AI CHAT BOT USING SHAN ALGORITHM

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 10 Issue: 05 | May 2023 www.irjet.net p-ISSN: 2395-0072
© 2023, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 112
AI CHAT BOT USING SHAN ALGORITHM
1. Mrs. S. Nandini (Ph.D)1, A.Nawas Hussain2, B.S. Haran Pranav3, A. L. Abdul Vahith4, J.
Samuel George5
1 Professor of Department of Information Technology, Meenakshi College of Engineering, Chennai, Tamil Nadu,
India
2,3,4,5 Student of Department of Information Technology, Meenakshi College of Engineering, Chennai, Tamil Nadu,
India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract-
Letters dominated previous communication. Then, when
telephones and, later, mobile phones became more
widespread, voice chats took over as the major mode of
communication. In many cases, a chatbot can be useful in
providing services. These services range from weather
forecasts to the option to purchase a new laptop,
smartphone, or anything in between. They also provide
life-saving health alarms. Numerous big firms, like Google
(Google Assistant), Amazon (Alexa), Microsoft (Cortana),
and Oracle, are investing substantial time and resources in
the research of personal assistants. The development of a
chat-bot would cover the following topics: Image
recognition with Custom Vision services is utilised with
Azure Bot Architecture. We present a SHAN algorithm that
combines NLP, RNN, and LSTM.
Keywords—azure bot, NLP, RNN, SHAN ALGO
1.INTRODUCTION:
In our daily interactions with friends, family, and co
workers, we learn about the context of the topic being
discussed. When someone states they are reading a book,
you might inquire about the author or whether they enjoy
the book rather than asking if they have read any other
books. You give the greatest response you can at the time.
A chatbot is a piece of software that mimics human
communication through text or voice exchanges. It is
intended to automate processes and give people
information.
Various platforms, including websites, messaging
services, and mobile applications, can incorporate
chatbots. An AI chatbot is a computer programme that
simulates human-like discussions with people using text-
based or voice-based interfaces. These chatbots can be
used for a wide range of applications, including customer
service, virtual assistants, and even entertainment. To
interpret and respond to user inputs, AI chatbots employ
natural language processing (NLP) and machine learning
algorithms. They are trained on massive volumes of data
and are able to learn from user interactions, allowing them
to improve their responses over time. Chatbots can be
connected into a variety of platforms, such as websites,
messaging applications, and social networking platforms,
allowing users to interact with businesses or services in a
seamless and convenient manner. They can also handle
many chats at the same time, decreasing human agents'
workload.
1.1 Existing System
The purpose of a chatbot system is to provide a
seamless and efficient means of communication between
humans and computers. The system is designed to
simulate a human conversation, where the user can input
their query or request in natural language, and the chatbot
responds with relevant information or assistance[1].
To achieve this goal, the chatbot system's architecture
integrates a language model and computational algorithm,
which work together to process and interpret the user's
input. The language model allows the chatbot to
understand and generate natural language responses,
while the computational algorithm enables the chatbot to
access and retrieve information from various sources to
provide accurate and relevant responses to the user.
One of the key benefits of using a chatbot system is its
accessibility. Anyone, from employees to the general
public, can freely upload their queries and receive
immediate responses. This makes the chatbot system a
useful tool for businesses and organizations that want to
provide efficient customer service or streamline their
internal communication processes. Additionally, the use of
chatbots can also reduce the workload of human customer
service representatives, allowing them to focus on more
complex tasks that require human intervention.
1.2 Proposed System
The AIML-based bots have been popularly used in the
past, other algorithms can also be implemented in chatbot
systems to provide improved functionality and
performance. For instance, advanced machine learning
algorithms such as deep learning models can be used to
improve the accuracy of the chatbot's responses.

In addition, chatbot systems can also be designed to
accept text-based queries from users. This means that
users can input their queries in the form of text, and the
chatbot system will provide a text-based output in
response. This can be useful for users who may have
difficulty with speech-based interactions or for situations
where privacy concerns make it difficult to use speech-
based inputs.
Once a chatbot system has been successfully executed in
one domain, such as a college domain, it can be extended
to other domains such as medical, forensic, sports, and
many others. This can be highly beneficial as it enables
users to access relevant information quickly and easily
without spending much time sorting through irrelevant
data.
For example, in the medical domain, a chatbot system
can be designed to provide users with information on
medical conditions, symptoms, and treatments. Similarly,
in the forensic domain, a chatbot system can be designed
to provide users with information on crime scene
investigations, evidence collection, and analysis. In the
sports domain, a chatbot system can be designed to
provide users with information on team statistics, player
performance, and game schedules.
Overall, chatbot systems can be highly beneficial in a
variety of domains as they provide users with quick and
easy access to relevant information. By implementing
different algorithms and text-based queries, chatbot
systems can be designed to meet the specific needs of
users in different domains.
2. ALGORITHM USED
The nlp, rnn, lstm are three well-known algorithms
capable of handling sequential structural data. The
combination of the above three is Shan algorithm This algo
is used to resolve the query of the user.
2.1. Natural Language Processing
The term "Natural Language Processing" (NLP)
refers to the area of artificial intelligence that studies how
computers and human language communicate. It entails
the creation of computational models and algorithms that
let machines decipher, comprehend, and produce natural
language. Language translation, sentiment analysis, voice
recognition, and chatbots are just a few of the many uses
for NLP. Enabling computers to comprehend the nuances
of human language, including grammar, syntax, semantics,
and context, is the main problem in NLP. Several methods,
including machine learning, deep learning, and natural
language comprehension, are used in NLP. Deep learning
models employ neural networks to comprehend the
content, while machine learning algorithms are used to
train models that can recognise patterns in language data.
It is designed to understand natural language inputs
and generate human-like responses
2.2. Recurent neural network
Chatbot learns the statistical relationships and
patterns between words and phrases in natural language
by using a significant quantity of training data. Massive
amounts of text from the internet and other sources make
up the training data, which is used to build the model using
unsupervised learning methods. Once trained, the model
can create text by anticipating the following word in a
sequence based on the words that came before.
When a user enters a message, the chatbot uses its
linguistic expertise to produce an answer that is most
likely to be logical and relevant to the message. A
probabilistic sampling process is used to produce the
response, where the model generates multiple potential
responses and chooses the one that is most likely to occur
based on a probability distribution
2.3. LSTM
The LSTM algorithm is a type of recurrent neural
network (RNN) that is designed to address the vanishing
gradient problem and enable effective processing of
sequential data.
The LSTM network consists of memory cells and
three gating mechanisms: input gate, output gate, and
forget gate. The memory cell is a long-term memory that
can store information over long periods of time. The input
gate regulates how much new information should be
added to the memory cell, while the forget gate
determines which information should be removed from
the cell. The output gate controls how much information
should be output from the cell to the next layer of the
network. The LSTM algorithm processes data in a
sequence, with each element in the sequence being
processed by the network one at a time. For each element,
the input gate determines how much new information
should be added to the memory cell, based on the current
input and the previous state of the memory cell.
The forget gate decides which information should
be removed from the memory cell, based on the current
input and the previous state of the memory cell. Finally,
the output gate controls how much information should be
output from the memory cell, based on the current input
and the current state of the memory cell. The LSTM
algorithm has been used in a variety of applications,
including speech recognition, natural language processing,
and image captioning. It has also been used to generate
new sequences of data, such as music and text, by training

the network on existing sequences and then using it to
generate new sequences based on the learned patterns.
Overall, the LSTM algorithm has proven to be highly
effective in processing sequential data and has opened up
new possibilities for the analysis and generation of
complex data sequences.
SHAN ALGORITHM
Natural language processing (NLP) combines
computational linguistics, machine learning, and deep
learning models to process human language.
Computational linguistics. Computational linguistics is the
science of understanding and constructing human
language models with computers and software tools.
STEP~: AUTHENTICATE TEXT MESSAGE
Step 2: search in global server
Step 3: send text message
LSTM used in chatbot
In the domain of chatbots for time series conversations,
LSTM is shown to perform well and maintain the context
for longer durations. LSTM network.
STEP 1: Message view
STEP 2: The adaptive message transmission
STEP 3: Request sever for shortes form and summary of
client query
RNN
The RNN is a stateful neural network, which means that it
not only retains information from the previous layer but
also from the previous pass. Thus, this neuron is said to
have connections between passes, and through time.
STEP1:relate to answer for previous query information
STEP2: adapt the previous message transfer
STEP3:answer related to previous search and adapt the
message query
2.4. MLP
MLP stands for Multilayer Perceptron, which is a
type of neural network architecture used in machine
learning for supervised learning tasks, such as
classification and regression. The MLP network consists of
an input layer, one or more hidden layers, and an output
layer. Each layer consists of a number of artificial neurons
or nodes, which are connected to the nodes in the previous
and next layers by weighted connections. In the MLP
architecture, the input layer receives the input data, which
is then passed through the hidden layers. The hidden
layers perform computations on the input data using non-
linear activation functions, and the output of each hidden
layer is passed to the next hidden layer until the output
layer is reached. The output layer produces the final -
classifications.
The training of an MLP network involves
adjusting the weights of the connections between the
nodes to minimize the error between the predicted output
and the actual output. This is typically done using
backpropagation, which involves calculating the gradient
of the error with respect to the weights and using this
gradient to update the weights.
MLP networks have been used in a variety of
applications, including image and speech recognition,
natural language processing, and financial forecasting.
They are particularly effective for classification tasks
where the input data has non-linear relationships between
the features.
Fig.1: Multilevel Perception
Fig.2: Architecture for Proposed System
Client query: client can request question through the text
via message box it will be resolve for query
Bot reply: bot can reply for a query and it can be search
for the query by the related answer
Knowledge base : Create a knowledge management
strategy.

Choose your infrastructure.
Determine and collect the data your AI needs.
Make the data simple and accessible for AI.
Adjust the language to fit your chatbot's persona.
Get started with self-learning AI from User like.
Data store: data are collected in a data base and solve for a
related query
Knowledge store: it can Knowledge store is a data sink
created by a Cognitive Search enrichment pipeline that
stores AI-enriched content in tables and blob containers in
Azure Storage for independent analysis or downstream
processing in non-search scenarios like knowledge mining.
Action: it can be accessed for a message what a chat bot
replies for client query
Modules:
1.Python library Authentification
Intents. json: intents classification or recognition it is a
type of getting a spoken or written text and
then classifying it based on what the user wants to
achieve.
Trainer.py: Defines the Chatbot overall file structure and
contains the intent, actions, slots, stories, domain, config
and endpoint details. The code will train an NLU and
dialogue model to retrieve weather from the Yahoo
weather API. Model folder contains the trained models. It
will also start the server with actions and also runs the
chatbot on the command line. Execute only this code as it
will trigger the actions and run.py.
Run.py: triggered by trainer.py. contains the modules to
run the chatbot module in the command line.
2. RNN (seq2seq model)
A Seq2Seq model is a model that takes a sequence of items
(words, letters, time series, etc) and outputs another
sequence of items.
In the case of Neural Machine Translation, the input is a
series of words, and the output is the translated series of
words.
Sequence to Sequence (often abbreviated to seq2seq)
models is a special class of Recurrent Neural Network
architectures that we typically use (but not restricted) to
solve complex Language problems like Machine
Translation, Question Answering, creating Chatbots, Text
Summarization, etc.
Fig.3: Performance Metrics
OLD ALGORITHMS LIKE NLP&LSTM&RNN CAND BE
DEFINED IN BLUE LINE
SHAN ALGORITHMS CAN BE DEFINED IN GREEN LINE IT
HAVE A HIGHER INFORMATION VALUES
SCREEN SHOTS
10. 12. 15. 17. 20.
0.0
0.2
0.5
0.7
1.0
1.2
1.5
1.7
Cr
os
s-
e
nt
ro

3. CONCLUSION
A chatbot is a piece of software that mimics human
communication through text or voice exchanges. It is
intended to automate processes and give people
information. Various platforms, including websites,
messaging services, and mobile applications, can
incorporate chatbots.
REFERENCES
[1] Milla T Mutiwokuziva, Melody W Chanda, Prudence
Kadebu, A neural network based chat-bot , 2nd
International Conference on Communication and
Electronics Systems (ICCES 2021).
[2] M. Young, The Technical Writer’s Handbook. Mill
Valley, CA: University Science, 1989.
[3] R. Nicole, “Title of paper with only first word
capitalized,” J. Name Stand. Abbrev., in press.
[4] Milla T Mutiwokuziva, Melody W Chanda, Prudence
Kadebu, A neural network based chat-bot , 2nd
International Conference on Communication and
Electronics Systems (ICCES 2017).
[5] T. Mikolov et al. “Efficient Estimation of Word
Representations in Vector Space”. In: CoRR
abs/1301.3781 (2013). arXiv: 1301.3781. url: http://
arxiv.org/abs/1301.3781.
[6] J. Pennington, R. Socher, and C. D. Manning. “GloVe:
Global Vectors for Word Representation”. In:
Empirical Methods in Natural Language Processing
(EMNLP). 2014, pp. 1532–1543. url:
http://www.aclweb.org/anthology/ D14-1162.
[7] R. Pascanu, T. Mikolov, and Y. Bengio. “Understanding
the exploding gradient problem”. In: CoRR
abs/1211.5063 (2012). arXiv: 1211 . 5063. .
[8] S. Hochreiter and J. Schmidhuber. “Long Short-Term
Memory”. In: Neural Comput. 9.8 (Nov. 1997), pp.
1735–1780. issn: 0899-7667. doi: 10.1162/neco.
1997.9.8.1735.
[9] I. Sutskever, O. Vinyals, and Q. V. Le. “Sequence to
Sequence Learning with Neural Networks”. In: CoRR
abs/1409.3215 (2014). arXiv: 1409.3215.
[10]O. Vinyals et al. “Show and Tell: A Neural Image
Caption Generator”. In: CoRR abs/1411.4555 (2014).
arXiv: 1411.4555.
[11]C. Olah. Understanding LSTM Networks. Aug. 2015.
url: http://colah. github.io/posts/2015-08-
Understanding-LSTMs/. 65 REFERENCES.
[12]K. Cho et al. “Learning Phrase Representations using
RNN Encoder-Decoder for Statistical Machine
Translation”. In: CoRR abs/1406.1078 (2014). arXiv:
1406.1078.
[13]J. Chung et al. “Empirical Evaluation of Gated
Recurrent Neural Networks on Sequence Modeling”.
In: CoRR abs/1412.3555 (2014). arXiv: 1412.3555.
[14]M. Lui, J. H. Lau, and T. Baldwin. “Automatic Detection
and Language deification of Multilingual Documents”.
In: Transactions of the Association for Computational
Linguistics 2 (2014), pp. 27–40. issn: 2307-387X.
[15]L. Meng and M. Huang. “Dialogue Intent Classification
with Long Short-Term Memory Networks”. In: Natural
Language Processing and Chinese Computing. Ed. by
X. Huang et al. Cham: Springer International
Publishing, 2018, pp. 42– 50. isbn: 978-3-319-73618-
1.
[16]K. Lee et al. “Conversational Knowledge Teaching
Agent that uses a Knowledge Base”. In: (Jan. 2015), pp.
139–143.
[17]S. Perez. Microsoft silences its new A.I. bot Tay, after
Twitter users teach it racism. 2016. url: https : / /
techcrunch . com / 2016 / 03 / 24 / microsoft -
silences - its - new - a - i - bot - tay - after - twitter -
users - teach - it - racism/.
[18]A. Tammewar et al. “Production Ready Chatbots:
Generate if not Retrieve”. In: CoRR abs/1711.09684

(2017). arXiv: 1711.09684. url: http://arxiv.
org/abs/1711.09684.
[19]C. Liu et al. “How NOT To Evaluate Your Dialogue
System: An Empirical Study of Unsupervised
Evaluation Metrics for Dialogue Response
Generation”. In: CoRR abs/1603.08023 (2016). arXiv:
1603.08023. url: http://arxiv. org/abs/1603.08023.
[20]F. Chollet et al. Keras. 2015. .
[21]K. Papineni et al. “BLEU: a Method for Automatic
Evaluation of Machine Translation”. In: 2002, pp. 311–
318. [21] C.-Y. Lin. “ROUGE: A Package for Automatic
Evaluation of summaries”. In: Proc. ACL workshop on
Text Summarization Branches Out. 2004, p. 10. M.
Svane. Zendesk..
[22]M. Abadi et al. TensorFlow: Large-Scale Machine
Learning on Heterogeneous Systems. Software
available from tensorflow.org. 2015. url:
https://www. tensorflow.org/.
[23]Theano Development Team. “Theano: A Python
framework for fast computation of mathematical
expressions”. In: arXiv e-prints abs/1605.02688 (May
2016).
[24]W. McKinney. pandas: a Foundational Python Library
for Data Analysis and Statistics.
[25]F. Pedregosa et al. “Scikit-learn: Machine Learning in
Python”. In: Journal of Machine Learning Research 12
(2011), pp. 2825–2830.
[26]D. P. Kingma and J. Ba. “Adam: A Method for Stochastic
Optimization”. In: CoRR abs/1412.6980 (2014). arXiv:
1412.6980. url: http://arxiv.org/ abs/1412.6980. [29]
C. Welch. Google just gave a stunning demo of
Assistant making an actual phone call. 2018..

AI CHAT BOT USING SHAN ALGORITHM

More Related Content

Similar to AI CHAT BOT USING SHAN ALGORITHM (20)

More from IRJET Journal (20)

Recently uploaded (20)

AI CHAT BOT USING SHAN ALGORITHM