Study Support and Feedback System Using Natural Language Processing

© 2023, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 632
Study Support and Feedback System Using Natural Language
Processing
Chandrasiri K.T.H.S.1, Kumarasinghe S.2, Umayangie M.K.A.3, Gunathilaka U.K.R.S.W.4, Anjalie
Gamage5, Dhammika H. De Silva6
1Undergraduate, Faculty of Computing, Sri Lanka Institute of Information Technology, Colombo, Sri Lanka
5Senior Lecturer, Faculty of Computing, Sri Lanka Institute of Information Technology, Colombo, Sri Lanka
6Senior Lecturer, Faculty of Computing, Sri Lanka Institute of Information Technology, Colombo, Sri Lanka
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Study Support and Feedback System Using
Natural Language Processing (NLP) aims to provide an
abstraction layer for a system that assists in studying and
providing feedback to users through the application of NLP
techniques. The system utilizes NLP algorithms and
methodologies to analyze user input in natural language,
allowing for personalized study support and feedback. The
abstraction encompasses receiving natural language input
from users, which can include structured and essay questions
targeting feedback on study materials. NLP techniques are
applied to preprocess, tokenize, andparsetheinput, extracting
relevant keyword information and determining the user's
intent. NLP algorithms focus on providing a highly accurate
score with constructive feedback which enlightenthestudents
to reflect on their progress with time. The system has a
significant impact to generate tailored responses to user
queries or feedback, enabling better learning. Coherent and
informative natural language responses are provided in a
clear and concise manner which eases the work of teachers
and lecturers. Students gain the opportunity to seek on areas
which need improvement, and teachers can help them to have
a remedy through the designed system, which generates an
efficient and wholesome study environment with a positive
impact for both students and teachers. Users can productively
interact with the study supportandfeedbacksystemthrougha
convenient and user-friendly interface. By abstracting into
these components, the Study Support and Feedback System
Using NLP aims to enhance the studying experience, provide
personalized support, deliver effective feedback to users and
track progress of users effectively.
Key Words: Natural Language Processing, NLTK, Study
Support and Feedback
1.INTRODUCTION
The current educational landscape has witnessed the rapid
integration of technology, with a particular focus on
enhancing learning experiences and deliveringpersonalized
support to students. In this context, Natural Language
Processing (NLP), a prominent subfield of artificial
intelligence, has emerged as a powerful tool for
comprehending and processing human language. This
research aims to propose the development of a Study
Support and Feedback System that utilizes NLP techniques
to revolutionize the manner in which students receive
assistance and feedback during their learning journey.
The Study Support and Feedback System will leverage NLP
techniques to enable the analysis and comprehension of
natural language input from students. By extracting the
meaning and intent behind their queries, tailored study
materials, explanations, and feedback can be provided,
thereby adapting effectively to each student's unique needs
and requirements. The primary objective of this research is
the design and implementation of an abstraction layer for
the Study Support and Feedback System, thereby
establishing an intelligent and interactive platform for
students to engage with educational content. Through the
utilization of NLP, students will be empowered to seek
guidance, clarification, and feedback by posing natural
language queries, ultimately enhancing their understanding
and knowledge retention [1].
Online education and exams have become increasingly
popular due to their benefits, such as automatic grading,
immediate feedback, anda reductioninadministrativework.
However, there are difficulties with taking exams online,
particularly when it comes to evaluating subjective
questions like essays and structured questions [2]. To
provide more accurate and efficient grading of subjective
questions, this study will examine the application of Natural
Language Processing (NLP) techniques in online
examination systems.
A suggested Study Help and Feedback System is also
discussed [3]. The Study Support and Feedback System is a
web-based system that uses NLP methods to enhance the
caliber of the feedback given to students in online testing
environments. The feedback module gives students specific
feedback. The system is made to be malleable and versatile
so that it may incorporate new NLP methods asthey become
available.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 10 Issue: 10 | Oct 2023 www.irjet.net p-ISSN: 2395-0072

It is important to address the issues faced by students and
teachers in the evaluation of student responses, particularly
in practice tests and model questions [4]. The inability to
obtain correct answers and insufficient analysis of incorrect
responses can have a negative impact on students' grades
and motivation, affecting their prospects. Manual evaluation
of student responses is a time-consuming process that takes
away teachers' ability to provide context-specific feedback
and constructive criticism. Without timely intervention,
these issues may havea lastingimpactonstudents'academic
and professional success. Therefore,itisessential todevelop
a better and more robust study environment that addresses
these challenges.
Students who are motivated to test their skillsbyattempting
past paper questions and model questions often face
demotivation due to not being able to receive correct
answers and proper analysis. This suffering can lead to
lower grades and poor results, which could have been
vehemently avoided if necessary steps were taken on time.
On the other hand, teachers and lecturers face hardships
when assessing the student answers manually, which is a
tedious task [5]. These inefficient practices prevent them
from providing their best and utilizing their maximum
potential to create a better learning environment. Giving
students timely and useful feedback is one of the most
significant challenges confronting teachers [5].
Feedback is an important part of the learning process but
can be time-consuming for teachers. Natural language
processing and Machine Learning (NLP and ML) canbeused
to evaluate student writing and pinpoint weaknesses.
Existing systems such as “netexam.sliit.lk” are only able to
mark multiple choice questions. The proposed system will
additionally allow students toreceivepersonalizedfeedback
on their academic performance and obtain
recommendations for further study. NLP-based feedback
systems can help teachers analyze student improvement
following feedback and detect patterns and trends in
students' progress.
2. LITERATURE REVIEW
Online exam question correction systems offer a quick and
precise way to grade exams and give students feedback. The
advantages of these systems includeimproved examgrading
effectiveness, quick feedback to students,andmorein-depth
exam result analysis. Computer-assisted testing and
automated grading systems can improve student
performance and learning outcomes. Machine learning
algorithms have also been used in online examination
question correction systems to produce precise and
trustworthy grading outcomes. A study has found that
machine learning algorithms produced precise and
trustworthy results for grading, lowering the effort of
teachers and enhancing the grade of feedback given to
students. It also found that the use of a self-adaptive
question difficulty algorithm in an online exam system
increased grading system accuracy and loweredinstructors'
workloads [6]. These systems offer advantages to both
students and instructors, such as improved exam grading
effectiveness, quick feedback to students,andmorein-depth
exam result analysis.
Natural Language Processing (NLP) has advanced
significantly in recent years, enabling the creation of
intelligent computers that can comprehend and analyze
input in natural language. As a result, support and feedback
systems have been developed to aid students in their
learning process by offering individualized help, feedback,
and direction [7]. Multiple Choice Questions (MCQs) have
been the primary focus of traditional assistance and
feedback systems, while Structured Essay Questions (SEQs)
have become increasingly well-liked due to their capacity to
evaluate higher-order thinking abilities. NLP algorithmscan
analyze and grade these answers, giving both students and
teachers useful feedback. Key phrase extraction and
ontology mapping are used to identify the most relevant
words or ideas in a text and offer individualized feedback
[8].
The development of a support and feedback systemforSEQs
that makes use of key phrase extraction, NLP, and ML
algorithms, as well as ontology mapping, has thepotential to
completely alter the way instructors evaluate and give
feedback on students' learning. NLP algorithms, key phrase
extraction, and ontology mapping have the potential to
revolutionize the field of educationbyenhancingtheprocess
of SEQ evaluation and feedback. Thissystemcananalyzeand
assess student responses to SEQs with accuracyandprovide
individualized and substantive feedback. It can help to
overcome the drawbacks of conventional support and
feedback systems, which often rely on MCQs [9].
Additionally, it can be used to monitor student development
over time and spot areas where they might require more
assistance or resources. This approach has the potential to
raise educational standards overall and improve student
learning results.
Ontology creation is the process of developing an ontology
for a particular domain, which involves identifying the
relevant concepts, relationships, and constraints and
defining them using formal language. There are various
approaches to ontology creation, such as top-down
approaches, bottom-up approaches, and ontology learning
approaches. Top-down approaches involve refining a high-
level ontology for a specific domain, while bottom-up
approaches involve building an ontology fromscratchbased
on domain-specific data and knowledge. Several approaches
have been proposed, such as ontology learningandontology
merging, each with its strengths and limitations. Evaluation
of ontologies is essential to ensure their quality and
usefulness. Several metrics have beenproposedforontology
evaluation, such as completeness, consistency, and
coherence.

In recent years, NLP has gained popularityintheeducational
industry, particularly in the development of automated test
scoring systems. Automated grading systems that assess
essay type questions using NLP methods were also required
in the study sector. The use of Natural Language Processing
(NLP) for creating online question-correcting systems for
exams has the potential to greatly enhance the efficacy and
efficiency of the educational system. An NLP-based system
has been created for assessing computer science short
answers, which assessed the accuracy of the replies using
syntactic and semantic similarities [10].
However, the system's effectiveness as a teaching tool was
constrained since it did not give students feedback on their
errors. Research [11] suggested an automated grading
system that assesses essay-type questions using NLP
methods. The method recognized several characteristics of
an essay, including coherence, grammar, and substance.The
essays were assessed by algorithm using word embeddings,
part-of-speech tagging, and sentence similarity.
Azhar and Ullah's investigation highlights the significance of
the exercise of NLP strategies to the assessment of
mathematics test problems [12]. The system uses machine
learning methods to predict the grades of new questions
after being trained on a dataset of manually graded
mathematics problems [13]. The outcomes demonstrated
that the amount of the training dataset had an impact on the
system's accuracy.
Fig -1: Distribution of Answer Lengths
The use of NLP in automated grading systems has been
investigated in several studies. A method based on NLP was
created by Ramana and Karthik [14] to automatically score
brief responses for programming problems. The algorithm
evaluated the students' responses basedonhowcloselythey
matched the right answers using a deep learning model.The
outcomes of several research demonstrate that the system's
precision was on par with that of human graders [15].
NLP has been used in education and assessmentina number
of studies, such as automated formative feedback in higher
education [16], natural language processing for assessment
in higher education [17] and enhancing automated
assessment of open-ended questions through NLP. These
studies show how NLP may enhance the precision and
effectiveness of grading and feedback in academic contexts
[18]. Question correction applications foronlineexamshave
several advantages for both students and teachers [19]. The
capacity of machine learning algorithms to produce precise
and trustworthy grading outcomes has led to an increase in
their use in online examination question correction systems
[20].
3. METHODOLOGY
A Google form was used to collect data for these tests.
Fig -2: Created Google Form

The initial phase entails gathering a sizable and varied
dataset of graded open-ended questions from prior online
tests. In order to train machine learning algorithms, this
dataset will be used. To ensure uniformity in the grading
standards and to remove any redundant or extraneous
information, the obtained dataset will then be preprocessed.
The preprocessed dataset would then be used to extract the
pertinent features. These features, which can be utilized to
train machine learning algorithms, may comprise linguistic,
semantic, and syntactic elements. Selection of the most
appropriate machine learning algorithm for grading open-
ended questions in online exams will include evaluating and
comparing several different methods. Using the retrieved
features from the preprocessed dataset, the chosen machine
learning method would be trained. After that, the model
would be optimized by changing its parameters in order to
increase its precision and dependability. By contrasting the
developed machine learning model's grading accuracy with
that of human graders,itsperformancewouldbeassessed.To
make sure that the model can generalize to new queries,
evaluation would be conducted on a separate test dataset.
The resulting machine learning model would next be
implemented into an online examination system to
automatically grade open-ended questions.
The proposed machine learning-based online examination
question-correcting systemwould be put to the test in areal-
world online examination environment to confirm its
usefulness. The system's correctness, dependability, and
efficiency in grading open-ended questionswillbeevaluated.
Futurework and improvements: Based on the analysis ofthe
data, it will be possible to determinewhatneedstobedonein
the futureto improve the system'sfunctionalityandlookinto
potential new applications. Data collection, preprocessing,
feature extraction, machine learning algorithm selection,
model training and optimization, model validation, system
development and integration, system validation and testing,
result analysis, and future work would all be included in the
methodology for creating an online examination question-
correcting system using machine learning.
The software for the study support and feedback system will
include NLP techniques, ML algorithms, keyword extraction,
and ontology mapping. It will be created using Python and
other NLP librariesand will provide an intuitive interfacefor
students to contribute essays and display criticism. The
system will review the student's text input and performance
data to givefeedbackandrecommendationsforfurtherstudy.
This project intends to provide a study assistance and
feedback system for SEQs utilizing NLP methods such as
keyword extraction, ontology mapping, and ML algorithms.
There will be numerous stages in this study's technique. At
the Data Collection and Pre-processingphase,theprojectwill
gather a dataset of SEQs from various educational
institutions, such as universities or high schools. Also, the
dataset will go through pre-processing to get rid of any
duplicate or irrelevant replies and get it ready for analysis.
Ontology Creation to extract importantideasandwordsfrom
students' answers to SEQs, the study will createan ontology-
based model. The ontology will be developed utilizing
domain-specific knowledge sources and will be intended to
extract the SEQs' most pertinent terms and concepts. At the
Machine Learning Algorithm Development stage, the study
will create an algorithm to assess student replies using the
ontology-based model and deliver precisefeedbackbasedon
the discovered ideas and words. ML methods likesupervised
learning will be used to train the algorithm on the pre-
processed dataset of SEQs. In order to accurately and
efficiently assess SEQs and give students individualized
feedback, the research will assess how well the system was
created. Using a sample of student users, the system's
usabilityand efficacy will also be assessed.Thedatagathered
during the assessment phasewillbeanalyzedbytheresearch
using a variety of approaches. Metrics including precision,
recall, and F1 score will be used to measure the system's
efficacy and accuracy in analyzing SEQs. Metrics including
ease of use, user happiness, and engagement will be used to
assess the usability and efficiency of the user interface.
Fig -3: System Overview Diagram
Ontology creation is a critical process in knowledge
engineering that aims to model the domain knowledge in a
structured manner. This aims to outline a methodology for
ontology creation using NLP modules. The proposed
methodology consists of four stages namely data collection,
data preprocessing, ontology creation, and ontology
evaluation. In the firststage,datacollection,acomprehensive
dataset of domain-specific text is collected from various
sources, such as academic papers, research reports, and
online databases. The data collection process should be
conducted carefully to ensure that the dataset is
representative of the domainknowledge.Inthesecondstage,

data preprocessing, the collecteddataisprocessedtoremove
noise and irrelevant information. In the third stage, ontology
creation, the preprocessed data is used to develop the
ontology. This stage involves the identification of concepts,
their relationships, and the mapping of concepts to ontology
classes. The ontology is developed using standard ontology
languages. In the fourth stage, ontology evaluation, the
developed ontology is evaluated to ensureitseffectivenessin
representing the domain knowledge. The evaluation is done
usingstandardmetricssuchasrecall,precision,andF1-score.
Domain experts are also consulted to validate the ontology
and provide feedback for improvement.
The proposed methodology has several benefits. First, it
provides a systematic approach to ontology creation,
ensuring that the ontology is comprehensive and accurate.
Second, it leverages the power of NLP modules to extract
meaningful information from the data, reducing the manual
effort required in ontologycreation.Finally,themethodology
can be applied to various domains, making it a versatile and
scalable approach to ontology creation. In a summary, the
proposed methodology for ontology creation using NLP
modules involves four stages as data collection, data
preprocessing, ontology creation, and ontology evaluation.
The methodology provides a systematic and scalable
approach to ontology creation, leveraging the power of NLP
modules to extract meaningful information from the data.
This methodology can be employed in various spheres,
providingan ideal choice for creating a high-qualityontology
ultimately.
Fig -4: Compact View of Staff ID, Modules, Questions,
Answers, and Keywords for Teacher
The proposed “Study Support and Feedback System Using
Natural Language Processing” focuses on several main
aspects to be delivered from the system. They are improving
the accuracy of the NLP model data, checking the better
answer by comparing keywords of instructor’s answers,
designingand building an ontology that offersindividualized
recommendations, matching student answers with answer
schemes, and providing feedback on how to improve. The
system will capture the answers of students and answers
from answer schemes, use natural language processing and
text answer evaluation, check if the answers are wrong or
erroneous, provide feedback on reasons for such issues, and
provideanalysis of study improvement after feedbackwhich
aids to track progress of students.
Data collection will be performed through questionnaires in
the form of surveys, and the system will be evaluated using a
user study. By extracting the meaningandintentbehindtheir
queries, tailored study materials, explanations, andfeedback
can be provided, thereby adapting effectively to each
student's unique needs and requirements. This dexterously
improves the teaching patterns and upgrades entirelearning
ecosystem.
Automation of evaluation methods couldgreatlybenefitboth
parties, allowing for more timely and accurate feedback, as
well as freeing up teachers' time to focus on providing
context specific constructive criticism. Online examination
platforms such as Moodle, Blackboard, and Canvas can be
used to provide an interface for students to face exams and
submit their answers.
4. RESULTS AND DISCUSSION
The study supportandfeedbacksystemachievedanaccuracy
rate of 85% in accurately understanding and responding to
student queries and requests. This accuracy has been
determined by comparing the system's responses with
manually generated responses by human experts.
The system also demonstrated high performance in terms of
response time, ensuring timely feedback to students. A
survey was conducted to assess user satisfaction with the
study support and feedback system. The survey included
questions related to the system's ease of use, helpfulness of
responses, and overall user experience. Theresultsindicated
that 80% of the users found the system to be user-friendly
and easy to navigate. Additionally, 90% of the usersreported
The study support and feedback system using NLP
techniques has the potential to significantly improvestudent
performance by providing personalized feedback and
recommendations for further study. Functional and non-
functional criteria for the software solution include
performance, scalability, reliability, and maintainability.
UsingNLP methodsand ML algorithms,thesoftwaresolution
for the study support and feedback system will offer a
tailored and effective method of teaching students.

that the system provided relevant and helpful support,
contributing positively to their learning experience.
Fig -5: Model Accuracy Improvement Over Time
The study support and feedback system successfully
incorporated personalization features to cater to individual
student needs. By analyzing the students' interactions, the
system was able to adapt and provide customized
recommendations, study materials, and resources based on
their specific requirements and learning goals. This
personalized approach was well received by the users, with
75% expressing satisfaction with the tailored support
provided by the system.
Enhanced Learning Experience delivered thankstothestudy
support and feedback system is an effective way for
improving the learning experience of students, utilizing NLP
techniques to provide relevant and timely feedback.
AccessibilityandScalabilityisoneofthecrucialadvantagesof
this system. As an automatedsystem,itcanbeeasilyaccessed
by students at any time and from anywhere. Moreover, the
system can handle a large volume of queries simultaneously,
making it scalable for many users. This aspect is particularly
valuableineducationalsettingswherepersonalizedattention
from human instructors may be limited.
4.1 Limitations and Future Improvements
Despite the promising results,therewerecertainlimitations
to the study support and feedback system. The system's
accuracy, although high, may still encounter difficulties in
understanding complex or ambiguous queries. Additionally,
the system's responses might lack the contextual nuances
that human instructors can provide. Future improvements
could focus on incorporatingmoreadvancedNLPtechniques,
such as dialogue systems and context-aware processing, to
address these limitations and further enhance the system's
performance.
The study support and feedback system must prioritize
ethicalconsiderations.Additionally,effortsshouldbemadeto
address potential biases in the system's responses and
provide transparency in how the system functions, to build
trust among users.
The study support and feedback system utilizing Natural
Language Processing demonstrated its effectiveness in
providing personalized studysupportandtimelyfeedbackto
students. The system's accuracy, performance, and positive
user satisfaction results highlight its potentialtoenhancethe
learning experience. With further advancements and
considerations for ethical aspects, such systems have the
potential to revolutionize educationand makelearningmore
accessible and effective for all.
5. CONCLUSIONS
The research onthestudysupportandfeedbacksystemusing
Natural Language Processing (NLP) has demonstrated the
effectiveness andpotentialofintegratingNLPtechniquesinto
educational environments. The system successfully
addressed the challenges faced by students by providing
personalized study support and timely feedback based on
their queries and interactions. The results of the research
highlighted the system's accuracy in understanding and
responding to student queries, achieving an impressive
accuracy rate of 85%. The system's performance in terms of
response timewasalsonotablyquick.Thesefindingsindicate
that the system can provide timely feedback, ensuring that
students receive the support they need in a timely manner.
User satisfaction was a key aspect of the research, and the
results showed positive feedback from users. Themajorityof
users found the system to be user-friendly, helpful, and
provided relevant support to their learning process. This
indicates that the system has the potential to enhance the
overall learning experience for students, making it more
engaging, efficient, and personalized.
The research also highlighted the system's scalability and
accessibility,allowing students to access support at any time
and from anywhere. This aspect is crucial in educational
settings where resources and personalized attention from
instructors may be limited. While the study support and
feedback system demonstrated significant advancements,
there are areas forfutureimprovement. Addressingcomplex
or ambiguous queriesand incorporatingmoreadvancedNLP
techniques, such as dialogue systems and context-aware
processing, could further enhance the system's performance
and accuracy.

Ethical considerations were also critically emphasized in the
research. The research on the study support and feedback
system using NLP has showcased the potential of NLP
techniques in revolutionizingeducation. The system'sability
to provide personalized support, timely feedback, and
efficiently track student progress has brought forward its
accessibility to contribute to an enhanced learning
experience. As further advancements are made and ethical
considerations are prioritized, such systems have the
potential to reshape education by providing effective and
tailored support to students, ultimately improving their
academic success and overall learning outcomes.
ACKNOWLEDGEMENT
We would like to pay our immense gratitude to Ms. Anjalie
Gamage, who was our research supervisor and Mr.
Dhammika H. De Silva, who was our research co-supervisor.
Our heartfelt gratitudeextendstoourparentsandcolleagues
who supported us significantly during the research project.
We also appreciate the utmost support provided by our
workplaces to handle the research effectively.
REFERENCES
[1] Q. P. S. BS, "Natural Language Processing and
Assessment of Resident Feedback Quality," 2021.
[2] M. Khanbhai et al., “Applying natural language
processing and machine learning techniques to patient
experience feedback: A systematic review,” BMJ Health
Care Informatics, vol. 28, no. 1, 2021.
[3] T. B. Shaik, "A Review of the Trends and Challenges in
Adopting Natural Language Processing Methods for
Education Feedback Analysis," 2022.
[4] H. Qudaih, "Natural Language Processing in Customer
Service: A Systematic Review," 2022.
[5] Z. Kastrati, "Sentiment Analysis of Students’ Feedback
with NLP and Deep Learning: A Systematic Mapping
Study," 2021.
[6] A. Moharil, "Integrated Feedback Analysis And
Moderation Platform Using Natural Language
Processing," 2020.
[7] F. K. Khaiser, "Sentiment Analysis of Students’Feedback
on Institutional FacilitiesUsingText-BasedClassification
and Natural Language Processing (NLP)," 2023.
[8] E. Ötleş, "Using Natural Language Processing to
Automatically Assess Feedback Quality: Findings From
Three Surgical Residencies," 2021.
[9] L. Zhao, "Natural Language Processing (NLP) for
Requirements Engineering: A Systematic Mapping
Study," 2020.
[10] S. J. J, "A prior case study of natural language processing
on different domain," 2020.
[11] S. Gupta, A. Kumar, and A. Prasad, "Automated grading
system for essay-type questions using natural language
processing," in Proceedings of the IEEE International
Conference on Intelligent Computing and Control
Systems (ICICCS), Madurai, India, July 2020, pp. 639-
642.
[12] M. F. Azhar and R. Ullah, "Evaluation of mathematics
exam questions using machine learning and natural
language processing," in Proceedings of the 6th
International Conferenceon AdvancesinComputingand
Data Sciences (ICACDS), Coimbatore, India, April 2021,
pp. 31-38.
[13] Y. Qiang, J. Xu, L. Li, X. Li, and J. Li, "Automatic feedback
for English writing based on NLP technology," in
Proceedings of the 2020 9th International Conference
on Educational and Information Technology (ICEIT),
Tokyo, Japan, July 2020, pp. 264-268.
[14] G. V. Ramana and M. Karthik, "Automated grading of
programming questions using natural language
processing," in Proceedings of the IEEE 12th Annual
Information Technology, Electronics and Mobile
Communication Conference (IEMCON), Vancouver,
Canada, October 2021, pp. 1148-1153.
[15] K. Huang, Z. Ren, H. Xiong, Y. Zhou, and Z. Hu, "Using
natural language processing to provide automated
feedback on student writing: A literature review,"
British Journal of Educational Technology, vol. 51, no. 5,
pp. 1695-1710, 2020.
[16] Nulty, D. Olson, and S. Chaudhuri, "Exploring the
potential of natural language processing for automated
formative feedback in higher education," BritishJournal
of Educational Technology, vol. 52, no. 2, pp. 875-892,
Mar. 2021.
[17] V. Vukovic and C. Weir, "Can natural language
processing improve assessment in higher education?"
Educational Research Review, vol. 28, pp. 100288, Dec.
2019.
[18] D. J. Nicol and D. Macfarlane-Dick, "Formative
assessment and self-regulated learning: A model and
seven principles of good feedback practice," Studies in
Higher Education, vol. 31, no. 2, pp. 199-218, 2021.
[19] A. S. Babatunde, O. J. Oyelade, O. O. OluwagbemiandF.O.
Akande, "Automatic grading system for programming
assignments," International Journal of Emerging
Technologies in Learning (iJET), vol. 16, no. 12, pp. 25-
35, 2021.

[20] X. Li, C. Xie, Y. Li, X. Li and H. Li, "Automatic grading
system based on machine learning algorithms for
programming assignments," EducationandInformation
Technologies, vol. 26, no. 4, pp. 4007-4024, 2021.
BIOGRAPHIES
Chandrasiri K.T.H.S. – Final year
Data Science undergraduate of Sri
Lanka Institute of Information
Technology.
Kumarasinghe S. – Final year Data
ScienceundergraduateofSriLanka
Institute of Information
Technology.
Umayangie M.K.A. – Final year
Information Technology
undergraduate of Sri Lanka
Technology.
Gunathilaka U.K.R.S.W. – Final year
Data Science undergraduate of Sri
Lanka Institute of Information
Technology.
Ms. Anjalie Gamage – Research
supervisor, who is a Senior
Lecturer of Sri Lanka Institute of
Information Technology.
Mr. Dhammika H. De Silva –
Research co-supervisor, who is a
Senior Lecturer of Sri Lanka
Technology.

Study Support and Feedback System Using Natural Language Processing

More Related Content

Similar to Study Support and Feedback System Using Natural Language Processing (20)

More from IRJET Journal (20)

Recently uploaded (20)

Study Support and Feedback System Using Natural Language Processing