PROJECT SYNOPSIS

of
CHATBOT

Submitted in partial fulfillment of Bachelors of Technology

In
Computer Science and Engineering

Submitted by
Lakshay Sareen | 2003657
Madhavi | 2003658
Mehak Dora | 2003666

Under the guidance

Of
DR. VINAY CHOPRA AP(CSE)

At

DAV Institute of Engineering and Technology, Jalandhar

(2020-2024)
 TABLE OF CONTENTS
S.NO. TOPIC PAGES REMARKS

1. INTRODUCTION 2-6

2. OBJECTIVE OF PROJECT 7-8

3. SCOPE OF THE PROJECT 9-10

4. TOOLS AND TECHNOLOGIES 11-16

5. FUNCTIONAL REQUIREMENTS 17

6. NON-FUNCTIONAL 18
REQUIREMENTS

7. UML DIAGRAMS 19-21

8. IMPLEMENTATION 22-24

2
 INTRODUCTION

Chatbots, often known as chatter robots, are software agents that mimic human
communication through text or voice messaging. One of the first & most important aims
of Chatbot was to resemble an intelligent human and make it difficult for others to realize
their true nature. Due to their capacity to offer all-day customer care, automate company
procedures, & increase user engagement, chatbots have grown in popularity in recent years.
A chatbot is an artificial intelligence computer that can mimic human -user dialogue using
natural language processing methods. A form of artificial intelligence called machine
learning enables chatbots to learn from previous encounters and get better at responding
less time.
Machine learning-powered chatbots are getting increasingly smart, capable of
comprehending real language and seeing trends in user interactions. They have thus
developed into a useful tool for companies trying to increase consumer interaction &
automate procedures. They may be used to gather information and help in research and
other fields of business.
A software agent known as a chatbot, also referred to as a chatter robot, simulates human
communication by sending text or audio messages. One of the original and primary
objectives of chatbots was to appear clever and deceive others about their true nature. The
use of chatbots has greatly increased as more of them with different designs and capabilities
have been developed.
These conversational agents can deceive users into thinking they are speaking to a human,
but they have very little ability to hone their knowledge base in real time. The chatbot
employs artificial intelligence and deep learning techniques to comprehend user input and
offer a relevant answer.
Doctors, nurses, patients, and their families might benefit from a chatbot. Chatbots may
be able to step in and relieve some of the pressure on medical staff by better organizing 4
patient data, managing medications, assisting in crises or with giving first aid and finding
a remedy for minor medical issues.
Software that can converse with a human using natural language is known as a chatbot or
conversational agent. The modelling of dialogue is one of the key tasks in artificial

3
intelligence and natural language processing. Making a smart chatbot has been the biggest
difficulty in artificial intelligence so far.
Although chatbots are capable of a variety of functions, their main responsibility is to
comprehend human speech and give appropriate responses. In the past, chatbot
architectures were built using straightforward statistical techniques or manually written
templates and rules. About 2015, end-to-end neural networks replaced these models due to
their increased learning capacity. The encoder decoder recurrent model dominates
conversation modelling, particularly at the moment. This architecture was developed for
the neural machine translation market, where it excelled. Up until this point, numerous
additions and modifications have been made that have significantly improved chatbots'
ability to converse.

• Limited Contextual Understanding: Chatbots may struggle to grasp the

complexities of human language as well as the context in which it is used. This
might result in incorrect replies & a misunderstanding of user input.

• Sentiment Analysis: The previous chatbots could not identify the emotions of any
subject about which a human talked. A chatbot should be capable of identifying
whether a human is happy, sad, or angry from the way any speech or text pattern is
presented to it.

• Accuracy: The chatbots should be designed in such a way that their conversation
is like a human to complete any task. But existing chatbots are bad at suddenly
changing subjects and providing unpredictable responses. Sometimes chatbots
respond without any context. Thus, we cannot achieve a satisfactory level of
accuracy.

4
• Predefined or Closed-domain: Previous research has shown that the majority of
chatbots only respond to queries that fall under a closed domain or those that have
been predefined.

Fig1: Different Chatbots Established Different Years

Fig 2: Chatbot Adoption Across Different years

5
Fig 3 Flowchart for chatbot assistant system for college

6
 OBJECTIVE OF THE PROJECT

A chatbot assistant system's aims and objectives might change based on the particular use case and
the requirements of the company. Below is an overview of the many aims and purposes of chatbot
helper systems in the literature.

1. Enhancing client interactions

Enhancing customer service is one of a chatbot assistant system's main objectives. Chatbots
can assist in lowering customer wait times and enhancing response times by offering rapid
and effective customer service. Higher customer satisfaction and loyalty may result from
this. By lowering the cost of customer service, chatbots are predicted by Juniper Research
to help businesses save over $8 billion annually by 2022.

1) Saving resources and time

Moreover, chatbots may assist the business and its clients both save time and resources.
Chatbots can assist in freeing up customer care agents to address more difficult or urgent
issues by automating repetitive chores and enquiries. Cost reductions as well as improved
production and efficiency can result from this. According to Grand View Research, the
demand for automated customer assistance solutions will boost the worldwide chatbot
industry to $1.25 billion by 2025.

2. Enhancing retention and engagement

By giving consumers individualized and pertinent support, a well-designed chatbot
assistant system may help to enhance user engagement and retention. Chatbots may offer
recommendations and guidance that are specifically suited to the user by comprehending
their preferences and habits. Via options for upselling and Cross selling, this may enhance
income while enhancing customer pleasure and loyalty.

3. Improving brand recognition

by offering a consistent and competent level of customer support, chatbots may also assist
to improve a brand's reputation. Chatbots may aid in establishing trust and credibility by

7
responding to users with precise and practical information. Research by Accenture found
that 67% of customers would be open to using a chatbot if it meant they could receive a
prompt answer.

4. Delivering data and insights

by monitoring user interactions and behavior, chatbots may also offer enterprises insightful
analytics. Organizations can discover chances for process and product innovation by
studying this data and learning more about consumer preferences and demands. In
conclusion, the goals and objectives of a chatbot assistant system can vary depending on
the specific use case and the organization's needs. However, the primary goals of chatbots
are typically to improve customer service, save time and resources, increase engagement
and retention, enhance brand reputation, and provide insights and analytics. With the
increasing demand for automated customer support solutions, chatbot assistant systems are
expected to continue to grow in popularity and importance.

1) The main objective of this project is to develop a chatbot software using data sources
available on the web, user generated content and providing knowledge from knowledge
databases.

2) The main purpose of an intelligent is to answer the questions the user may have.

3) It will work on following principles:

i) Natural language processing (NLP)

ii) Artificial Intelligence
iii) Inter process communication
iv) RASA: A inbuilt library for chatbot. Through this virtual assistant, we have automated
various services using a single line command. It eases most of the tasks of the user like
searching the web, retrieving weather forecast details, vocabulary help and medical related
queries. We aim to make this project a complete server assistant and make it smart enough
to act as a replacement for a general server administration.

8
 SCOPE OF THE PROJECT

1. The future of chatbots is still up in the air. What is evident, however, is that these
initiatives are still in their infancy and will very certainly undergo numerous revisions
before becoming widely adopted.

2. Businesses that favor one-on-one or telephone talks are now obsolete as the world of
technology expands. Customers are now demanding quicker forms of communication
via messenger programs.

3. The only way to improve conversion rates in the market is through conversion rate
optimization. Experiences that make the lives of customers and employees more
accessible, safer, more enjoyable, and, of course, more productive!

4. As the world of social media has become increasingly influential, buyers have realized
that they can contact their contacts at any time of day. Email-based messages and other
messaging tools are losing their appeal because buyers cannot always get in touch with
their contacts when they need to.

5. Chatbots can provide buyers with a great experience across all devices, offering deep
insights about the buyer that businesses can use to make personalized offers. Chatbots
help people obtain the information they need and solve their problems.

6. A company's credibility and the quality of the customer experience are two essential
components of any company wishing to achieve a successful business model.

7. Many well-known companies are now using ai chatbot to keep up with the changing
world. The technology sector has seen a massive increase in AI development, which

9
 has led to the creation of chatbots that can help users easily find the information they
need online.

8. Chatbots powered by artificial intelligence effectively create a strong brand image.

They will continue to evolve and play an important role in customer service for
businesses.

9. Chatbots are constantly evolving and becoming more widely used to offer the best
possible customer service. By interacting with customers in real-time, customer service
chatbots can provide a level of service that is otherwise not possible. By the end of
2023, chatbots will have helped businesses gain market share and be a significant
investment in customer service experience.

Fig 4

10
 TOOLS AND TECHNOLOGIES

As shown in the image below, chatbot analyses the request from the user and then identify the
intent of the user, what they need and then generate an appropriate reply and automatically replies
to the user.

Fig 5

There are many tools that help in Chatbot development. Some of the best tools are:

1. Google Dialogflow

Google Dialogflow is a chatbot development tool that helps developers design, develop, and
integrate a user interface into a mobile app, web app, bot or device. It is one of the most popular
tools that support Natural Language Processing in more than 20 languages. Dialogflow supports
nearly all famous messaging platforms including Facebook etc.

11
 Fig 6

2. Microsoft Bot Framework

Microsoft Bot Framework is a comprehensive framework that has the ability to build amazing AI
experiences for your brand. With its Azure cognitive services, you can create a bot for your
business that can speak, listen, and understand your users. Microsoft Bot Framework is packed
with open-source SDK and useful tools that let your team of designers and developers build, test,
and connect bots or sophisticated virtual assistants.

Fig 7

12
3. Amazon Lex

Amazon Lex is an amazing service for building omnichannel Q&A chatbots or conversational
interfaces. It is based on a complete natural language model that allows users to interact with voice,
and text, enabling your brand to build applications that can offer highly engaging user experiences.

Fig 8

4. BotMan

BotMan is a chatbot development framework that allows you to build a chatbot using PHP and
integrate it into your business website and various messaging services, such as Amazon Alexa,
Facebook Messenger, Slack, Telegram, WeChat, and more. One of the best parts of using BotMan
is that it is framework agnostic, allowing us to use the existing codebase, regardless of the
framework type.

13
 Fig 9

5. GupShup

GhupShup is a messaging platform that allows developers to build bots that create highly
interactive conversational experiences for a myriad of messaging channels using a single API. The
API is bundled with highly advanced features, allowing us to perform a lot of tasks. Gupshup’s
single API allows us to develop interactive conversational experiences across voice, SMS, and IP
messaging.

Fig 10

14
6. Wit.ai

Wit.ai is a chatbot development tool that allows businesses to create a natural language experience.
This platform allows you to create bots that enable users to interact with your products with voice
and text. Wit provides developers an open and extensible platform that allows developers to build
apps, bots, and virtual assistants.

Fig 11

7.Botsify

Botsify is one of the famous chatbot development platforms that helps in building a chatbot for
business by simply using its drag-and-drop UI. Botsify helps us build a highly efficient chatbot
that you can use for managing multiple tasks. It makes your chatbot conversational by supporting
more than 190 languages. With Botsify, we can publish the chatbot on your company website,
popular messaging platforms, such as Facebook Messenger, WhatsApp, Slack, and much more.

15
 Fig 12

8. Rasa

Rasa helps to build a text or voice-based assistant for your business. Rasa is a machine learning
framework, allowing developers to build great AI assistants. Rasa is packed with amazing tools
that are necessary for creating high-performing and smart AI assistants that can help solve
customer problems. Developers and enterprises widely use Rasa due to its cutting-edge natural
language understanding, dialogue management, transparent architecture, and integrations.

Fig 13

16
 FUNCTIONAL REQUIREMENTS

Here are some functional requirements for a chatbot assistant system for a college admission
system:

1. User Authentication: The chatbot should allow users to authenticate themselves using their
personal details such as name, email, phone number, and application number.
2. Admission Queries: The chatbot should be able to answer admission related queries such
as application deadlines, admission requirements, fee structure, and admission process.
3. Program Queries: The chatbot should be able to provide details of various courses and
programs offered by the college, such as course duration, eligibility criteria, and course
curriculum.
4. Application Assistance: The chatbot should be able to guide users through the application
process, such as filling out application forms, uploading necessary documents, and
payment of application fees.
5. Schedule Appointments: The chatbot should be able to schedule appointments with the
admission department or faculty members for further queries or assistance.
6. Chat History: The chatbot should maintain a record of the user's chat history to provide
personalized assistance and avoid repetition of information.
7. Integration with Database: The chatbot should be integrated with the college's database to
fetch and update information in real-time.
8. Feedback Collection: The chatbot should collect user feedback to improve its performance
and provide better assistance.
9. Multilingual Support: The chatbot should be able to communicate with users in multiple
languages to cater to a diverse user base.
10. Assistance with Accommodation and Transportation: The chatbot should provide
information about accommodation options and transportation facilities available to
students.

17
 NON-FUNCTIONAL REQUIREMENTS

Non-functional requirements are characteristics or qualities that a chatbot assistant system should
possess to operate effectively and efficiently. Some examples of non-functional requirements for
a chatbot assistant system are:

1. Scalability: The chatbot assistant system should be able to handle a growing number of
users and requests without compromising its performance.
2. Reliability: The system should be dependable and operate consistently without crashing or
experiencing downtime.
3. Security: The chatbot assistant system should be designed with security features to protect
user data, such as encryption and access control.
4. Usability: The system should be easy to use, with a user-friendly interface and clear
instructions for users.
5. Responsiveness: The chatbot assistant system should respond quickly to user requests,
minimizing wait times and improving user experience.
6. Accuracy: The system should provide accurate and relevant responses to user queries, with
high levels of precision and recall.
7. Availability: The chatbot assistant system should be always available to users, with
minimal downtime or maintenance disruptions.
8. Maintainability: The system should be easy to maintain and update, with clear
documentation and modularity.
9. Performance: The system should operate efficiently, with minimal processing time and
resource utilization.
10. Compatibility: The chatbot assistant system should be compatible with different platforms,
devices, and operating systems, ensuring its broad usability.

18
 UML DIAGRAMS

Here are a few UML diagrams that could be useful for a chatbot assistant system:

1. Use Case Diagram:

A use case diagram shows the interactions between the user and the system. In the case of a
chatbot assistant system, the user would be interacting with the chatbot through text-based
conversations. The use case diagram would show the various use cases for the system, such as
asking for help, setting reminders, or providing information.

Fig 14

2. Class Diagram:

A class diagram shows the different classes and their relationships in the system. For a chatbot
assistant system, the classes might include User, Chatbot, Message, and Reminder. The
relationships between these classes would show how they interact with each other, such as a
User sending a Message to the Chatbot, or the Chatbot creating a Reminder for the User.

19
 Fig 15

3. Sequence Diagram:

A sequence diagram shows the interactions between objects in a system over time. In the case
of a chatbot assistant system, the sequence diagram would show the flow of messages between
the User and the Chatbot, as well as any other objects in the system. This could be useful for
debugging or understanding how the system operates.

Fig 16

4. Activity Diagram:

An activity diagram shows the flow of activities or tasks in a system. For a chatbot assistant
system, an activity diagram could be used to show the flow of a conversation between the User

20
and the Chatbot. This could help to identify any potential issues or bottlenecks in the
conversation flow.

Fig 17

Overall, UML diagrams can be a useful tool for designing and understanding a chatbot assistant
system. Depending on the complexity of the system, you may need to create additional
diagrams or modify these diagrams to suit your needs.

21
 IMPLEMENTATION

Here's a possible algorithm for implementing a chatbot assistant system for a college admission
system:

1. Define the scope of the chatbot assistant system: Identify the purpose of the chatbot, the
user types, the key features and functionalities, and the channels through which the
chatbot will be accessible.
2. Collect data and create a knowledge base: Collect the relevant information about the
college admission process, courses, fees, eligibility criteria, admission requirements, etc.
Create a knowledge base that includes this information and is accessible to the chatbot.
3. Choose a chatbot platform or framework: Select a chatbot platform or framework that
suits your needs and preferences. Examples include Dialogflow, Botpress, Rasa, and
Microsoft Bot Framework.
4. Design the conversation flow: Define the conversation flow by creating a chatbot script
that outlines the user's journey from the beginning of the conversation to the end. This
should include different user scenarios and possible responses.
5. Train the chatbot: Train the chatbot by providing it with sample conversations and real-
world data. This will help the chatbot learn how to recognize user intents and respond
appropriately.
6. Integrate the chatbot: Integrate the chatbot with the college admission system's backend
and frontend. This may require using APIs or webhooks to communicate with other
systems and databases.
7. Test and refine the chatbot: Test the chatbot to ensure that it is working correctly and that
it meets the user's needs. Refine the chatbot by analyzing user feedback and improving
the conversation flow, response accuracy, and user experience.
8. Deploy and monitor the chatbot: Deploy the chatbot to the desired channels and monitor
its performance. Use analytics to track user engagement, conversion rates, and user
satisfaction. Make adjustments as necessary to improve the chatbot's performance over
time.

22
Fig 18

23
Fig 19

24