See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/363763025

Implementation of a Web-based Chatbot Assisted Services for Results

Information System

Article · September 2022

DOI: 10.9790/0050-09013543

CITATION READS

1 1,129

2 authors:

Onuoha Chibuike Nwokonkwo Obi

The University of Aizu Federal University of Technology Owerri
4 PUBLICATIONS 2 CITATIONS 54 PUBLICATIONS 57 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Nwokonkwo Obi on 22 September 2022.

The user has requested enhancement of the downloaded file.

IOSR Journal of Mobile Computing & Application (IOSR-JMCA)
e- ISSN: 2394-0050, P-ISSN: 2394-0042.Volume 9, Issue 1 (Jan. – Feb. 2022), PP 35-43
www.iosrjournals.org

Implementation of a Web-based Chatbot Assisted Services

for Results Information System
Nwokonkwo O. C1, John-Otumu A. M2 and Onuoha, M. C3
1, 2, 3
Department of Information Technology, Federal University of Technology, Owerri, Nigeria

Abstract
An interactive chatbot enabled service for result information system was built in response to the challenges
students had dealing with the existing result information system for students' retrieval. The proposed system was
created using a natural language processing model, CSS and Tailwind for graphical design, HTML for page
alignment, and the flask framework to enable web-based chatbot development easier. According to the results,
the designed system received an overall score of 87.35 percent comprehension, 91 percent functionality, 76.63
percent speed, 80 percent interoperability, 88.05 percent engagement, and 87.23 percent scalability.
Keywords: NLP, Chatbot, Information system, Artificial Intelligence, Agents
---------------------------------------------------------------------------------------------------------------------------------------
Date of Submission: 28-02-2022 Date of Acceptance: 23-03-2022
---------------------------------------------------------------------------------------------------------------------------------------

I. Introduction
The increasingly advancement in technology has really paved way for the market demand of mobile
devices which has experienced a whole new phase over the last decades due to the exponential development and
application of intelligent features. This feature has replaced several traditional methods of handling operational
activities for schools, financial institutions, organizations, and other aspects in a society.
In recent times, Artificial intelligence (AI) is seen to be one of the most important intelligent features
that are widely used because of its ability to act and produce results like humans. A major Artificial intelligence
technique that is progressively taking the lead in both application and popularity is the use of chatbots. Chatbots
can be defined as conversational representatives that make use of natural languages (or human languages) to
conduct conversations with human beings. Chatbots can also be referred to as an AI software that is able to
communicate with its users in a natural language format through websites, messaging apps, school portals and
several other software instruments. Simply, it is a computer programming tool that simulates conversations with
humans either through textual or spoken languages. Chatbots ability to comprehend and analyze natural
languages context cannot be overemphasized as it uses an AI method known as Natural Language Processing
(NLP) for both in analysis and comprehension.
NLP, is also an AI technique that enables chatbots to simply control, understand and interpret users
queries in natural languages. It is built from the fields of AI, computer science and linguistics which aim
towards achieving easy interactions between computers and users in human languages. Importantly, there is
embedded information or large chunk of natural language data being stored in a chatbot database by the
programmer to recognize different structures of sentences (Hussain, Folu, and Shrish, 2019).
A Chatbot refers to a chatting robot. It is a communication simulating computer program, it is all about the
conversation with the user. The conversation with a Chatbot is very simple, it answers to the questions asked by
the user (Dahiya, 2017).
A dialogue system is an automated system that is developed to interact with an end user with natural language. it
renders a platform between the end user and the automated application which allows communication with the
system in a relatively natural order (Arora, Batra, and Singh, 2012).
Dialog systems consists of two types which includes, the Conversational agents which receives input and sends
feedback automatically with natural language examples include but not limited to Chatbot, Virtual personal
assistants etc., another type of a dialogue system is the Inter- active Voice Response (IVR) which allows an
automated system to communicate with the end user with voice through a keypad. An illustration of the various
type of a dialogue system is given in Figure 1.

DOI: 10.9790/0050-09013543 www.iosrjournals.org 35 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

Figure 1: An illustration of the various types of dialogue system (Candela, 2018)

1.1 Chatbot Architecture

A chatbot is made up of four major parts which includes the front-end, knowledgebase, backend, and
the data corpus (data that is used to train the chatbot). The front-end is responsible for allowing interactions
between the end user and the chatbot, it makes use of Natural Language Understanding (NLU) and Artificial
Intelligence (AI) to dictate the end-user's in- tent, the front end indicates the user whenever more information is
needed and also gives feed- back to the user's input. The knowledge base contains the knowledge residing
within a chatbot and in a format that can be displayed in the frontend (Gregori, 2017).

Figure 2: Components of a Chatbot (Gregori, 2017)

Types of Chatbots
Chatbots can be classified into different types, such as:
Rule-based chatbots: these are simple chatbots that deliver answers to questions based on pre-defined rules. It
reacts in a predetermined way to particular questions. They excel at giving users with answers to frequently
asked questions.
Intelligent Chatbots: This form of bot uses machine learning to interpret the inquiries and commands of its
users. They are programmed to understand specific terms and, over time, to learn to understand and interpret a
wider range of user questions.
AI-powered chatbots: These bots are usually cleverer in their responses to users' questions, as they combine
rule-based chatbots with intelligent software programs. They can remember discussions and grasp customer
preferences because to its AI capability. It is feasible to understand consumers using a combination of natural
language processing (NLP), machine language, and artificial intelligence (AI).
In Nigeria, educational institutions, particularly at the higher level, have shifted from paper-based information
management to web-based information management, where student and possibly school information may be

DOI: 10.9790/0050-09013543 www.iosrjournals.org 36 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

accessible via a school portal. Regardless, there is undoubtedly a series of incompetence in school portals in
accurately and promptly providing college-related information to students, which may include information about
student results, cultural activities and events, exam timetables, multiple locations and departments around
school, and guidelines on paying tuition fees; school registration (especially for newly admitted students).
While efforts have been made to allow students to access material on the internet, more effort is needed
to develop approaches that will aid in the updating of daily trends in the classroom. Furthermore, schools have
replaced this ineffectiveness with a physical conversational agent (school personnel), which will unavoidably
result in inefficiencies, rigidity in accessing information due to possible crowding or limited staff, and response
time delays. As a result, a technique is required to address these numerous setbacks faced by pupils,
necessitating the use of a chatbot.
The mechanisms through which students receive results, updated information, and rules about college-
related activities are strenuous in most Nigerian universities. Despite the university's web portal's effectiveness,
it is generally ineffective since it lacks various desirable features; hence, a chatbot-assisted service can
overcome numerous restrictions. Because of the multitude, this produces an unfriendly climate for both
employees and students. As a result of massive workloads and limited manpower, it also increases inefficiencies
and ineffective services provided by employees.
With the aforementioned issues that students encounter, the creation of a chatbot to aid students
becomes important, where the chatbot has AI competence in understanding, interpreting, and replying to student
queries efficiently and at a low cost. The major goal of this study is to create a web-based chatbot-assisted result
information system that allows students to check their grades on a specific exam, learn when an exam is
planned, and receive comments on any exam.

II. Related Works

This section discusses some related works on chatbot assisted services.
Hussain, Folu, and Shrish (2019) designed and studied the factors influencing chatbot adoption for
family planning-related information. The chatbot was created using a decision-tree structure to provide
comments on a specific topic, and the medium for communication was a mobile phone's Short Message Service
(SMS). The Unified Theory of Acceptance and Use of Technology (UTAUT) model was used to assess the
factors that influence end-users' adoption and use of chatbots. The results show that "attitude toward using
chatbot" and "reduced anxiety in acquiring information" are the two most important factors influencing chatbot
adoption for family planning-related information.
Neema (2019) created an interactive health information chatbot for non-communicable diseases using
natural language processing in Dar Es Salaam Tanzania. The chatbot was built utilizing the Facebook
Application Programming Interface (API), BotSociety, PHP, Botman Framework, and Dialogflow. Satisfaction,
usability, responsiveness, efficiency, and accuracy of the chatbot were all assessed, and the results were
positive.
Ardiana, Joni, and Udayana (2020) developed a mobile chatbot application using the Artificial
Intelligence Markup Language (AIML) approach, the HIV counselling chatbot was developed using four phases
of development; User requirement, System design, System development and System evaluation. The chatbot
application used black box testing approach to test the func- tionality of the system.
Palkar, Maurya, and Balpande (2017) suggested an intelligent career counselling bot, which was built
with the Android studio kit, a pattern matching algorithm, and a relational database management system table.
To promote drug adherence, Fadhil and Schiavo (2017) created a conversational interface. The chatbot
recognizes the time and amount of medication, and it was integrated with a web application so that the
healthcare provider may monitor the patient's condition and intervene as needed. Telegram API, Microsoft Bot
Framework, APImedic, Ruby, and Ruby on Rails framework were used to create the chatbot.
Comendador, Francisco, Medenilla, Sharleen, and Serac (2015) created "Pharmabot," a pediatric
generic drug consultant chatbot, which was implemented using Visual C# for the graphical interface (front end)
and Microsoft Access (MS Access) for the back end. The chatbot was tested on two sets of people: experts
(pediatricians) and students (pharmacy students). The results revealed that the built system is extremely user-
friendly, accurate, and consistent.
Bushra and Nadesh (2020) created a diagnostic chatbot to assist primary health care providers. The
chatbot was created using a decision tree algorithm that allows users to simulate a diagnosis of their ailment
based on the symptoms they are experiencing.
A smart health care prediction chatbot was presented by Jayashree, Monika, Preetha, and Piraisoodan
(2020). The chatbot was created using an AIML algorithm that predicts illness based on the user's symptoms
and provides contact information for a doctor for further consultation.

DOI: 10.9790/0050-09013543 www.iosrjournals.org 37 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

III. Methodology
The development specification, high-level model design, flowchart, and process flow diagrams are all covered
in this part.

Table I: System / Development Specification

S/N Environment
1 Python Programming Language v3.7
2 Hyper Text Markup Language (HTML)
3 Flask v1.1.1
4 Tailwind CSS
5 Dependencies: Numpy v1.9.2, Chatterbot v1.0.5, pyYaml v3.11, NLTK
v3.5, Chatterbot_corpus v1.2.0
6 Anaconda IDE
7 Web browser
8 Connected device

3.1 High Level Model

The high-level model of the system only depicts how the system development environment will seem.

Knowledge
Web Page
Machine
Learning Human input

Chat application
Logic and Rules Parser for
result management system
Figure 3: Anatomy of the result chatbot system

The architecture of the result chatbot system is depicted in Figure 3. The chatbot system retrieves
information from its knowledge base and connects it to the inference engine at this point. The connection
explains the relationships throughout the full evolved system. This design's dataflow easily explains the
chatbot's concept.

DOI: 10.9790/0050-09013543 www.iosrjournals.org 38 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

Figure 4: Proposed system flowchart

DOI: 10.9790/0050-09013543 www.iosrjournals.org 39 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

Get Input
Get Input from user response Text input

Process input
Select from trainer, a known statement that most closely Return a known response to the selected match from the
matches the input statement database trainer and a confidence value based on the matching

Return esponse
Return the response from the input query

Figure 5: Process flow diagram of the result assisted chatbot management system

IV. Results and Discussion

This section reports the various interfaces designed and developed as a result of research work.

Figure 6: Web view landing page

DOI: 10.9790/0050-09013543 www.iosrjournals.org 40 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

Figure 7: Mobile view landing page

Figure 8: Chatting Interface

4.1 System Performance Evaluation

This section details the system's performance as determined by the evaluation.

Table II: Performance Metric

Metrics Accuracy
Comprehension 87.35%
Functionality 91.00%
Speed 76.63%
Interoperability 80.00%
Engagement 88.05%
Scalability 87.23%

DOI: 10.9790/0050-09013543 www.iosrjournals.org 41 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

Accuracy

95.00%
90.00%
85.00%
80.00%
75.00%
70.00%
65.00%

Accuracy

Figure 9: diagram showing the performance metric

Chat Analytics

Figure 10: Line graph of sentiment analysis

Figure 8 depicts a line graph of sentiment request session and interaction counts based on sentiment result data
from the developed model

V. Conclusion
This research work centers its focal point on the usage of web based chatbot assisted services for result
information system; thus, it eliminates and replaces the traditional methods of accessing information with the
use of chatbot to provide electronic and digital services to students. Therefore, students are no longer faced with
the backlogs that that is associated with manually checking or enquiring about results information. The
development of the web based chatbot information system experienced some level of constraints which affected
the speed by which such system was built. Despite the anticipated obstacles, the web-based chatbot for
information system was designed to solve a variety of challenges for pupils. The chatbot is designed to provide
the most benefits in terms of responding to student questions and making life on campus easier in terms of
information retrieval. Deep Learning models will be used in the future to improve the design of intelligent
chatbot assisted services.

DOI: 10.9790/0050-09013543 www.iosrjournals.org 42 | Page

 Implementation of a Web-based Chatbot Assisted Services for Results Information System

References
[1]. Ardiana, D. P. Y., Joni, I. D. M. A. B., and Udayana, I. P. A. E. D. (2020). Mobile based chatbot application for HIV / AIDS
counseling using artificial intelligence markup language approach Mobile based chatbot application for HIV / AIDS counseling
using artificial intelligence markup language approach. International Journal of Physics, 1469, 2–8. https:// doi.org/10.1088/1742-
6596/1469/1/012041
[2]. Arora, S., Batra, K., and Singh, S. (2012). Dialogue Systemn: A Brief Review. 3(2), 2–5.
[3]. Bushra, K., and Nadesh, R. (2020). Design and Development of Diagnostic Chabot for supporting Primary Health Care Systems.
Nternational Conference on Computational Intelligence and Da- ta Science, 167, 75–84.
https://doi.org/10.1016/j.procs.2020.03.184
[4]. Candela, E. (2018). Consumers' perception and attitude to wards chatbots' adoption: A focus on the Italian (Master's Thesis, Aalbor
University; Vol. 1). Retrieved from https:// projekter.aau.dk/projekter/files/281244069/IMthesis
[5]. Comendador, B. E. V, Francisco, B. M. B., Medenilla, J. S., Sharleen, M. T., and Serac, T. B. E. (2015). Pharmabot : A Pediatric
Generic Medicine Consultant Chatbot. 3(2), 137–140. https:// doi.org/10.12720/joace.3.2.137-140
[6]. Dahiya, M. (2017). A Tool of Conversation : Chatbot. International Journal Of Computer Sciences and Engineering, 5(5), 2–5.
https://doi.org/2347-2693
[7]. Fadhil, A., and Schiavo, G. (2017). Designing for Health Chatbots. Retrieved from https://
www.researchgate.net/publication/331343433 designing for health chatbots&ved+2ahukewivibXhAC&usg=16209980178 67
[8]. Gregori, E. (2017). Evaluation of Modern Tools for an OMSCS Advisor Chatbot. Retrieved from https://www.researchgate.net/
[9]. publication/338868364 Evaluation of Modern Tools for an OMSCS Advisor Chatbot
[10]. Hussain, S. A., Folu, O., and Shrish, B. (2019). Chatbot for Family Planning Counseling. https:// doi.org/10.1007/978-3-030-
22219-2 26
[11]. Jayashree, K., Monika, K. A., Preetha, R., and Piraisoodan, S. P. (2020). The Smart Health Care Prediction using Chatbot.
International Journal of Recent Technology and Engineering, 9(2), 75–78. https://doi.org/10.35940/ijrte.A3007.079220

DOI: 10.9790/0050-09013543 www.iosrjournals.org 43 | Page

View publication stats