene generation is set forth by word embedding with sentiment classification where the emotionally analyzed words are vectorized using the EmoVec algorithm performing sentiment analysis. Based on the sentiment and location associated with each scene, context for the phrases is identified and proceeded to build a well-defined script. Bidirectional long short-term memory BLSTM with multi-head attention is used to capture the information processed in both forward and backward propagation in order to understand future context. The vocabulary is built using Stanford’s Internet Movie Database IMDB datasets to perform word-based encoding for which requirement of an extensive vocabulary is imminent.
Chapter
Educational data mining is a research field that is used to enhance education system. Research studies using educational data mining are in increase because of the knowledge acquired for decision making to enhance the education process by the information retrieved by machine learning processes. Sentiment analysis is one of the most involved research fields of data mining in natural language processing, web mining, and text mining. It plays a vital role in many areas such as management sciences and social sciences, including education. In education, investigating students' opinions, emotions using techniques of sentiment analysis can understand the students' feelings that students experience in academic, personal, and societal environments. This investigation with sentiment analysis helps the academicians and other stakeholders to understand their motive on education is online. This article intends to explore different theories on education, students' learning process, and to study different approaches of sentiment analysis academics.
Chapter
Educational data mining is a research field that is used to enhance education system. Research studies using educational data mining are in increase because of the knowledge acquired for decision making to enhance the education process by the information retrieved by machine learning processes. Sentiment analysis is one of the most involved research fields of data mining in natural language processing, web mining, and text mining. It plays a vital role in many areas such as management sciences and social sciences, including education. In education, investigating students' opinions, emotions using techniques of sentiment analysis can understand the students' feelings that students experience in academic, personal, and societal environments. This investigation with sentiment analysis helps the academicians and other stakeholders to understand their motive on education is online. This article intends to explore different theories on education, students' learning process, and to study different approaches of sentiment analysis academics.
Article
Full-text available
In this paper, we propose a sentiment analysis model for the assessment of teacher performance in the classroom by tweets written by a pilot group of college students. Naive Bayes (NB) is the technique to be applied to classify tweets based on the polar express emotion (positive, negative and neutral), to carry out this process, a dataset fits adding distinctive terms of context as possible features to support the classification process.
Conference Paper
Full-text available
Large amount of data is available in the form of reviews, opinions, feedbacks, remarks, observations, comments, explanations and clarifications. In Education system, main focus is given to quality of teaching. That quality depends on coordination among teacher and student. Feedback analysis is more important to measure the faculty performance. Performance of faculty should be evaluated so that we can enhance our education quality. To measure the performance of faculty, we use classification technique by using opinion mining. We also use this technique on GPU architecture using CUDA-C programming to evaluate performance of a faculty in very less time. This paper uses opinion mining concept with GPU to extract performance of a faculty.
Conference Paper
Full-text available
Learning diaries are instruments through which students can reflect on their learning experience. Students' sentiments, emotions, opinions and attitudes are embedded in their learning diaries as part of the process of understanding their progress during the course and the self-awareness of their goals. Learning diaries are also a very informative feedback source for instructors regarding the students' emotional well-being. However the number of diaries created during a course can become a daunting task to be manually analyzed with care, particularly when the class is large. To tackle this problem, in this paper we present a functional system for analyzing and visualizing student emotions expressed in learning diaries. The system allows instructors to automatically extract emotions and the changes in these emotions throughout students' learning experience as expressed in their diaries. The emotions extracted by the system are based on Plutchik's eight emotion categories, and they are shown over the time period that the diaries were written. The potential impact and usefulness of our system are highlighted during our experiments with promising results for improving the communication between instructors and students and enhancing the learning experience.
Conference Paper
Full-text available
Educational data mining (EDM) is an important research area that is used to improve education by monitoring students performance and trying to understand the students’ learning. Taking feedback from students at the end of the semester, however, has the disadvantage of not benefiting the students that have already taken the course. To benefit the current students, feedback should be given in real time and addressed in real time. This would enable students and lecturers to address teaching and learning issues in the most beneficial way for the students. Analysing students’ feedback using sentiment analysis techniques can identify the students’ positive or negative feelings, or even more refined emotions, that students have towards the current teaching. Feedback can be collected in a variety of ways, with previous research using student response systems such as clickers, SMS and mobile phones. This paper will discuss how feedback can be collected via social media such as Twitter and how using sentiment analysis on educational data can help improve teaching. The paper also introduces our proposed system Sentiment Analysis for Education (SA-E).
Article
Full-text available
Even though considerable attention has been given to the polarity of words (positive and negative) and the creation of large polarity lexicons, research in emotion analysis has had to rely on limited and small emotion lexicons. In this paper we show how the combined strength and wisdom of the crowds can be used to generate a large, high-quality, word-emotion and word-polarity association lexicon quickly and inexpensively. We enumerate the challenges in emotion annotation in a crowdsourcing scenario and propose solutions to address them. Most notably, in addition to questions about emotions associated with terms, we show how the inclusion of a word choice question can discourage malicious data entry, help identify instances where the annotator may not be familiar with the target term (allowing us to reject such annotations), and help obtain annotations at sense level (rather than at word level). We conducted experiments on how to formulate the emotion-annotation questions, and show that asking if a term is associated with an emotion leads to markedly higher inter-annotator agreement than that obtained by asking if a term evokes an emotion.
Conference Paper
The research aims to develop a teacher's performance evaluation tool using opinion mining with sentiment analysis. The study may help to identify the strengths and weaknesses of the faculty members based on the positive and negative feedback of the students either in English or in Filipino language. The proposed system provides the sentiment score from the qualitative data and numerical response rating from the quantitative data of teachers evaluation. It will also graphically represent the evaluation result including the percentage of positive and negative feedback of the students. Thus, the school administrators and educators will be more aware about the sentiments and concerns of the students.