20CS813 CAPSTONE PROJECT

Chatbot For Food Delivery System

A PROJECT REPORT

Submitted by

SOGASANI REETHIKA 111721102144

TRIPURAVARAM SNEHITHA 111721102160
UKKU SUSMITHA 111721102161

in partial fulfillment for the award of the degree

of
BACHELOR OF ENGINEERING
IN
COMPUTER SCIENCE AND ENGINEERING

R.M.K. ENGINEERING COLLEGE

(An Autonomous Institution)
R.S.M. Nagar, Kavaraipettai-601 206

November 2024
 R.M.K. ENGINEERING COLLEGE
(An Autonomous Institution)
R.S.M. Nagar, Kavaraipettai-601 206

BONAFIDE CERTIFICATE

Certified that this project report “Chatbot for Food Delivery System” is the bonafide work
of Sogasani Reethika (111721102144), Tripuravaram Snehitha (111721102160), Ukku
Susmitha (111721102161) who carried out the 20CS813 Capstone Project work under my
supervision.

SIGNATURE SIGNATURE

Dr. T. Sethukarasi, M.E., M.S. Ph.D., Ms.D.Paulraj, M.E., (Ph.D)

Professor and Head Professor
Department of Computer Science and Department of Computer Science and
Engineering Engineering
R.M.K. Engineering College R.M.K. Engineering College
R.S.M. Nagar, Kavaraipettai, R.S.M. Nagar, Kavaraipettai,
Tiruvallur District– 601206. Tiruvallur District–601206.

Submitted for the Project Viva–Voce held on ..................................... at R.M.K. Engineering

College, Kavaraipettai, Tiruvallur District– 601206.

INTERNAL EXAMINER EXTERNAL EXAMINER

I
 ACKNOWLEDGEMENT

We earnestly portray our sincere gratitude and regard to our beloved Chairman Shri. R. S.

Munirathinam, our Vice Chairman, Shri. R. M. Kishore and our Director, Shri. R. Jyothi

Naidu, for the interest and affection shown towards us throughout the course.

We convey our sincere thanks to our Principal, Dr. K. A. Mohamed Junaid, for being the

source of inspiration in this college.

We reveal our sincere thanks to our Professor and Head of the Department, Computer

Science and Engineering, Dr. T. Sethukarasi, for her commendable support and

encouragement for the completion of our project.

We would like to express our sincere gratitude for our Project Guide Ms. P. Baby Shamini,

Assistant Professor for her valuable suggestions towards the successful completion for this

project in a global manner.

We take this opportunity to extend our thanks to all faculty members of Department of

Computer Science and Engineering, parents and friends for all that they meant to us during

the crucial times of the completion of our project.

II
 ABSTRACT

This project aims to create a user-centric food delivery chatbot system that redefines the food ordering

experience. It leverages Dialogflow, Python, and MySQL to facilitate responsive and efficient

interactions.Dialogflow plays a pivotal role in enabling natural language understanding. We explore key

components such as intents, entities, and contexts, enhancing the chatbot's conversational

capabilities.Our Python and FastAPI backend bridges Dialogflow and the database, managing user

authentication, order processing, and data retrieval. This ensures a seamless user experience.Integration

with a MySQL database enables secure data management, including user profiles, menus, and order

information.The project prioritizes user experience by offering personalization features, enabling users

to create accounts, store preferences, and access order history.The chatbot facilitates the entire ordering

process, from order placement to real-time tracking, with automated notifications.Multi-platform

accessibility ensures a consistent user experience across web and mobile applications.Robust error

handling mechanisms guide users through potential issues during the ordering process.After thorough

testing, the chatbot system is ready for deployment in a cloud environment.This project combines

Dialogflow , Python, and MySQL to create an efficient and user-friendly food delivery chatbot,

promising a convenient and satisfying food ordering experience.

 TABLE OF CONTENTS
PAGE
CHAPTER TITLE NO

ABSTRACT 3

1 INTRODUCTION
1.1 Problem Statement 6
1.2 Project Scope and Objectives 6
1.3 Project Specification 7
1.4 Literature Survey 8
1.5 System Requirements 12

2 SYSTEM ANALYSIS

2.1 Existing System 14

2.2 Proposed System 17

3 SYSTEM DESIGN

3.1 System Architecture 21

3.2 User interface 22

3.3 UML Diagrams 23

4 MODULE DESCRIPTION

4.1Modules 27
4.2 Alogrithm 32
4.3 Test Cases 33
44.Test procedure 34

5 RESULT AND DISCUSSION 35

6 CONCLUSION AND FUTURE ENHANCEMENTS 38
 CHAPTER 1

INTRODUCTION

1.1 Problem Statement

"In the food delivery industry, current systems lack efficiency and a user-friendly experience.
Traditional methods often lead to user frustration and business challenges .Integrating a
Responsive chatbot system, powered by dialogflow, Python, and MySQL, can streamline
the food ordering process and enhance user engagement, setting a new industry standard.

1.2 Project Scope and Objective:

1.1.1 Scope of the Project:

The scope of this project is to develop a chatbot for a food delivery system, offering a comprehensive
solution that encompasses user interaction and ordering, real-time order tracking, menu information
and personalized recommendations, user authentication and profiles, efficient database management
with MySQL, Dialogflow integration, a robust FastAPI backend, scalability for future growth,
customization options for businesses, potential payment integration, feedback and support channels,
stringent security measures, data analytics and reporting, comprehensive documentation, and plans
for continuous improvement. This project aims to streamline the food delivery process, enhance user
experiences, and provide businesses with a versatile tool for improving operations and customer
satisfaction.

1.1.2 Objective of the Project:

The primary objective of this project is to create a highly efficient and user-friendly chatbot tailored
for a food delivery system. Our focus is on enhancing the user experience by providing a
conversational interface that allows users to seamlessly place food orders, track their order status in
real-time, and receive personalized menu recommendations. To achieve this, we aim to efficiently
manage user data, order details, and menu items through the implementation of a MySQL database,
ensuring data security and accessibility. Integration with Dialogflow is another core objective,
enabling context-aware conversations with users through the effective use of intents, entities, and
contexts.
1.3 Project Specification:

The project's primary goal is to develop a chatbot-based food delivery system that simplifies and
enhances the process of ordering and receiving food for users. This chatbot will serve as a user-
friendly and efficient interface, connecting users with the food delivery system. Key components
of the system include Dialogflow integration for natural language understanding, a Python-based
FastAPI backend, and a MySQL database for streamlined data management. Users will have the
ability to place food orders, customize their selections, and specify delivery details, all through a
conversational interface. The chatbot will provide real-time order tracking, offering users updates
on their order status and location, thereby improving transparency and user satisfaction. It will also
offer comprehensive menu information, personalized recommendations based on user preferences
and order history, and the ability to create and manage user profiles. Additionally, the option for
secure payment integration within the chatbot is available. To ensure user support, feedback
channels will be in place, addressing concerns and enhancing overall customer satisfaction.
Stringent security measures will safeguard user data, and data analytics and reporting features will
provide insights into user behavior and system performance. The project will offer comprehensive
documentation for users and administrators, along with plans for ongoing maintenance, updates,
and enhancements, ensuring the chatbot remains competitive and adaptable to evolving user and
industry requirements. Ultimately, the chatbot-based food delivery system aims to offer an
efficient, secure, and user-centric solution that optimizes the food delivery process and meets the
changing needs of both users and businesses.
1.4 LITERATURE SURVEY

[1] "Chatbots and the Food Delivery Industry: A Review"

This review offers a comprehensive look at how chatbots have revolutionized the food delivery industry, making
it easier and faster for users to place orders, track delivery status, and get assistance. Chatbots streamline the
ordering process by reducing human involvement in repetitive tasks and enabling users to interact at any time,
leading to greater convenience and satisfaction. Metrics like average response time, accuracy in interpreting
requests, and customer retention rates are discussed to assess a chatbot’s effectiveness. Case studies highlight
successful implementations by leading food delivery companies, such as Domino’s and Swiggy, which have
effectively used chatbots to enhance user engagement. The paper also considers cost savings, as automated bots
reduce the need for a large customer service team, allowing businesses to allocate resources elsewhere.
Additionally, limitations are noted, such as difficulties in handling complex user requests or complaints, which
may require human intervention. Potential future advancements are discussed, including adaptive learning
models that allow chatbots to understand nuanced language better and respond in more conversational, human-
like ways. This could eventually lead to more autonomous, intelligent chatbots that enhance user experience
and streamline operations.

[2] "Natural Language Processing in Food Delivery Chatbots: A Survey"

This survey explores the application of natural language processing (NLP) in food delivery chatbots,
emphasizing how NLP improves a chatbot's ability to understand and respond to user queries with high
accuracy. Key NLP components, such as intent recognition, sentiment analysis, and entity extraction, enable
chatbots to detect user intent, gauge emotional tone, and identify keywords related to food items, order
modifications, and delivery preferences. The review highlights current challenges, such as managing
ambiguous language, handling user-specific dialects, and providing coherent responses under various language
conditions. Advancements in NLP, including deep learning models like transformers, have enhanced chatbots'
conversational abilities, allowing them to engage with users more naturally. Additionally, the paper addresses
the implementation of regional dialect support, which could help chatbots cater to a broader audience. With
further improvements, NLP-powered chatbots could bridge language gaps, reducing language barriers in
diverse markets. The review concludes by discussing the potential for NLP advancements to make chatbots
capable of handling complex conversations, supporting multiple languages, and personalizing interactions
based on historical user data.
[3] "User Experience and Chatbots in Online Food Ordering: A Literature Review"
This literature review focuses on the impact of chatbots on the user experience (UX) in online food ordering,
examining aspects like ease of use, accessibility, and personalization. Chatbots are designed to create a smooth,
efficient ordering process by reducing the number of steps users need to complete a transaction, which
contributes to a more seamless UX. Key UX metrics, such as user satisfaction, task completion rates, and
engagement duration, are discussed to measure chatbot effectiveness. Chatbots use conversational interfaces
that allow users to place orders, make modifications, and inquire about delivery status in a more interactive
manner, enhancing engagement. Furthermore, personalization features, such as remembering past orders and
offering tailored recommendations, make the experience feel more individualized. The study highlights how the
chatbot's tone, personality, and error-handling capabilities also play a role in shaping the user's perception of
the brand. It also considers the potential for frustration if a chatbot cannot effectively address complex requests
or adequately understand user inputs.
[4] "Security and Privacy Concerns in Chatbot-Based Food Delivery Systems: A Review"
This review addresses crucial security and privacy issues associated with chatbot-based food delivery systems.
It discusses how chatbots, which collect and process sensitive user data such as addresses, payment information,
and order history, are vulnerable to privacy risks and data breaches. Security challenges, such as unauthorized
data access, phishing attacks, and identity theft, are identified, highlighting the need for stringent security
protocols. Various mitigation strategies, including encryption, multi-factor authentication, and regular security
audits, are explored. The paper also examines regulatory frameworks, like GDPR and CCPA, which require
businesses to implement data protection measures. Additionally, the importance of transparency in data handling
and user consent mechanisms is emphasized, as it fosters user trust. The review discusses real-world cases where
security lapses in chatbot systems led to data breaches, reinforcing the necessity for continuous updates and
monitoring. Future trends suggest implementing AI-driven security models capable of detecting and responding
to unusual patterns or suspicious activities autonomously. By strengthening security protocols and adopting
privacy-focused designs, chatbot-based food delivery systems can enhance user trust and regulatory compliance.
[5] "Emerging Technologies in Food Delivery: A Comprehensive Review"
This literature review explores the integration of emerging technologies, such as artificial intelligence (AI),
machine learning (ML), and the Internet of Things (IoT), in the food delivery industry.It highlights how these
technologies improve operational efficiency and user experience, with chatbots as a core technology facilitating
user interactions. The paper discusses the use of AI for real-time route optimization, which reduces delivery
times and operational costs. IoT-enabled sensors track food freshness and ensure safe handling during delivery.
Additionally, machine learning algorithms help chatbots predict user preferences, improving the accuracy of
recommendations.
Case studies showcase businesses that have successfully combined these technologies, such as Uber Eats using
ML to estimate delivery times. The review also explores how emerging technologies can be further integrated,
like using augmented reality (AR) to enhance the food ordering experience and using drones for last-mile
delivery. Future trends indicate a shift toward even more autonomous food delivery systems, where AI-driven
chatbots interact seamlessly with IoT devices to provide a more responsive and efficient user experience.

[6] "Personalization and Recommender Systems in Food Delivery Chatbots: A Survey"

This survey investigates the personalization capabilities of food delivery chatbots, focusing on recommender
systems that suggest menu items based on individual user preferences and historical data. It examines various
recommendation techniques, including collaborative filtering, content-based filtering, and hybrid methods,
which allow chatbots to cater to unique user tastes and dietary restrictions. The study highlights how
personalization contributes to user satisfaction, as tailored recommendations make the ordering experience
more relevant and enjoyable.
By analyzing user behavior patterns, chatbots can also provide dynamic recommendations that evolve over
time. However, challenges include balancing personalization with privacy concerns, as excessive data
collection could lead to user discomfort. Techniques like differential privacy are discussed to protect user data
while maintaining recommendation accuracy. The survey concludes with insights into future trends, suggesting
that more advanced machine learning models may enable chatbots to predict orders or provide real-time
customization, leading to a highly personalized food ordering experience.

[8] "Data Analytics in Food Delivery: A Review of Applications and Trends"

This review discusses the role of data analytics in enhancing food delivery operations, with chatbots playing a
key role in collecting and leveraging customer data for actionable insights. By gathering data on user
preferences, order frequency, and popular items, chatbots help businesses optimize menus, forecast demand,
and manage resources more effectively. Predictive analytics can anticipate peak order times, enabling better
staffing and delivery preparation, while prescriptive analytics suggest actionable steps to improve service. The
review highlights data-driven success stories in major food delivery platforms that have increased operational
efficiency and customer satisfaction. Additionally, ethical considerations surrounding data collection and usage
are discussed, as there is a need to balance analytics-driven insights with user privacy. As data analytics
technology advances, chatbots could offer even more granular insights, allowing for hyper-personalized
marketing campaigns and real-time adaptive strategies to meet evolving customer demands.
[9] "Conversational Restaurant Recommendation with Spoken Language Understanding.":

Natural Language Processing (NLP) offers a compelling solution to these shortcomings. Hakkani-Tur,
HakkaniTur, & Heck (2016) delve into the development of an NLP-powered chatbot for restaurant
recommendations in their research titled "Conversational Restaurant Recommendation with Spoken Language
Understanding." [1].Their findings highlight a strong user preference for conversational interfaces, particularly
for tasks like menu exploration and filtering based on dietary restrictions. This underscores the potential of NLP
to streamline menu navigation and cater to specific user needs.

[10] "Factors Influencing User Satisfaction with Online Food Ordering Platforms: A Meta-Analysis."

Understanding user experience (UX) considerations for online food ordering systems is paramount for designing
effective NLP-powered interfaces. Luo, Li, & Sun (2021) investigate the factors influencing user satisfaction in
online food ordering platforms in their meta-analysis titled "Factors Influencing User Satisfaction with Online
Food Ordering Platforms: A Meta-Analysis." Their research identifies ease of navigation, personalization options,
and order confirmation features as key elements contributing to a positive UX. This suggests that the NLP
interface should be intuitive and easy to navigate, allowing users to find desired items and complete orders
effortlessly. Personalization options, enabled by NLP, can further enhance user satisfaction by catering to
individual preferences.

[11] "The Effect of Website Interface Design on User Experience in Online Food Ordering Systems."

Building upon the importance of interface design, Sun, Wang, & Wang (2019) explore the impact of interface
design on user experience in online food ordering systems within their research titled Their findings reveal that
users favor interfaces that are visually appealing, intuitive to navigate, and provide clear information about menu
options and order status.
1.5 System Requirements
1.5.1 Hardware Requirements
Operating System: Windows, macOS, or Linux
RAM: Minimum 8 GB (16 GB recommended for optimal performance)
Hard Disk or SSD: At least 256 GB SSD (512 GB or higher recommended for faster data access)
Processor: Intel Core i5 or higher / AMD Ryzen 5 or higher (Quad-core processor or above
recommended)
1.5.2 Software Requirements
Front End: HTML, CSS, JavaScript
Framework: React.js or Angular for front-end development
Monitor: Minimum resolution of 1920 x 1080 (Full HD) recommended
Server Side Script: Node.js with Express (for backend API development)
Scripts: Python scripts for data processing, JavaScript for client-side interactivity

1.5.3 Feasibility Study

The feasibility of the project is analysed in this phase and business proposal is put forth with
a very general plan for the project and some cost estimates.During system analysis the feasibility
study of the proposed system is to be carried out. This is to ensure that the proposed system is
not a burden to the company.For feasibility analysis, some understanding of the major requirements
for the system is essential.
Three key considerations involved in the feasibility analysis are
ECONOMICAL FEASIBILITY
TECHNICAL FEASIBILITY
OPERATIONAL FEASIBILITY

ECONOMICAL FEASIBILITY:
The cost of development is manageable given the open-source nature of the core technologies used, such
as Python, React.js, and MySQL. These technologies are widely adopted and do not require expensive
licensing, reducing the overall project cost. Additionally, cloud deployment ensures scalability at an
affordable cost, with the ability to scale up as user demand grows. The choice of commonly used
technologies also simplifies the hiring process, as there is a large pool of developers with expertise in
these areas, making labor costs reasonable. With the availability of cloud platforms offering pay-as-you-
go pricing models, the system can be efficiently scaled without significant upfront investment.
 TECHNICAL FEASIBILITY:

The hardware and software requirements outlined are well within the capabilities of current technology.
The system is designed to run on widely available operating systems like Windows, macOS, and Linux, and
the chosen components (Dialogflow, Python, MySQL) are industry-standard technologies with extensive
community support. The integration of Dialogflow for natural language processing ensures efficient and
accurate user interactions, while the use of Python and FastAPI provides a robust backend architecture for
managing user data, orders, and authentication. MySQL is a reliable database management system for
handling secure and scalable data storage, including user profiles, menus, and order histories. Given that the
required hardware (8 GB RAM, SSD storage, and modern processors) is standard in most development
environments, the technical setup is entirely feasible for development and deployment.

OPERATIONAL FEASIBILITY:

The proposed chatbot system enhances the user experience by focusing on ease of use and personalization.
The use of React.js or Angular for the front-end ensures a responsive, user-friendly interface, while the
backend setup with Node.js and FastAPI guarantees smooth data processing. By supporting multi-platform
access (web and mobile), the system ensures consistency in the user experience across devices. The chatbot’s
ability to handle everything from order placement to real-time tracking, along with robust error-handling,
improves both functionality and user satisfaction. Moreover, the integration of personalization features, like
storing preferences and order histories, aligns with modern user expectations for convenience and efficiency.
 CHAPTER 2
SYSTEM ANALYSIS
2.1 Existing System

The current food delivery systems primarily rely on traditional mobile apps or web-based platforms that allow
users to manually browse menus, place orders, and track deliveries. These systems are generally designed with a
standard user interface that requires customers to navigate through several screens and select their preferred food
items, quantities, and payment options. Although functional, these platforms often lack the intuitive,
conversational interface that users increasingly expect in the digital age. Users are required to manually input
information such as delivery addresses, contact details, and payment methods for each order, leading to potential
friction in the user experience.Additionally, many existing food delivery apps are not optimized for handling
personalized interactions. They rely on static user profiles and do not offer dynamic, context-aware responses
during the ordering process. As a result, customers are often limited to repetitive tasks, such as manually entering
information for every order or navigating through menus without receiving helpful recommendations based on
past preferences. This can lead to inefficiencies and frustration, particularly for users who frequently order food
and expect a faster, more personalized experience.

Current systems also often fail to provide real-time assistance to users, especially during critical stages of the
ordering process. Issues such as incorrect order details, payment errors, or delivery problems require customers
to interact with customer support teams, which can be time-consuming and frustrating. Furthermore, the
verification and validation of orders can involve multiple steps that require human intervention, reducing the
overall efficiency of the process. This lack of automation in key areas, including order tracking and notification,
makes the current systems less efficient and prone to delays.Another limitation of the existing systems is their
lack of integration across multiple platforms. Many food delivery services are limited to a specific device or
platform, such as a mobile app or desktop website, which restricts access for users who may prefer to place orders
on different devices. This results in a fragmented user experience that is inconsistent across different platforms,
ultimately leading to user dissatisfaction. Users may encounter performance or display issues when switching
between devices, further diminishing the overall experience.

In terms of data management, traditional food delivery platforms typically store user information, orders, and
payment details in centralized databases. While these systems work adequately, they may not offer the same level
of security and transparency that modern, decentralized technologies like blockchain and cloud computing can
provide. Data breaches and privacy concerns are common in centralized systems, and users may have limited
control over their personal information.
Furthermore, these platforms often lack robust data-driven insights into customer preferences or behavior,
limiting their ability to offer personalized recommendations or promotions.Finally, existing systems may not offer
seamless error handling or support during the ordering process. Users who encounter issues, such as unavailable
menu items, payment failures, or delivery delays, often face lengthy customer support interactions to resolve these
problems. While some platforms may offer chatbot assistance, these are often rule-based systems that fail to
provide meaningful or contextually appropriate responses. As a result, many users continue to rely on traditional,
less efficient methods of order verification and issue resolution.

Overall, while the existing food delivery systems serve their purpose, they suffer from a range of limitations,
including a lack of personalization, slow response times, and fragmented user experiences across platforms. The
need for more intuitive, conversational, and efficient food ordering processes has led to the development of a new
generation of systems, such as the proposed food delivery chatbot system, which seeks to address these
shortcomings and provide users with a more convenient, personalized, and seamless ordering experience.

2.1.1 Disadvantages of Existing System

1. Manual Input Requirement:

Users are required to manually enter delivery details, payment methods, and food selections for each order,
leading to a time-consuming process.
2. Lack of Personalization:
Existing systems often do not offer personalized recommendations or remember user preferences, which
results in a less efficient experience for frequent customers.
3. Navigation Complexity:
The multi-step process of browsing menus, selecting items, and filling out order details can be overwhelming,
especially for new users.
4. Limited Interaction:
Users interact primarily through static forms and interfaces, with minimal conversational or context-driven
feedback, reducing engagement.
5. Inefficient Customer Support:
In case of order issues, users must manually contact support teams, which can be slow and frustrating.
6. Fragmented User Experience:
Many systems are limited to specific platforms (e.g., mobile apps or desktop), which causes a lack of
consistency across different devices.
7. Security Concerns:
Centralized systems are prone to data breaches, exposing user information to security risks.
8. Lack of Real-Time Assistance:
Users may struggle with problems like incorrect orders or payment failures without immediate assistance,
leading to delays and frustration.
9. Slow Order Processing:
Multiple manual steps in verifying and processing orders can slow down the overall experience.
10. No Real-Time Tracking:
Most existing systems do not offer real-time order tracking, causing anxiety and uncertainty for users waiting
for their food.
11. Inconsistent Performance:
Users may experience performance issues, such as slow loading times or crashes, when switching between
devices.
12. Limited Data Analytics:
Traditional platforms lack advanced data-driven insights into user behavior, limiting their ability to offer
targeted promotions or personalized recommendations.
13. Customer Dependency on Support:
Users have to rely on human support for order verification, which increases response times and operational
costs.
14. Payment Failures:
Payment errors can be common, and users may not receive instant feedback or support to resolve them
quickly.
15. Order Accuracy Issues:
Manual input and human interventions can lead to errors in orders, including wrong food items or incorrect
addresses.
16. Lack of Automation:
Existing systems require significant manual effort for tasks like updating user profiles or managing inventory,
reducing efficiency.
17. Difficulty in Scaling:
Traditional systems may struggle to scale efficiently during high-demand periods, causing slowdowns or
service interruptions.
18. Lack of Transparency:
Users may not have full visibility over the status of their order, resulting in uncertainty and dissatisfaction.
19. Over-reliance on Reviews:
Users often rely on reviews to judge restaurant quality, but without real-time feedback from a chatbot, the
reviews can be outdated or inaccurate.
2.2 Proposed System

The proposed system aims to revolutionize the food ordering experience by creating a user-centric food
delivery chatbot powered by Dialogflow, Python, and MySQL. By incorporating artificial intelligence
(AI) and natural language processing (NLP), the chatbot will facilitate intuitive, conversational
interactions, enabling users to place orders with ease. Dialogflow, with its advanced capabilities in
understanding user intent and context, forms the core of the system, allowing for efficient and natural
conversations. Users will no longer need to navigate through complex menus or forms; instead, they can
interact with the chatbot as they would with a human assistant, making the ordering process more seamless
and efficient.One of the key features of the proposed system is personalization. The chatbot will allow
users to create accounts, store their preferences, and access their order history. This personalized
experience will not only make the ordering process quicker for repeat customers but will also enable the
chatbot to suggest food items based on the user’s past behavior and preferences. For example, the chatbot
could offer meal recommendations based on dietary preferences, frequently ordered items, or popular
local dishes. This level of personalization will improve user satisfaction and drive customer loyalty by
offering a more tailored experience.

The backend of the system, developed using Python and FastAPI, will act as the bridge between
Dialogflow and the MySQL database. This setup will ensure that the system can securely manage critical
data such as user profiles, order histories, menus, and real-time order tracking. FastAPI, known for its
high-performance asynchronous capabilities, will ensure the system remains responsive and capable of
handling multiple concurrent requests from users, especially during peak hours when the demand is high.
Integration with MySQL will allow for secure, scalable storage of user data and order information,
maintaining both the integrity and privacy of sensitive data.
In terms of user interaction, the chatbot will manage the entire ordering process, from placing orders to
tracking deliveries in real-time. Once the user places an order, the chatbot will send automated
notifications at key stages of the process, such as confirming the order, updating the user about the status
of the delivery, or alerting them about potential delays. This proactive communication will keep users
informed and engaged, reducing the need for manual updates from customer support teams and enhancing
overall user experience.Another significant feature of the proposed system is multi-platform accessibility.
The chatbot will be accessible across web and mobile applications, ensuring that users have a consistent
experience regardless of the device they use. This multi-channel approach ensures that users can place
orders from their smartphones while on the go or from their computers while at home or work, providing
flexibility and convenience. The design of the chatbot interface will be responsive, ensuring that it adapts
to different screen sizes and resolutions, maintaining a smooth and engaging user experience on all
devices.The proposed system will also focus on robust error handling. During the ordering process, users
may encounter issues such as unavailable menu items, payment errors, or delivery delays. The chatbot
will be equipped with advanced error-handling mechanisms to guide users through these situations,
offering clear instructions and support to resolve issues quickly. This will reduce user frustration and
ensure that users feel confident and supported throughout their entire interaction with the system.

Once thoroughly tested, the chatbot system will be deployed in a cloud environment to ensure scalability
and availability. Cloud hosting will provide the flexibility to scale the system based on user demand,
ensuring it can handle peak traffic during high-order volumes without performance degradation. The
cloud environment will also ensure that the system is always accessible, offering reliable uptime and
minimizing downtime due to hardware or infrastructure failures.

In conclusion, the proposed food delivery chatbot system will redefine the food ordering experience by
integrating conversational AI with backend technologies to deliver a seamless, personalized, and efficient
service. By providing real-time tracking, automated notifications, multi-platform support, and robust error
handling, the system promises to enhance user satisfaction and reduce the friction often associated with
traditional food delivery methods. This user-centric approach, combined with a scalable and secure
infrastructure, will make the chatbot a valuable asset to both customers and food delivery services.
2.2.1 Advantages of proposed System

1. User-Centric Experience:
The chatbot offers an intuitive, conversational interface, making food ordering more accessible and
engaging for users.
2. Personalization:
Users can create accounts, store preferences, and access order history, leading to a tailored food
ordering experience based on past behavior.
 3. Real-Time Order Tracking:
Users can track their orders in real-time, reducing uncertainty and improving satisfaction during the
delivery process.
4. Automated Notifications:
The chatbot sends automated updates at each stage of the order process, keeping users informed and
minimizing the need for customer support interventions.
5. Time-Efficiency:
The conversational flow streamlines the ordering process, reducing the time spent navigating complex
menus and forms.
6. Multi-Platform Accessibility:
The chatbot can be accessed on both mobile and web platforms, providing users with flexibility to
place orders from various devices.
7. Improved Order Accuracy:
With natural language understanding, the chatbot ensures that user input is correctly interpreted,
reducing errors compared to traditional manual input.
8. Reduced Customer Support Load:
Automated issue handling and notifications minimize the need for customer support, allowing staff
to focus on more complex queries.
9. Scalable Infrastructure:
Cloud deployment allows the system to scale easily according to demand, ensuring smooth operation
during peak times.
10. Security and Data Privacy:
Integration with MySQL ensures secure data storage and management, protecting user profiles and
payment information.
11. Fast Performance:
FastAPI’s asynchronous capabilities ensure that the system handles multiple requests efficiently,
providing quick responses even under heavy load.
12. Seamless Integration:
The system integrates Dialogflow with the backend, ensuring smooth communication between the
chatbot and the database for accurate data retrieval and order processing.
13. Enhanced User Engagement:
Personalization and real-time feedback foster continuous user engagement, making the experience
more interactive and enjoyable.
14. Error Handling:
The chatbot features advanced error handling, guiding users through issues like unavailable items or
payment failures, reducing frustration.
15. Cost-Effective:
The automation of order processing and customer support reduces operational costs compared to
traditional food delivery methods.
16. Consistency Across Platforms:
Users experience a consistent and unified interface across different devices, ensuring no loss of
functionality or design elements when switching platforms.
17. Advanced Analytics:
The system can collect and analyze data on user preferences and order history, which can be used to
improve future interactions and offer personalized promotions.
18. Convenient User Experience:
Users can place orders quickly, receive timely notifications, and enjoy a smoother process from start
to finish, improving overall satisfaction.
19. Accessibility for All Users:
The chatbot interface, with its natural language capabilities, makes food ordering accessible to users
who may struggle with traditional forms or apps.
20. Competitive Edge for Businesses:
The implementation of AI and a streamlined food delivery process provides a competitive advantage
by offering a modern, efficient service that enhances customer satisfaction.

These advantages highlight how the proposed system significantly enhances the user experience, reduces
operational complexity, and promotes scalability, ultimately offering a more efficient and satisfying food
ordering process.
 CHAPTER 3

SYSTEM REQUIREMENTS AND ANALYSIS

3.1 SYSTEM ARCHITECTURE:

The architecture diagram defines the fundamental structures of a software system and the discipline
of creating such structures and the structured solution of the proposed idea.

Figure 1: Architecture Diagram

 The architecture of our chatbot-based food delivery system is an intricate blend of innovation and
functionality, designed to provide a seamless and efficient experience for users.
 At its core, the chatbot interface acts as the primary point of interaction, offering a user-friendly
platform to place orders, inquire about menu items, and manage food deliveries.
 Beneath the chatbot lies the powerful Dialogflow integration, a natural language understanding
and processing platform that processes user queries and commands.
 Dialogflow interprets intents, entities, and contexts, facilitating meaningful and context-aware
conversations with users.
 The Python-based Fast API backend serves as the backbone of the system, connecting Dialogflow with
the rest of the infrastructure. Data management is key, ensuring data integrity, security, and
accessibility. Real-time order tracking, one of the system's standout features, enhances user experience
and transparency. It provides users with up-to-the-minute information on their order's status and
location. The system also accommodates secure payment integration, offering a comprehensive end-
to-end solution.
In essence, our system architecture is a symphony of technology, data management, and user- centric
design, creating an ecosystem where the food delivery process is streamlined and enriched, while
continuously evolving to meet changing user and industry demands.

3.2 USER INTERFACE:

Our user interface is designed for simplicity and efficiency. It's a user-friendly platform that enables
easy food ordering and customization through a convers ational interface. Real-time order t racking
adds transparency and satisfaction. Users can create profiles and save preferences for a personalized
experience. Optional payment integration streamlines the process. It's all about making food ordering
straightforward and enjoyable.
 3.3 UML Diagrams

3.3.1 USE CASE DIAGRAM

This use case diagram describes the interaction between the user and the website.

Figure 2: Use case Diagram

A use case diagram for a chatbot in a food delivery system represents the interactions between the
system’s key stakeholders and the chatbot’s functionalities. This chatbot serves as a virtual
assistant to handle customer inquiries, manage orders, track delivery status, and facilitate overall
customer engagement.The main actors in this use case diagram would include customers, delivery
personnel, and the admin team.
These actors interact with the chatbot to perform various tasks, streamlining the order and delivery
process and enhancing user experience.The *customer* is a primary actor in the system. In the use
case, customers can engage with the chatbot to *browse available menu items*, select items, place
an order, make payments, and check delivery updates. The chatbot is designed to assist customers
through a series of interactive messages that guide them from menu selection to finalizing their
orders. Additional features allow customers to modify orders, request refunds, and rate their
experience, all of which the chatbot can facilitate, creating a smooth and user-friendly
experience.For *delivery personnel*, the chatbot provides real-time order details, delivery
addresses, and customer contact information to enable efficient delivery service. Delivery
personnel can also update the order status through the chatbot (e.g., marking the order as “on the
way” or “delivered”).
This communication keeps both the customers and admin team informed of the delivery status
without requiring direct human intervention, improving transparency and tracking within the
delivery process.The *admin team, who manages the backend of the food delivery service, uses
the chatbot for **order management and tracking analytics*. Through the chatbot’s functionalities,
admins can monitor order volumes, customer interactions, and delivery statuses. In case of any
discrepancies or issues reported by customers or delivery personnel, the chatbot can escalate the
issue to an admin, who can then take corrective action. This use case also ensures that the chatbot
aids in collecting valuable customer feedback for continuous improvement. The use case
diagram’s core *functionalities* reflect various scenarios the chatbot must handle efficiently. Core
use cases include “Browse Menu,” “Place Order,” “Track Order,” “Manage Payments,” and
“Provide Feedback.”
Each of these use cases captures specific interactions that customers or delivery personnel may
initiate with the chatbot. By mapping out these functions, the diagram helps in visualizing how
each task is organized within the chatbot’s system, ensuring that all critical interactions are
accounted for.The *scope and limitations* of the chatbot’s functionality also need to be
represented in the use case. For instance, the chatbot should be able to handle common queries and
standard order processes but may escalate more complex issues (e.g., technical payment errors or
delivery location issues) to human support.
 3.3.2 CLASS DIAGRAM

Figure 3: Class Diagram

In a class diagram for a food delivery chatbot system, the core classes are designed to handle different
aspects of user interaction, order management, menu display, payment, and feedback collection. The
*Chatbot* class acts as the central controller, responsible for interpreting user input, managing the
conversation flow, and generating responses. It processes user requests by determining intent and passing
requests to relevant classes, such as handling orders, payments, and feedback. The chatbot interacts
extensively with other classes to deliver a smooth user experience.

The *User* class stores individual user data, including user ID, name, contact information, and location. It
manages communication with the chatbot, enabling users to send messages, view the menu, place orders,
and access past order history.
The *Menu* class provides access to the available food items and consists of a list of *Menu Item* objects,
each containing details like item name, price, description, and availability. Users can browse and search for
specific items, with the chatbot fetching this information from the Menu class to present it during the
conversation. The *Order* class plays a crucial role in the order management process, tracking the items
selected by the user, calculating the total cost, and updating the order’s status (e.g., pending, completed, or
canceled). Each order links to both the User and the Payment classes, ensuring that user details are connected
with the transaction, and the payment status is accurately reflected. The *Payment* class then handles the
financial aspects, including managing payment methods (such as credit card, digital wallet, or cash on
delivery), processing payments, and validating payment details. This ensures that each transaction is
securely completed before confirming the order.Additionally, the *Feedback* class allows users to provide
feedback after their order is delivered. This class is linked to both the User and Order classes and includes
attributes for storing ratings and comments. Feedback information can be retrieved later to assess user
satisfaction and improve service quality.The *Location* class supports the logistics of the delivery by
managing delivery addresses, postal codes, and coordinates. This class calculates estimated delivery times
based on the user’s location and the restaurant's address, enabling the chatbot to inform the user of expected
delivery timelines. In a more advanced setup, Location could also interact with third-party mapping services
to provide real-time updates on delivery progress.Together, these classes facilitate a comprehensive food
delivery chatbot system. The classes interact closely, allowing the chatbot to assist users with browsing the
menu, placing and tracking orders, processing payments, and gathering feedback, thus creating a seamless
and user-friendly experience. This design also offers flexibility to scale, allowing additional features, such
as delivery tracking or promotional offers, to be added in the future.
 CHAPTER 4

4.1 MODULE DESCRIPTION:

4.1.1 User Interface Module

The User Interface Module is the interactive front end that connects users to the chatbot system, designed
with ease of use and responsiveness in mind. It supports multiple communication channels, such as mobile
apps, web applications, and popular messaging platforms like WhatsApp, Facebook Messenger, and
Instagram. Users can explore the menu, place and customize orders, check order status, and make
payments directly through this module.
The UI also provides a guided flow for new users, assisting them with initial commands and teaching
them how to interact with the chatbot effectively.This module also incorporates real-time notifications
and error handling to manage unexpected issues. For example, if a user tries to order an unavailable item,
it provides suggestions for similar items. If there are network or processing delays, it displays messages
to keep users informed, reducing frustration. Additionally, the UI can integrate with accessibility features,
such as voice-based input or screen reader compatibility, ensuring inclusivity for all users. By delivering
a responsive, intuitive, and user-centered experience, the User Interface Module plays a crucial role in
user retention and satisfaction.

4.1.2 Dialogflow Integration Module

The Dialogflow Integration Module leverages Dialogflow’s NLU capabilities to interpret and process
user queries accurately. This module configures intents for specific actions (e.g., placing an order,
modifying an order, checking delivery time) and employs entities to capture relevant details like item
names, quantities, and user locations. For example, it can handle complex requests where a user specifies
dietary preferences, like "Show me vegan options under $10."

Dialogflow’s context management also enables this module to maintain ongoing conversations, making
it possible to manage follow-up questions seamlessly. If a user asks, “How much is the burger?” followed
by “Add one to my order,” the chatbot understands the context and completes the action without needing
further clarification. This module also integrates with Dialogflow’s language support for multilingual
experiences, enabling users to interact with the chatbot in their preferred language. Advanced error-
handling features are configured to guide users in case of misunderstood queries, such as suggesting
rephrased questions. The Dialogflow Integration Module essentially powers the chatbot’s “conversational
intelligence,” making it adaptable, context-aware, and user-friendly.

4.1.3 FastAPI Backend Module

The FastAPI Backend Module is the system’s powerhouse, coordinating various processes and managing
data flow across modules. It handles core backend logic such as validating requests, managing sessions,
and coordinating API calls between Dialogflow and the database. This module executes complex
workflows, like processing an order, which involves checking item availability, calculating prices
(including taxes and delivery fees), and updating order status in real-time.
It also provides secure authentication mechanisms, managing user accounts through login, registration,
and token-based authentication, ensuring only authorized users can access their order history and sensitive
information. The backend supports customizations, allowing users to adjust their orders according to
preferences or allergies. Additionally, FastAPI’s asynchronous capabilities allow the system to handle a
high volume of concurrent requests, ensuring stability during peak times. Advanced error handling is
implemented, logging errors and sending feedback to Dialogflow, so the chatbot can communicate helpful
messages to the user. This module is also designed for scalability, making it easy to add new features,
such as promotional offers or loyalty programs, with minimal code changes.
The FastAPI Backend Module is the core engine of the chatbot system, seamlessly managing and
coordinating all processes to provide a smooth and responsive user experience. As the main orchestrator,
it handles backend logic, including request validation, session management, and integration with external
APIs like Dialogflow and the MySQL database. This module manages complex workflows, such as
processing orders, by performing a series of steps: verifying item availability, calculating costs (including
taxes and delivery fees), and updating the order status in real time. It also logs transactions, creating a
record of each interaction for further analysis or troubleshooting.

Security is a top priority for the FastAPI module. The backend includes secure authentication and
authorization processes using token-based authentication, ensuring only authorized users can view their
order history, preferences, and other sensitive information. User accounts are managed through secure
login and registration processes, while encryption and secure handling practices are enforced for sensitive
data. To personalize the user experience, the backend allows for various customizations, enabling users
to adjust their orders based on preferences, dietary restrictions, or allergies. FastAPI’s asynchronous
processing capabilities make it highly efficient and capable of handling a large volume of concurrent
requests without compromising performance.
During peak hours, this capability ensures stability, allowing users to place orders or interact with the
chatbot smoothly and promptly.The module also implements comprehensive error handling and logging,
capturing detailed error information and generating logs for each issue. This helps in diagnosing and
resolving errors efficiently, while feedback is sent to Dialogflow to relay helpful messages to the user.
This approach not only improves user satisfaction but also supports continuous system monitoring and
improvement.To support future growth, the FastAPI Backend Module is designed with scalability in
mind. Its modular structure enables developers to introduce new features, such as promotional offers,
loyalty programs, or user feedback mechanisms, with minimal code changes. This adaptability also allows
for easy integration with additional APIs or services that may enhance functionality, such as integrating
new payment gateways or real-time analytics tools.

Another critical feature is session management. Each user session is tracked to maintain continuity,
allowing users to return to their orders or queries without losing information. This session management
also aids in providing more context-aware interactions, enabling the chatbot to understand user
preferences and offer relevant suggestions.

Moreover, FastAPI’s support for OpenAPI documentation means that the module is self-documenting,
making it easier for developers to understand and work with the API. This documentation is auto-
generated and accessible, simplifying maintenance and onboarding for new developers.The module also
prioritizes performance optimization. Techniques like data caching, query optimization, and background
task processing are employed to minimize latency and improve response times. For instance, frequently
accessed data, such as common menu items or promotions, can be cached to reduce database load,
improving speed.Ultimately, the FastAPI Backend Module’s architecture and capabilities enable the
chatbot system to operate reliably, securely, and efficiently, providing a seamless experience for users.
By integrating request validation, session management, error handling, scalability, and security, it lays a
strong foundation for delivering responsive, data-driven, and user-friendly interactions.

4.1.4 MySQL Database Module

The MySQL Database Module is responsible for securely storing, organizing, and managing all relevant
data for the chatbot system. It stores user profiles, including preferences, order histories, and payment
details, allowing for a personalized experience.
This module uses a relational database structure, which helps in organizing data efficiently, especially
when handling relationships between users, orders, and menu items. Tables are structured to optimize
data retrieval, with indexing and caching mechanisms ensuring quick access to frequently requested
information like popular items or regular orders.

Data integrity is maintained through strict validation protocols to prevent inconsistent data entries, while
access control mechanisms ensure data security, safeguarding user privacy and payment information. This
module also includes backup and recovery features to protect against data loss and support data analytics,
allowing the system to analyze trends, such as popular menu items or peak order times. By providing
reliable, secure, and well-organized data storage, the MySQL Database Module underpins the chatbot’s
ability to offer timely, accurate, and data-driven responses.
The MySQL Database Module is a critical component for the chatbot system, responsible for securely
storing, organizing, and managing data essential to user interactions. It holds user profiles, preferences,
order histories, and payment details, supporting a personalized user experience by leveraging data
insights. Built on a relational database structure, it efficiently manages relationships between users,
orders, and menu items, making data retrieval faster and more accurate.Each table in the database is
designed with efficient schema structures. The `users` table contains information like user ID, name,
contact info, and preferences.

The `orders` table logs past orders, including order dates, item details, quantities, and costs, while the
`menu` table lists all available menu items with descriptions, prices, and categories. Foreign keys link
these tables, enforcing relational integrity and helping prevent inconsistencies.To further enhance
performance, indexing and caching mechanisms are employed, enabling fast access to frequently
requested data, such as popular menu items or repeat orders. Primary and secondary indexes optimize
query speed, reducing response times for data retrieval operations. Caching ensures frequently accessed
data is readily available without querying the main database repeatedly, improving efficiency.

Strict validation protocols ensure data integrity by preventing invalid entries and maintaining data
consistency across tables. Access control mechanisms protect user privacy, with role-based permissions
allowing only authorized personnel to access sensitive data, such as payment information.
 This helps safeguard user data from unauthorized access or manipulation.The database also includes
robust backup and recovery systems to prevent data loss in case of system failures or crashes. Scheduled
backups and real-time replication mechanisms ensure that data remains protected, enabling rapid recovery
and continuity of service.

Analytics capabilities are integrated to analyze trends, such as frequently ordered items, peak ordering
times, and popular menu categories. These insights allow the chatbot to offer data-driven
recommendations, predict user preferences, and optimize the menu and inventory accordingly. Such
analytics also support business decisions by identifying areas for improvement or high demand.Finally,
the MySQL Database Module’s secure, reliable, and structured data storage underpins the chatbot's ability
to provide accurate, personalized, and timely responses, creating a seamless and engaging user
experience. By using well-defined schema design, indexing, caching, validation protocols, access
controls, and analytics, this module establishes a robust data management system tailored to the chatbot's
needs.

4.1.5 Real-Time Order Tracking and Payment Module

The Real-Time Order Tracking and Payment Module enhances user convenience by providing up-to-date
order information and secure payment processing.For order tracking, it integrates with GPS-enabled
delivery systems, providing users with real-time updates on the location and estimated arrival time of
their orders. Through push notifications, users are informed about key events, such as order confirmation,
preparation, dispatch, and delivery completion. This module can also display an interactive map that
shows the driver’s location, improving user transparency and trust.

For payment, this module supports multiple methods, such as credit cards, debit cards, and digital wallets,
with additional payment options like cash on delivery. Each transaction is encrypted and processed
securely to ensure user privacy. Payment statuses are updated in real time, with notifications confirming
successful payments or alerting users of payment failures. In case of payment failures, the module
provides retry options and suggests alternative payment methods.
4.2 Algorithms
4.2.1 Natural Language Processing:
Natural language processing(NLP) is confined to the stream of computer science and artificial
intelligence. Natural language processing mainly deals with the interaction between computers and
human language which mainly focuses on how computers process and analyze a large amount of data.
The NLP techniques help computers to understand contexts in the document. Natural Language
Processing (NLP) plays a significant role in document ranking, especially in information retrieval
systems such as search engines. Document ranking refers to the process of determining the relevance of
documents to a user's query and presenting them in a ranked order.
4.3 TEST CASES:
Scope - The scope of the testing is to make the user to go through all the web pages easily.
Test Deliverables – An enjoyable and worthful destination will be provided as the outcome.

TEST CSE ID TEST NAME EXPECTED ACTUA

STATUS
RESULTS RESULTS
1. User Order Order is As expected Pass
Placement successfully
placed, and a
confirmation is
received.
2. Real-Time Order Real-time As expected Pass
Tracking updates on
the order's
status and
location are
provided.
3. User Profile User profile is As expected Pass
Creation and successfully
Authentication created, and
login is
successful.

4. Dialogflow Chatbot As expected Pass

Interaction and effectively
Intent interprets user
Recognition queries,
demonstrating
accurate intent
recognition.

5. Secure Payment is As expected Pass

Payment securely
Integration processed, and
an order
confirmation is
received.
4.4 TEST PROCEDURE:

The test plan is shown below

Name of the test Things to be tested

Data Integrity and Security Protecting user data and preventing

unauthorized access.

Data Analytics and Reporting Gathering insights into user behavior and system
performance.

Personalized Menu Recommendations Recommending menu items based on user

preferences and history.

Real-Time Order Tracking Requesting and tracking order status and

location in real-time.
 CHAPTER 5

RESULT AND DISCUSSION

 CHAPTER 6

CONCLUSION AND FUTURE ENHANCEMENTS

In summary, the chatbot-based food delivery system has revolutionized the way users order and
receive their favorite meals. It offers a user-friendly experience with real-time tracking, menu
personalization, and robust security. Looking ahead, we plan to further enhance the system by
incorporating features like machine learning for smarter recommendations, geolocation services for
precise tracking, and voice recognition for hands-free interactions. Additionally, we aim to support
multiple languages, collaborate with third-party services, and continuously improve security
measures. Our goal is to keep evolving and providing a cutting-edge and engaging food delivery
experience for our users.
 REFERENCES

[1] Yuhua Li, David McLean, Zuhair A. Bandar, James D. O’Shea, Keeley Crockett, “Sentence
Similarity Based on Semantic Nets and Corpus Statistics”, IEEE Transactions on Knowledge and
Data Engineering, Volume 18 - No. 8, August 2006
[2] Emanuela Haller, Traian Rebedea, “Designing a Chat-bot that Simulates an Historical Figure”,
IEEE Conference Publications, July 2013.
[3] Pratik Slave, Vishruta Patil, Vyankatesh Gaikwad, Girish Wadhwa, “College Enquiry Chat Bot”,
International Journal on Recent and Innovation Trends in Computing and Communication, Volume
5, Issue 3, March 2015.
[4] “AIML Based Voice Enabled Artificial Intelligent Chatterbot”, International Journal of u- and e-
Service, Science and Technology Volume 8 - No. 2, 2015.
[5] Conversational Restaurant Recommendation with Spoken Language Understanding by Dilek
HakkaniTur, Dur Sun Hakkani-Tur, and Lawrence Heck (2016).
[6] An Overview of Chatbot Technology by Chathurika Wickramaratna, Dimuthu Udugama, K. P.
Sumudu Manthilake, and Manoj Wijewickrema (2017)
[7] The Effect of Website Interface Design on User Experience in Online Food Ordering Systems by
Yuhui Sun, Yuhong Wang, and Jing Wang (2019).
[8] Chatbots in Education and Research: A Critical Examination of Ethical Implications and Solutions
by Ana-Paula Correia, Paula de Jesus Pereira, and Ana Alice Baptista (2020)
[9] A Personalized E-commerce Recommendation System with Natural Language Processing by
Jingjing Niu, Yuxia Zhang, and Jing Li (2020)
[10] The Impact of Chatbot Design on User Experience in Online Shopping by Yujia Xu, Li Luo, and
Zhiwei Liao (2020)
[11] Research on Natural Language Understanding and Chatbot Development by Yang Liu, Xiaoyue
Shang,and Qingming Liu (2021).
[12] Factors Influencing User Satisfaction with Online Food Ordering Platforms: A Meta-Analysis by
Liang Luo, Hang Li, and Yuhui Sun (2021).
[13] Future directions for chatbot research: an interdisciplinary research agenda by P.M. Botturi (2019)
[14] Role of AI chatbots in education: systematic literature review by A. A. Ziatun, S. A. Abdul-
Rahman, and R. Alias (2020)