© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.

org(ISSN-2349-5162)

Chatbot using Natural Language Processing (NLP)

Techniques
1Sunny Kaushik, 2Rahul
1.2
Department of Computer Science & Engineering, Ganga Institute of Technology & Management, Kablana

ABSTRACT

The advent of chatbots powered by natural language processing (NLP) has changed the way humans interact with
machines. This article focuses on the process of building a chatbot using NLP techniques, exploring key concepts such
as natural language understanding, speech management, and natural language design. Through extensive literature
review, various NLP algorithms, tools, and techniques are reviewed for developing context-aware chatbots. The
process includes data collection, prioritization and application of NLP techniques. Measure the effectiveness of
chatbot design using appropriate metrics, including user satisfaction. The findings highlight the potential and
limitations of chatbots, paving the way for future improvements in chatbot design and operation. Overall, this
research contributes to the field of NLP-driven chatbot development, providing insight into the potential of these
systems to improve human-powered communication.

Inevitable Keywords- Natural Language Processing (NLP), Chatbots, Human-Machine Interaction, Natural Language
Understanding, Speech Management, NLP Algorithms, Data Collection, Human-Powered Communication.

INTRODUCTION
Technological advances have changed the way we interact with systems and access information. In recent years, chatbots
have become a common solution for human-machine communication. Chatbots are computer programs designed to
simulate human interaction, provide users with interactive interactions, and receive information or services. Natural
language processing (NLP) technology has become a tool to make chatbots smarter and able to understand and respond
to natural language. By harnessing the power of NLP technology, chatbots can interpret user questions, extract valuable
information, and generate meaningful responses in conversations. NLP technology enables chatbots to understand and
respond to user intent, manage context and uncertainty, and provide personalized and context-aware interactions.

The purpose of this article is to explore the process of creating chatbots using NLP techniques, focusing on key concepts
such as comprehension, speech management, and natural language design. Understanding the relevant language
reflects the purpose and meaning of the user's questions while managing the conversation to provide consistent answers
and appropriate content throughout the interaction. Natural language design helps create a more relevant and relevant
human response.

The potential applications of NLP-powered chatbots are enormous. Customer service, healthcare, e-commerce,
education, etc. they can be used for many purposes. Chatbots can perform routine tasks, provide instant answers to
frequently asked questions, provide personalized advice, and assist users in complex journeys or processes. They can
also improve the user experience, increase productivity and reduce the burden on human resources.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d200
© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

Figure 1.1 How a Chatbot Works

However, creating chatbots using NLP techniques presents many challenges. These challenges include
addressing ambiguity, properly understanding and achieving user goals, managing content in ongoing
conversations, and creating appropriate content and responses. Overcoming these challenges requires
carefully selected and implemented NLP algorithms, robust training data, and effective speech management
strategies.
This article aims to explore the intricacies of creating chatbots using NLP techniques and explore available
methods, frameworks and tools. This study will provide an in-depth analysis of NLP algorithms, their strengths,
limitations, and usability for chatbot development. Additionally, the chatbot creator will be evaluated using
appropriate indicators to measure its effectiveness efficiency in providing a smooth conversation.

By understanding the process of creating NLP-powered chatbots, researchers and practitioners can understand
their potential and limitations. In addition, this research will contribute to the advancement of NLP technology
and its effective use in improving human-computer communication.

In the next section, this article will introduce the current literature on NLP techniques, discuss the chatbot
creation process, introduce the design and use of content, evaluate the effectiveness of chatbots, and review
their research work, NLP-focused Contributions, and possible future directions in chat bot development.

2. BACKGROUND DETAILS

2.1 Impact on society

Beyond customer service, chatbots have significantly improved information retrieval processes. They excel in quickly
providing users with answers to their queries, offering a user-friendly and efficient way to access information. Moreover,
chatbots have been instrumental in automating repetitive tasks across various industries, freeing up human resources to
focus on more complex and creative endeavors. The advent of chatbots powered by Natural Language Processing (NLP)
has left an indelible mark on society. One of the most notable impacts is the transformation of customer service and
support. Chatbots have revolutionized customer interactions by providing 24/7 assistance, reducing response times, and
enhancing user experiences. This has not only improved customer satisfaction but also enabled businesses to operate more
efficiently, as they can handle a higher volume of customer inquiries without the need for extensive human intervention.
2.2 Pros & Cons

Chatbots come with an array of advantages that have contributed to their widespread adoption. Perhaps one of the most
significant benefits is their 24/7 availability. Chatbots can operate round the clock, ensuring that support or assistance is

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d201
© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

available at any time, accommodating users from different time zones and catering to their needs promptly. This
uninterrupted service can lead to increased user satisfaction and loyalty. From a cost perspective, chatbots are cost-
effective solutions. They significantly reduce operational costs associated with customer service and support staff. Once
developed and deployed, they require minimal maintenance and can handle a high volume of inquiries simultaneously,
making them a cost-efficient choice for businesses. Scalability is another advantage of chatbots. They can handle multiple
conversations simultaneously, scaling effortlessly to meet the growing demands of users. Furthermore, chatbots offer
consistency in their interactions. They provide standardized responses and information, eliminating human errors that can
occur due to factors such as fatigue whereas, despite their numerous benefits, chatbots also come with certain limitations.
One of the primary drawbacks is their limited understanding of nuanced or complex queries. While they excel at handling
straightforward and structured inquiries, they may struggle with comprehending the intricacies of human language,
leading to user frustration. Another critical limitation is the lack of empathy exhibited by chatbots. Unlike humans, they
cannot genuinely understand or empathize with the emotions and concerns of users. This can be a significant disadvantage
in customer interactions that require a compassionate or empathetic response, such as dealing with emotional or sensitive
issues. The initial development costs associated with creating advanced chatbots can be substantial. Building and training
chatbots that are capable of sophisticated language processing and context-aware responses can require a considerable
investment in terms of technology, expertise, and resources. Lastly, there is a concern about over-reliance on chatbots
potentially diminishing human interaction and impacting social skills. While chatbots are invaluable tools for automation
and efficiency, there is a balance to strike between human and automated interactions to maintain the richness of human
communication.
2.3 Algorithm used

Chatbots leverage various algorithms to perform their tasks effectively. Rule-Based Algorithms are a foundational
approach where chatbots operate based on predefined rules and patterns. They are suitable for handling structured queries
and straightforward tasks. Machine Learning Algorithms, including decision trees, random forests, and support vector
machines, are used in chatbots that can learn from data and improve over time. These algorithms enable chatbots to adapt
to changing user needs and preferences. Natural Language Processing (NLP) algorithms are at the heart of chatbots that
deal with human language. This includes recurrent neural networks (RNNs), LSTM (Long Short-Term Memory)
networks, and advanced Transformer models like BERT and GPT-3. Intent Recognition Algorithms are crucial for
understanding user intent, helping chatbots route conversations or make decisions based on user queries. Dialog
Management Algorithms maintain context in conversations, ensuring that chatbots provide coherent and contextually
relevant responses. Sentiment analysis algorithms gauge user emotions based on their language, allowing chatbots to tailor
responses or assess customer satisfaction. Reinforcement Learning may also be employed in some chatbots to optimize
responses based on user feedback, improving their performance over time. These algorithms collectively empower
chatbots to function intelligently and efficiently in a variety of applications, making them a transformative force in modern
technology.
3. LITRATURE SURVEY

This is current research and progress in the development of chatbots using natural language processing (NLP).
It explores the various methods, algorithms and techniques used in the development of NLP-driven chatbots,
highlighting their strengths, limitations, and potential applications. The field of Natural Language Processing
(NLP) contains many techniques and algorithms that enable machines to understand and reproduce human
language. This section introduces the basics of NLP about building chatbots.

It covers prewriting techniques such as word segmentation, root extraction and lemmatization. It also explores
modeling techniques and recurrent neural networks (RNNs), which form the basis of many NLP applications,
including chatbot development.
3.1 Approaches to chatbot development
Various approaches have been proposed to develop chatbots, each with its own advantages and limitations.
Rule-based systems use rules and precedents to determine appropriate responses. Structural models,
including contiguous and variable models, generate responses based on learning models and contexts. This
section explores the strengths, weaknesses and viability of each approach to building NLP-powered
chatbots.
JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d202
© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

3.2 NLP Techniques for chatbot development

Building effective chatbots requires a combination of NLP techniques. This section covers specific techniques
related to chatbot development. It includes functions such as natural language understanding (NLU), target
recognition, location recognition, and content analysis, and enables chatbots to understand user queries.
Conversation management focuses on keeping conversations meaningful and coherent using techniques such
as situation monitoring, policy management, and process support work. Natural Language generation (NLG)
technology enables chatbots to generate human-like and contextually balanced responses.

This section examines the algorithms, models, and applications in each NLP, providing information on their use
and potential problems.

Figure 3.2: NLP Techniques for chatbot development

Available chatbot platforms and frameworks

Various chatbot development platforms and frameworks have been introduced to simplify the process of
building NLP-driven chatbots. It discusses the features, capabilities, and ease of use of these platforms,
including their support for NLP activities, integration options, and easy deployment.
Understanding platforms and operating systems helps researchers and professionals make informed
decisions when choosing appropriate chatbot development tools.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d4

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

Table 1: Comparison in NLP Techniques for chatbot development table

3.3Chatbot performance Evaluation Metrics

Evaluating a chatbot's performance is important for evaluating its effectiveness and user satisfaction. This
section covers various metrics used to measure chatbot performance. Objective measures such as accuracy,
precision, recall, and F1 score can be used to evaluate the effectiveness of NLU and interactive
management.
Confusion is a measure of the quality of language construct that evaluates the effectiveness and
consistency of the responses produced. User-based metrics, including user satisfaction, engagement, and
completion rates, provide insight into the chatbot's overall performance from the user's perspective. This
section highlights the importance of choosing the right metrics based on the chatbot's purposes and context.

4. Methodology

4.1 Data collection and preprocessing

Data collection and preprocessing are important steps in building chatbots using NLP techniques. This step
includes collecting important data and preparing for training and evaluation. The collected data will serve as
the basis for training NLP models and improving chatbot performance. Data collection and pre-processing can
be divided into the following stages:

1.Define the required information: To create a chatbot, start by specifying the specific information. Determine
the purpose, the type of interaction the chatbot will handle, and the messages it will support.

2.Check data source: Check possible data based on chatbot name and user interaction. These resources may
include customer support information, online forums, social media platforms, or general information. Also,
consider creating a unique document using methods such as crowdsourcing or manual annotation if needed.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d5

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

3.Data Collection: Collect data from data analysis. This may include web downloads, API integrations, or
writing guides through user research or interviews. Make sure the documentation covers a wide range of user
questions, including different language styles, concepts, and organizations.

4.Data cleaning and processing: When data is collected, it must be cleaned and processed to eliminate noise,
inconsistencies and sensitive data. The following operations are performed before and during the maintenance
of the data:
a. Text normalization: Uniform text by removing capital letters, converting to lowercase, and handling
abbreviations or typos.

b. Stop word deletion: remove words that have no significant meaning, such as "the", "a" and "is".

Tokenization: Splitting text into individual words or symbols for further analysis and processing.
Lemmatization / Stemming: Simplifies words into their basic or root forms to address differences and improve
consistency.

5. Notes: According to Chatbot's requirements, mark the document with important text such as thoughts and
names. Annotations can be made manually or with mechanical equipment. The collected data can lead to
supervised learning and help train good NLP models.

6. Data augmentation: If data storage is limited, consider augmenting data by creating additional models such
as data links, annotations, or translation. This helps to diversify the data and improve the performance of the
chatbot when processing different user queries.

4.2 Natural Language Understanding (NLU)

NLU is an essential part of building chatbots using NLP techniques. It focuses on extracting meaning and
purpose from user questions, ensuring that chatbots understand and respond correctly. NLU includes many
functions such as target recognition and name recognition, which are important for effective communication
between chatbots and users. The table below shows the importance of NLU when building a chatbot:

4.2.1 Purpose Goals

Purpose Goals attempt to identify the intent or purpose behind the user's question. It involves defining the
goal or task that the user wants to accomplish. For example, in a customer support chatbot, various targets
might include "information requests", "status queries" or "request requests". Authentication enables chatbots
to provide relevant and helpful responses.

There are several methods for setting targets, including:

-Policy Based Policy: Establishing predefined criteria or standards that specify certain phrases or terms
important to target consistency. This approach is easy to use but requires manual development and
maintenance of the code.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d6

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

Machine Learning Algorithms: This model is trained on collected data, where each user's question is written
with a corresponding target. Features such as word embeddings, n-grams, or syntactic models can be used to
represent questions.

The model learns to classify new questions according to these features.

-Deep Learning Models: Deep Learning Models such as RNN or Transformers have shown great performance
in cognitive processing. These models can capture the gist of queries and learn complex patterns in data.
Training deep learning models often requires larger datasets and large computational resources.

Evaluation of objective recognition models can be done using metrics.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d7

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

Figure 3.1: Flowchart of Work

4.2.2 Name Entity Recognition

Named Entity Recognition (NER) focuses on identifying specific entities or information related to user queries.
Entities may include names of persons, organizations, places, dates or other relevant information.

NER is essential to extract important information from user queries and provide accurate and personalized
answers.

There are many methods for NER, including:

-Rule-based methods: Create a predefined set of rules or patterns to recognize domain names based on usage
patterns or language patterns. This approach is suitable for certain species or restricted areas.

-Statistical Models: Use statistical models such as random field (CRF) or hidden Markov models (HMM) to label
each word or token in the query with its corresponding properties. This model learns from recorded data and
stores the details of language progress.

-Deep learning models: Deep learning models such as Bidirectional Long-Term Memory (BiLSTM) or Transformer-
based models such as BERT have been shown to be successful in NER projects. These patterns can capture
complex speech patterns and contexts to accurately describe nouns.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d5

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

The NER model is trained using a dataset of entities registered in the user query. Continuous development and
updating of NER modules are necessary to maintain new facilities and improve accuracy. By integrating good
authentication and naming authentication processes into the NLU module of chatbots, chatbots can better
understand user queries and provide relevant information.

This improves the overall user experience by enabling chatbots to provide more accurate and meaningful
responses.

Figure 3.2: Name Entity Recognition

4.3 Conversation Management

Conversation Management is the key to creating chatbots using NLP techniques. It aims to control the flow of
conversation between the chatbot and the user to ensure the right relationship and content. Session
management has two main components: state monitoring and policy management.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d6

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

4.3.1 Status Monitoring

Status monitoring refers to the process of maintaining and updating the internal representation of the
conversation status. It contains important information such as session state, user preferences, system activity,
and details of past interactions. Discovery allows chatbots to monitor ongoing conversations and provide more
personalized and relevant responses.

There are several methods for health monitoring, including:

-Transactional rules: Define a predefined process or state variable that stores relevant information during a
session.

These changes are updated based on user input and system activity. While legal systems are easy to implement,
they may not be successful and may require manual development and management of policies.

-Pattern Collection: Use pattern collection based on machine learning algorithms or deep learning models. This
model tracks features or differences during the session and updates them based on customer input. Slot filling
models often require a data-driven approach, in which every session row is filled with proximity and value.

-Session State Monitoring with Memory Networks: Memory Networks uses external memory to store and store
information from previous sessions. This model can capture long-term prospects and store details of many
intersections.

It is important to update and improve the status monitoring module as the chatbot meets new speech patterns
and user preferences.

4.3.2 Control Policy

Control Policy includes determining the appropriate response or the response to the chatbot. The policy defines
the chatbot's decision-making process by specifying the actions to be taken to provide conversational content.

Different approaches to policy management include:

-Policy-based: Create a predefined process or decision tree that specifies specific dialogs for actions. These
guidelines may be based on firm standards or expert knowledge.
Rule-based rules are defined but can be limited to handling complex negotiation situations.

-Rule-Based Reinforced Learning: Uses reinforcement learning techniques to learn good rules through trial and
error. Chatbots interact with users and their actions are encouraged as feedback or rewards. Rule-based learning
support works best when a simulation or environment for training is available.

These models can manage complex interactions and learn from big data.

Policy management models are trained and evaluated using interactive data, including user input, performance,
and corresponding rewards or evaluations. Metrics such as success, user satisfaction or success rate can be used
to measure the effectiveness of the policy management module.

By combining good health analysis and policy management, chatbots can manage context-sensitive
conversations and make decisions based on the context of the conversation.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d8

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

4.4 Natural Language Generation (NLG)

NLG is an essential part of building chatbots using Natural NLP. NLG focuses on creating human-like responses
and appropriate content to engage users in a positive and meaningful way. NLG plays an important role in
making chatbot responses integrated, informative and tailored to the user's questions. The following sections
discuss the importance of NLG when creating a chatbot:

Figure3.3: Natural Language Generation architecture

4.4.1 Response Generation

Response generation involves the translation of structured or semantically represented information into natural
languages. It aims to create effective and coherent sentences that convey the desired content to the user. The
answers created should be topic-oriented and address the user's question or goal.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d9

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

There are many ways to respond:

- Template-based approaches: Template-based approaches involve pre-defining a set of response templates

covering different scenarios or purposes.
This structure can contain fields for dynamically populated data. The template-based approach is easy to use,
but may not be easy and natural.

- Rules-Based Approach: A rule-based approach uses rules or predefined patterns to generate answers based on
questions or discussion points. These rules can be language or information specific. A policy-based approach
was developed, providing greater flexibility and control over the response.
- Statistical methods: Statistical models such as linguistic models or convolutional models can be trained to
generate answers based on input questions or discussion points. This model learns the possible effects of words
or strings and generates responses accordingly. Neural language models such as neural networks (RNNs) or
Transformer-based models such as GPT have shown good results in response.
- Reinforcement learning techniques: Reinforcement learning techniques can be used to train design models.
Chatbots interact with users and the responses generated are supported as user feedback or rewards. Supportive
learning can help improve user satisfaction and response to other pre-existing conditions.

Evaluation of the response pattern can be evaluated by human evaluation or using a range of criteria such as
intelligence, relevance, and relevance to the user's question. It is important to continually adjust and improve
the response design to improve the conversational capabilities of chatbots.

4.4.2 Personalization and Customization

In NLG, personalization and customization includes creating responses that suit the user's preferences,
characteristics or past interactions. Personalization can increase the effectiveness of chatbots, creating a more
engaging and tailored user experience. Techniques such as user profiling, content monitoring or network
memory can be used to customize NLG.
Combining effective response generation techniques and personalization techniques, chatbots can generate
effective responses and content that improves the overall user experience. The ability to generate consistent and
appropriate responses plays an important role in creating chatbots that users find useful and engaging.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d9

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

Figure4.4: Personalization and Customization

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d10
© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

4.5 Chatbot Integration and Deployment

The next step after creating a chatbot using Natural Language Processing (NLP) techniques is to integrate it
into the desired platform or application and send it to users for interaction. Integrating and deploying a Chatbot
involves many considerations to ensure seamless integration, scalability, and performance. The following
article discusses the main points of chatbot integration and referrals:

4.5.1 Platform Integration:

Chatbot Integration involves integrating the chatbot into the desired platform or application on which it will
be deployed. This may include websites, mobile apps, messaging platforms or voice assistants. The integration
process will vary based on current platform requirements and APIs for communication.

For web integration, chatbots can be embedded in web pages using HTML, JavaScript, and CSS. The chatbot's
interface and functionality can be customized to fit the overall design and user experience of the platform.
For mobile application integration, chatbots can be integrated using the software development kit (SDK) or
API provided by the platform. This allows chatbots to integrate into the UI and functionality of mobile apps.

Bot integration for messaging platforms like Facebook Messenger or Slack will create a bot app or bot account
and install the necessary APIs and webhooks to enable communication between the bot and the distribution
platform.

Figure3.5: Platform Integration

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d11
© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

4.5.2 Backend Infrastructure:

Chatbots need a strong backend infrastructure to be reliable and efficient. This includes servers, databases, and
other components needed to process user requests, perform NLP operations, and store related information.

Backend infrastructure must be able to manage user interactions, use NLP techniques to process and analyze
user queries, and provide real-time responses. It may involve using chatbots on cloud-based platforms such as
Amazon Web Services (AWS), Microsoft Azure, or Google Cloud Platform (GCP) to take advantage of their
robustness and flexibility.

To ensure efficient operation, infrastructure must be designed to handle the largest vehicles and state-of-the-art
equipment.

It should also include monitoring and recording processes to track usage, identify bottlenecks, and resolve issues.
Figure4.6: Backend Infrastructure

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d2

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

4.5.3 Integration with NLP services:

Depending on the complexity and needs of the Chatbot, integration with external NLP services may be
required. These services may provide additional functionality such as translation, opinion analysis or site
validation.

Popular NLP services include Google Cloud Natural Language API, IBM Watson NLU, or Microsoft Azure
Cognitive Services. Integration with these services usually includes configuring API access, managing data
exchanges, and processing responses received from the service

4.5.4 Development and Ongoing Maintenance: After the Chatbot is deployed, continuous development and
maintenance is essential to ensure its effectiveness and user satisfaction. User feedback, notes, and reviews can be
analyzed to identify areas for improvement and improve chatbot functionality.

Chatbot updates may include improving NLP models, adding new concepts or entities, or supporting resource creation.
It is important to monitor user interactions, solve problems or limitations, and adapt the chatbot to user needs and
changing needs.

Regular maintenance and updates may include bug fixes, security patches and optimizations to ensure chatbot
reliability and security.

In summary, it includes chatbot integration and commissioning, integrating the chatbot into the desired platform,
establishing a secure back-end infrastructure, providing integration with NLP service if necessary, making
improvements and maintenance.

A well-integrated and effective chatbot can ensure efficient and effective communication, ensuring user engagement
and satisfaction.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d3

© 2023 JETIR September 2023, Volume 10, Issue 9 www.jetir.org(ISSN-2349-5162)

References

[1] WHO survey for motorcycle crash deaths.

[2] Joseph Redmon, Santosh Divvala, Ross Girshick, Ali Farhadi, You Only Look Once: Unified, Real-Time Object
Detection.

[3] Veras, Romuere R.V. Silva Kelson R. T. Aires Rodrigo de M. S., "Detection of helmets on motorcyclists".

[4] Felix Wilhelm Sieberta, Hanhe Linb, "Detecting motorcycle helmet use with deep learning," 2020.

[5] M. Swapna, Tahniyath Wajeeh, Shaziya Jabeen, "A Hybrid Approach for Helmet Detection for Riders".

[6] Khan, F.A., Nagori, N.S., & Naik, A.K., "Helmet and Number Plate detection of Motorcyclists using Deep
Learning and Advanced Machine Vision Techniques," 2020.

[7] M. Dasgupta, O. Bandyopadhyay and S. Chatterji, "Automated Helmet Detection for Multiple Motorcycle
Riders using CNN".

[8] Yonten Jamtsho, Yonten Jamtsho, Rattapoom Waranusast, "Real-time license plate detection for non-
helmeted motorcyclist using YOLO," 2022.

[9] P. Sathe, A. Rao, A. Singh, R. Nair and A. Poojary, ""Helmet Detection And Number Plate Recognition Using
Deep Learning,," 2022.

[10] JOUR, Sajid Muhammad, Waris, Tasbeeha, "CNN-Based Automatic Helmet Violation Detection of
Motorcyclists for an Intelligent Transportation System," 2022.

[11] "Image from stackoverflow".

[12] Jain, Harshil & Nandy, S., "Incremental Training for Image Classification of Unseen Objects.," 2019.

[13] Mrs. Vinaya Kulkarni, Dhanashree Pawar, Sanskruti Talwekar,Rupali Bharambe, Akshata Mahadik, "Helmet,
Number Plate Detection and Stolen," 2023.

JETIR2309320 Journal of Emerging Technologies and Innovative Research (JETIR) www.jetir.org d4