Chat D

International Journal of Research Publication and Reviews, Vol 4, no 4, pp 1981-1993, April 2023
International Journal of Research Publication and Reviews

Journal homepage: www.ijrpr.com ISSN 2582-7421
Chatbot for Mental Health Care
R. Shanmukha Priya1, Ch. Navyamani2, M. MohammedAeliya3, Dr. V Ravi Kishore4

1,2,3
CSE, B. Tech, Aditya Engineering College(A). Surampalem
4
Ph.D, Associate Professor (CSE), Aditya Engineering College (A)
ABSTRACT—
This paper introduces a Healthcare Chatbot that utilizes Artificial Intelligence to assist in resolving basic health-related queries before consulting a doctor, thereby
reducing time and cost. The Chatbot uses Natural-Language-Processing to communicate with users and a Sequential algorithm to make decisions based on the
user's input. Additionally, users can create a profile specifying their symptoms, which the Chatbot utilizes to suggest doctors and provide dosage reminders. This
Chatbot is a useful tool for individuals seeking primary care advice and can also serve as a virtual doctor for minor mental health issues. The Chatbot can diagnose
diseases based on the symptoms provided and suggest appropriate remedies and precautions. The system also features voice output music for relax and games, as
well as appointment scheduling with time-slots. Conversational agents, such as this Chatbot, utilize NLP and ML techniques to understand user queries and provide
intelligent responses. The Chatbot is available 24/7, free of cost, and can facilitate more open communication between users and healthcare providers. In addition
to its primary functions, the Healthcare Chatbot can be programmed to offer a variety of healthcare services, including medical check-ups, therapy sessions, and
lifestyle advice. For instance, the Chatbot can offer personalized fitness plans based on the user's health status, dietary habits, and exercise preferences. It can also
provide mental health counseling and support for individuals struggling with stress, anxiety, or depression. One of the key advantages of the Healthcare Chatbot is
its accessibility. Since it can be accessed through any device with an internet connection, individuals living in remote areas or with limited access to healthcare
facilities can benefit from its services. The Chatbot can also be programmed to support multiple languages, making it a valuable tool for individuals with diverse
linguistic backgrounds. Overall, the Healthcare Chatbot represents a significant breakthrough in healthcare technology, offering an effective, accessible, and
affordable means of accessing primary healthcare services. With its intelligent algorithms and advanced NLP capabilities, the Chatbot has the potential to
revolutionize the way individuals receive healthcare services, providing personalized, convenient, and reliable care to people across the globe. In conclusion, this
Healthcare Chatbot is a useful tool for individuals seeking preliminary healthcare advice and can assist in reducing healthcare costs and time.
KeyWords:- Machine Learning (ML), Natural Language Processing (NLP), Automation , Intelligent task.
I. Introduction
Recent research from the World Health Organization reveals that depression affects approximately 264 million people worldwide. Stress and anxiety are
prevalent among individuals who are between the ages of 15 and 28 and experiencing symptoms of depression may be eligible for manifest in various
ways within this age group. Unfortunately, many people suffering from mental illnesses often perceives it as a phase and fail to comprehend its impact
on their brain cells. Chatbots could provide real-time assistance for individuals who lack the financial resources to seek professional help. The internet's
greatest advantage is its ability to share information and offer services to all users, and chatbots can address the shortcomings of mental health
professionals in any region.
Despite having access to health assistance, some individuals may delay seeking treatment due to various reasons, including underestimating the severity
of their emotions or fearing judgment from others. Chatbots are available 24/7, eliminating the need for appointments and making mental healthcare more
accessible to people. Chatbots are not designed to replace human health practitioners; instead, they assist them in providing better services. These
interactive and user-friendly tools, particularly the chatbot therapist, are more efficient and cost-effective in treating depression.
Therefore, self-mental healthcare techniques using chatbots have been developed to reduce stress and motivate users. The chatbot therapist is an advanced
method of therapy, consisting of different modules that work together to provide assistance to users. The chatbot analyzes the user's inputs to identify
their stress level and provides constructive thoughts and self-help techniques to help them cope. By utilizing Natural Language Processing technology,
the software is able to recognize the mood of the user and provide relevant content accordingly. Previous studies have utilized various technologies to
improve chatbot performance and achieve better results.
II. Related Work
[1] In 2019, A Mental Healthcare Chatbot System used SAT Counseling Method:
International Journal of Research Publication and Reviews, Vol 4, no 4, pp 1981-1993 April 2023 1982
In 2019, T. Kamita, A. Matsumoto, T. Munakata, and T. Inoue, along with T. Ito, conducted a study on the effectiveness of a chatbot for enhancing user
motivation and reducing stress in DIY or self-directed mental healthcare courses. However, the study had a limited sample size, and the assessment of
user motivation and stress reduction was based on a single use of the course. The authors began by digitizing the structured association technique
counseling method into a digital format and developed a independent mental health support system to achieve their objective. The system incorporated a
Virtual Reality (VR), which had previously received positive evaluations for stress relief in other studies conducted by Kamita et al..[1,12,13,14]
[2] In 2019, A Mental Health Care Chatbot using CBT Method:
The automated bot offers assistance to individuals dealing with mental health issues, while adhering to certain minimum standards. These standards
include respecting the privacy of users, relying on evidence-based practices, and ensuring their safety. To get started, the user is asked to complete a brief
survey, and based on their responses, a report is generated to assess their progress. While analyzing the report may take some time, it is an important step
in providing effective support to those in need.[2]
[3][4]AI for Chatbots in Mental Health-2019:
One approach to developing chatbots for mental health support is to ensure that the system aligns with the recommendations of healthcare
professionals and is cognizant of the treatment goals outlined in therapeutic sessions. However, a major hurdle in this process is accurately identifying
emergency situations and responding appropriately once an emergency has been detected. This presents a significant challenge that must be overcome in
order to provide effective and safe support to those in need. [3, 4]
[5] In 2018, Review on Analysis of Chatbot in Mental Healthcare :
The purpose of this chatbot is to offer mental support to students who experience varying levels of stress, which may potentially trigger depression.
However, further research is necessary to explore different aspects of its effectiveness. [5]
[6] M. Househ, A Review of Their Self-Care Features:
The prevalence of mental health disorders such as anxiety, depression, nervousness, and melancholy has surged to an all-time high globally, affecting
over 264 million individuals of all ages with depression alone. To locate existing chatbot applications designed for managing anxiety and depression, a
thorough search was carried out in both the App Store and Google Play Store, following the PRISMA protocol, which is a standardized reporting guideline
for systematic reviews and meta-analyses. The apps' eligibility was evaluated based on predetermined criteria by two reviewers, and meta-data regarding
the chatbots and their attributes were extracted from the descriptions and post-installation. The quality of the information was assessed based on adherence
to the mHONcode principles.. [1, 6, 15]
[7] Depression Detection and Emotion Recognition and using Deep Learning-2020:
The Recognition and Detection Chatbot is developed using Recurrent Neural Network (RNN) for its encoding and decoding mechanism, which
allows for the extraction of key characteristics from sentences. This approach is utilized to implement a chatbot with the necessary features for effective
recognition and detection. [7]
[8] In 2022, Divya Madhu -Voice assist using Python in AI-based HealthCare Chatbot system:
Divya Madhu suggests that Artificial Intelligence (AI) can be utilized to predict potential illnesses and recommended list of possible treatments
after considering the symptoms that were given. With regular analysis, AI can even assist in the early detection of diseases before they become apparent.
However, the expenses associated with researching and developing the technology, along with government backing to ensure proper implementation,
must be taken into consideration. [8].
[9] In 2022, Hameed-Ullah Kazi-Voice assist using Python in AI-based HealthCare Chatbot system:
Hameed-Ullah Kazi suggested the creation of an open-source medical chatbot called Chatter Bean, specifically intended for use by medical students.
The chatbot uses AIML and can accurately interpret human language queries and translate them into SQL queries. However, generating and organizing
groupings based on the number of issues in each group can be a lengthy process.[9].
[10] Zhang, X.-D. A Matrix Algebra Approach, The development of a chatbot:
In the past, "chatbots" or Artificial Intelligence systems for human-machine interaction were widely used. To interact accurately with users, a chatbot
to be effective, it needs to possess the ability to comprehend the intent of the user, along with the context in which the interaction is taking place, and the
sentiment behind their messages. Machine Learning has played a crucial role in chatbot development, where the process of creating a chatbot involves
using a mathematical framework that is trained on data to make predictions or decisions without explicit programming. Recent research has focused on
utilizing Convolutional Neural Networks (CNNs) as classifiers, combined with specific tokenization tools. CNNs are a type of deep learning algorithm
with a hierarchical structure, where each layer learns from the previous one. Text is processed using a technique called "Word Embedding," which
converts it into numerical form. For this purpose, Word2Vec technique was employed, which offers a vast amount of pretrained data and enables users
to train their own datasets. Additionally, Word2vec technique enables obtaining vectors from other vectors using vector operations. This preprocess is
necessary to insert data into the CNN. Various CNN architectures were utilized in this study of research. [10, 11]
[11] In 2017, K. Oh, D. Lee, B. Ko and H. Choi, "A Chatbot for Psychiatric Counseling in Mental Healthcare Service Based on Emotional
Dialogue Analysis and Sentence Generation,":
Early studies have explored the use of chatbots to encourage users to seek psychiatric counseling, leading to positive changes in their drinking habits
through intervention approaches. However, these studies did not take into account the user's psychiatric to seek psychiatric counseling. Intervention
approaches used in such studies have led to positive changes in drinking habits. However, limitations in these studies include a lack of consideration of
the user's psychiatric status during conversations, inadequate monitoring, and ethical considerations in the intervention process. Therefore, we propose a
conversational service for psychiatric counseling that incorporates more accurate and continuous emotion recognition, providing better satisfaction for
users in need of mental health care. Our approach includes high-level natural language understanding (NLU) methodologies and multi-modal emotion
recognition to enable the sensitive observation of emotional changes over time. Additionally, we provide appropriate and suitable responses for clinical
psychiatric counseling. [22,27]
[12] Du Preez, S.J, Lall, Manoj, Sinha, S. “An intelligent web-based voice chatbot” in 2009:
This article describes the development of an intelligent chatbot that utilizes voice recognition technology. The chatbot is designed to function as a
web service and a framework is proposed to support its operation. Through controlling the communication structure between the server and client
platforms, the web service enables clients to access it from any platform. An interface is generated to facilitate XML processing and ensure extensibility,
thus improving the service's lifespan. The chatbot employs an artificial brain to provide customized user responses that are consistent with the desired
character. In cases where the chatbot is unable to comprehend a question, the system employs an online intelligent research assistant, which is a third-
party expert system, to process the inquiry and archive the response, thereby enhancing the chatbot's artificial intelligence for future responses. [23, 27]
[13] In Ferry Wahyu Wibowo, "Chatbot Using a Knowledge in Database: Human-to- Machine Conversation Modeling", Intelligent Systems
Modelling and Simulation :
A chatbot, also known as a chatterbot, the design is used to facilitate a conversation between human and machines responses. The machine is
programmed with the ability to comprehend sentences and independently make decisions in order to respond to questions. The response mechanism
involves matching the user's input sentence with a reference sentence and determining their similarity score. To calculate the similarity score, bigram is
used to divide the input sentence into two-letter segments. The chatbot's knowledge base is stored in a database, while its core and interface access this
information using a RDBMS. The database is a storage unit for the chatbot's knowledge, while an interpreter employs stored program of functions and
procedures sets to meet pattern-matching as per requirements. The interface is a defines as a standalone unit that is built using programming languages
like Pascal and Java. [24, 27]
[14] In 2017, Dahiya, Menal- A Tool of Conversation Chatbot and C.P. Shabariram, V. Srinath, C.S. Indhuja, Vidhya-Chatbot Application
Using Expert System:
Communication through speech and text is a crucial aspect of human interaction. According to an article," adults now days spend over 8-9 hours a day
on computer or mobile display, leading to increased use of web applications such as WhatsApp, Facebook, Twitter and other Apps for speech and text-
based conversations. This paper focuses on designing a chatbot application for educational purposes that can answer user-provided questions. To
developed the system with ensembled learning method and random forest, was employed along with extracted features from dataset. The validation of a
system produced an average F-measure score of 0.870 for various K-values under random forest, demonstrating the effectiveness of the proposed chatbot.
The system has been deployed as a Telegram bot for convenient access. [25, 26, 27]
III. Methodology Used
There are various studies have been conducted on mental health chatbots and control systems, categorized based on the implemented methodologies and
technology utilized in the system. The exchange of messages is a crucial aspect of providing a chatbot for problem-solving and improving system
efficiency. The proposed systems are classified based on the tools and implementation of the architecture. This paper provides a brief description of the
various methodologies implemented in mental health chatbot and control systems.
A Chatbot for Mental Health Care using SAT Counseling Method
One of the biggest challenges in the conventional self-support mentalcare method is user withdrawal. To overcome this, the use of frequent smile face
images and course repetition is necessary to improves the mental health improvement effect. To address this problem, a Chatbot course was developed
using conversational guidance. The self-mental care system was developed using LINE, a widely used chat tool in Japan. During emotional stabilization
therapy, a counselor observes the effects of the conversation on the client, such as facial expression, body language, and other gestures, and may suggest
eye closure if the effect is deemed insufficient. With the digital-SAT approach, individuals can access self-guided intellectual healthcare based on the
SAT technique without requiring a counselor's education. This method simplifies the questionnaire by dividing the unique questions for emotional
stabilization therapy into individual queries. The Chatbot route was developed using the digital-SAT technique.
System Configuration
The system configuration is illustrated in Figure 1, which depicts the data processing flow through the system. In this system, the chatbot service is
offered through the LINE application, which is one of the most widely used social networking services (SNS) in Japan. The chatbot applications server
is hosted on a commercial of HEROKU server, and the LINE server is provided by LINE. To connect the servers, the messaging API program offered
by LINE is utilized. This program serves as an application programming interface, enabling data exchange between the sending and receiving server of
LINE.
Flowchart of system configuration
1) The system configuration for user and LINE server interaction involves the user connecting to the LINE application, which is a social
networking service used to send requests. The LINE server, which comprises of a reception component for transferring messages to the
prepared server, and a sending server for replying responses to the user, is responsible for facilitating the interaction.
2) To enable developers to build, run, and manage applications entirely in the cloud, without having to worry about infrastructure setup, scaling,
and maintenance, LINE provides the Heroku platform as a service (PaaS). Heroku is a cloud-based solution that supports various
programming languages like Ruby, Node.js, Python, PHP, and, more. It also includes a MariaDB database for storing the requests and
responses from the web server, corresponding to SSL. It is essential to note that "Heroku" is a registered trademark of Heroku server.
A Chatbot System For Mental Health Care using CBT Method
Cognitive behavioral therapy (CBT) is a method, aims to replace negative and unhelpful behaviors, thoughts, and emotions with positive ones.
Similar principles have been identified in various traditions.This has influenced modern CBT strategies, these techniques aim to identify cognitive
distortions that can lead to the development of depression and anxiety. CBT was developed by Aaron Beck in cognitive therapy and later evolved into
CBT, while Albert Ellis invented rational emotive behavior therapy (REBT) and introduced the ABC method. The ABC model involves the situation (A),
our interpretation of it (B), and our reaction or behavior (C). The final goal of this CBT is to helps to people respond correctly, rather than replacing
their thoughts or feelings.
During CBT, the therapist works on the client's thoughts, feelings, and behaviors and how they intersect. It examines how their thoughts contribute to
difficult emotions and how their thoughts may be helping to maintain negative feelings. The therapist helps the client to consider situations, state of
people, or themselves from a different attitudes in order to feel better. CBT is an evidence-based approach that is commonly used to treat a range of
problems and issues. It is a collaborative process, where the therapist and client work together to analyze the issues being faced. Homework assignments
are often given to clients to practice the skills and techniques learned in therapy outside of sessions. CBT is the combination of behavioral and cognitive
psychology, where emotions, thoughts, and behaviors are all interconnected.
Flow of CBT mechanism
Cognitive Behavioral Therapy (CBT) is a highly structured approach to therapy that involves identifying goals that both the client and therapist
agree to work on together. A distinctive feature of CBT is that the therapist actively seeks and welcomes feedback on how the therapy is progressing.
This reflects the collaborative nature of the therapy, where the therapist is not seen as having greater expertise than the client, but rather brings specialized
knowledge and skills to help the client address their issues.
Divya Madhu and Hameed-Ullah Kazi, Artificial Intelligence-based HealthCare Chatbot system with voice assist using Python,2022
In their research describe a chatbot system designed to provide basic health information to users. The chatbot employs artificial intelligence and
machine learning techniques to interpret natural language and respond to user inquiries. To use the chatbot, users must first register on the website. If the
chatbot does not have an answer in its database, an expert system is utilized to provide In their research paper titled "Artificial Intelligence-based
HealthCare Chatbot system with voice assist using Python," Divya Madhu and Hameed-Ullah Kazi describe a chatbot system designed to provide basic
health information to users. The chatbot employs AI and ML techniques to interpret Natural language and respond to clients inquiries. To use the chatbot,
users must first register on the website. If the chatbot does not have an answer in its database, an expert system is utilized to provide a response. Domain
experts can also register and provide specific information for the chatbot. The data generated by the chatbot is stored in a database as a sample template.
To interact with the chatbot, the user types their inquiry into the user interface, which is then received by the program. The chatbot undergoes pre-
processing procedures, including tokenization, stop word removal, and, features extraction based on N-grams, TF-IDF, and cosine similarity. The chatbot
retrieves solution to queries from the knowledge database. The text is tokenized by splitting the words or phrases. Punctuation is removed, and stop words
are eliminated to reduces the time and complexity of it . Feature extraction involves ranking the attributes of a document, and the frequency of each
keyword is extracted using TF-IDF. This process is used to compute the weight of each phrase in the sentence, resulting in increased accuracy and speed
of document processing.
Flowchart for User and System Interaction

Flowchart for voice assist
Emotion Recognition and Depression Detection using Deep Learning
The use of Convolutional Neural Networks (CNNs) is a popular method in Deep Learning for image analysis and recognition tasks. A CNN is capable
of assigning significance to different objects and features in an input visual format
through weights. This process requires less pre-processing compared to other classification algorithms. The advantage of using CNNs for image
recognition tasks is that they are able to capture spatial aspects of the input image due to the numerous filters employed in the algorithm.
1. Dataset of Facial Expression :
Numerous open access facial expression datasets are available in literature. For our study, we utilized a facial expression dataset sourced from Kaggle.
The dataset consists of grayscale images of faces with 48x48 pixel resolution. The training set contains a total of 28,709 samples, encompassing different
emotions – happy, surprised, and, angry etc.
2. Image Pre-processing:
Image preprocessing was performed to prepare the facial expression dataset for analysis. The Haar-Cascade library was utilized to detect the face
boundaries in the images, and these regions of interest were cropped and saved. Additionally, the images were converted into grayscale format and fed
into the neural network. This preprocessing step helped to reduce the complexity of the neural networks and improve their computational efficiency.
3. Convolutional neural community architecture:
Convolutional neural networks are used in this study to extract features from facial expression data. The proposed system CNN architecture consists with
three levels of convolutional layers with 'relu' is a activation function, max-pooling layers, and three fully connected layers with 'relu' and 'softmax'
activation functions. The input to the CNN is the rectangular region containing facial expressions. After passing through the convolutional and max-
pooling layers, the output is fed into the fully connected layers for prediction of the seven different emotional states. Prior to this, the face circumference
is detected using the Haar Cascade library, and the images are converted to grayscale to reduce the complexity of the neural networks.
CNN Model Architecture- The layers used in the construction of the CNN Model
4. Network training:
The network training was performed in Python using Keras with a TensorFlow backend. The model was trained over the course of 50 epochs.
5. Real time testing:
To test the effectiveness of the CNN model, real-time testing was conducted. The initial step was to detect human faces using the Haar Cascade library
via the computer camera. The detected images were then forwarded to the trained model, and the predicted facial expression categories were retrieved.
A review of mobile chatbot apps for anxiety and depression and their self-care features:
It appears that you have provided a comprehensive overview of the methodology used in the evaluation of mobile chatbot apps for stress and depression
and their self-support and care features. The study appears to have followed a rigorous search strategy, eligibility for this , app selection, information
extraction, and data process. The use of the PRISMA guidelines and the adaptation of the mHONcode principles demonstrate a commitment to
transparency, replicability, and quality assurance. The addition of the General Information and Technological Features categories shows the study's
relevance to current needs and trends. Overall, the methodology seems sound, and the study provides a useful reference for future research and app
development in this area., easy to comply with and met to aims we desired to acquire from our overview of study.
Principal Model with new version model
The development of a chatbot using CNN method
Developing a chatbot requires selecting an appropriate methodology for training it to understand user input in the text of the conversation. The
flowchart in Fig 7 outlines the necessary steps to process. First, the data must undergo a preprocess procedure, which involves separating the raw data
input into single word, known as tokens. The tokenized strings go through input text for normalization, which can be executed through stemming or
lemmatization techniques. Stemming removes word affixes to simplify the word, while emmatization looks up each token in a dictionary and to determine
the root lemma, making it is slower but it give more accurate. This paper utilizes lemmatization.
Since algorithms are difficult to apply to text, Word Embedding is used to convert text into numbers. Word2Vec and an extension called Doc2Vec are
used in this paper, with Doc2Vec taking entire sentences as input. Word2Vec offers pre-trained data online, which can be supplemented with user data.
Word Embedding is followed by entity classification, which involves identifying information items such as names, numeric expressions, and location
names. Intent classification follows, in which the chatbot attempts to understand the user's intention before providing a response or solution to the problem.
Flowchart depicts the process of how a chatbot engine receives an input and generates output
IV. IMPLEMENTATION
The proposed system utilizes the Sequential Algorithm for developing a mental healthcare chatbot that offers various features such as text and
voice-based input and output. Users can ask questions related to mental health issues, and the chatbot provides solutions as well as diagnoses diseases
based on symptoms. Additionally, it suggests doctors for specific ailments, acting as an AI healthcare consultant. The Sequential Algorithm is used for
accurate prediction and high accuracy.
The dialogue manager manages text and voice-based input from users, collects and records queries, and provides responses. The chatbot selects a
response by identifying specific keywords and patterns, but it does not comprehend the user's input or its own output. Responses are collected from the
user's text message and then classified using the Sequential methodology, which involves applying filters to recognize input in the correct format. The
result is an accurate output of text or voice related to mental health problems, which makes communication easy and saves time. The Sequential Algorithm
is applicable to various classification tasks in natural language processing, and it involves sliding a window over a larger input data with an emphasis on
a subset of the input matrix of data. It is critical to ensure that the data is in the correct dimensions for the learning algorithms.
Implementation Design for Chatbot.
Flowchart for chatbot processes sing input data and gives a response.
Output Screen for Implementation.
V. Results and Discussions
Table 1 provides a detailed explanation of the techniques utilized in the development of Chatbot for Mental HealthCare, including machine learning, deep
learning, CNN, SAT Counseling Method, CBT Method, and various other techniques. The table presents the advantages and disadvantages of each
technique and also highlights future directions for further research and development.
Various techniques in Chatbot system
S.No Techniques Used Advantages Disadvantages Future Directions

1. In 2019, A Mental Healthcare The chatbot system provides The study's sample size is Advanced algorithms be
Chatbot System used SAT users with increased limited, and the impact of the explored and enhancement in
Counseling Method[1] motivation and support to mental healthcare course on the future.
manage and reduce their user motivational and stress
stress levels. level reduction.
2. In 2019,A Mental Health Care The chatbot proved to be It requires a survey and In future increases the chatbot
Chatbot using CBT effective in helping clients creates a report of based on therapy.
Method[2] identified the emotions, and performance of individual
understand negative person. It takes time to
emotions, thereby simplifying analysis the report
the process for them.
3. AI for Chatbots in Mental It detects the emergency The most significant obstacle In the future improves the
Health-2019[3,4] situation. is the accurate identification accuracy and effective
of emergency situations. detecting process in chatbot.
4. In 2018, Review on Analysis It helps to references of all It requires additional review In the future, the system can
of Chatbot in Mental review details for developing. information. be improved by incorporating
Healthcare[5] additional review information
5. M. Househ, A Review of It has the capability to It is difficult to use and To gain greater trust and
Their Self-Care Features [1, 6, increase the maintain, because of high confidence in chatbot
15] quality based on our technology, it is important for
capacity of mental health care assessment according to the future developers to prioritize
and providing benefits. mHON code principles. evidence-based interventions.
6. Depression Detection and The emotion recognition and The current working desktop Chatbots is rising for a few
Emotion Recognition and depression analyzer which is application is local. No access years and
using Deep Learning-2020[7] implemented in desktop to everyone. have already gained
application based chatbot widespread acceptance with
the growth of developing
technologies.
7. In 2022, Divya Madhu -Voice It offers a high-quality In a positive dataset possible In future work involves to
assist using Python in AI- responses in a short amount of to get the information other improves the accuracy.
based HealthCare Chatbot time and more efficient. redundancy.
system[8].
8. Hameed-Ullah Kazi-Voice It offers a more-quality to In a given dataset possible to Our future work involves to
assist using Python in AI- responses in a less amount of get the details of repetition explored the other methods
based HealthCare Chatbot time. problem. and technics.
system[9].
9. Zhang, X.D. and A Matrix It has the best accuracy, It uses Server for managing In future, plan to increase
Algebra Approach, The fastest training time and the large data-set and cost is high. accuracy and keeping the
development of a chatbot least losses. training time within limits.
[10, 11]
Chatbot system implementation details
S.No Techniques Used Tools Used Dataset Layer

1. In 2019, A Mental Healthcare Chatbot LINE server, VR, and Real-time Network layer
System used SAT Counseling Method[1] HMD Dataset
2. In 2019,A Mental Health Care Chatbot CBT tool Real-time Convolutional Layer
using CBT Method[2] Dataset
3. AI for Chatbots in Mental Health- DMHIs Real-time Network and
2019[3,4] Dataset Transport Layer.
4. In 2018, Review on Analysis of Chatbot Reviews on chatbot Real-time Application Layer

in Mental Healthcare[5] Dataset and PRISMA
protocal
5. M. Househ, A Review of Their Self-Care mHON code, Android Real-time Application Layer
Features [1, 6, 15] and iOS systems Dataset.
6. Depression Detection and Emotion GUI tools Real-time Internet Layer
Recognition and using Deep Learning- Database
2020[7]
7. In 2022, Divya Madhu -Voice assist using TFIDF and N-grams Training and Network Layer and
Python in AI-based HealthCare Chatbot test data sets Internet Layer
system[8].
8. Hameed-Ullah Kazi-Voice assist using TFIDF and N-grams Training and Network Layer and
Python in AI-based HealthCare Chatbot test data sets Internet Layer
system[9].
9. Zhang, X.D. and A Matrix Algebra AlexNet, LeNet5, Real-time Application Layer
Approach, The development of a chatbot ResNet and VGGNet Dataset and and Convolutional
[10, 11] CNN architectures were Training Layer
utilised Dataset
Chatbot system using techniques and its accuracy
S.No Techniques Used Input Output Parameters Accuracy

Parameters
1. In 2019, A Mental Healthcare Chatbot Images Digital Output 80%
System used SAT Counseling Method[1]
2. In 2019,A Mental Health Care Chatbot using Virtual face-to face Voice Output 91%
CBT Method[2]
3. AI for Chatbots in Mental Health-2019[3,4] Text and Images Relatively good accuracy and 94%
speed. And text and Voice
output.
4. In 2018, Review on Analysis of Chatbot in Text Response from output screen 79%
Mental Healthcare[5]
5. M. Househ, A Review of Their Self-Care Text Content and Discussions of results 94%
Features [1, 6, 15] symptoms
6. Depression Detection and Emotion Virtual data Tags Displays on Screen 60%
Recognition and using Deep Learning-
2020[7]
7. In 2022, Divya Madhu -Voice assist using Text format Text and Voice format 89.5%
Python in AI-based HealthCare Chatbot
system[8].
8. Hameed-Ullah Kazi-Voice assist using Text format Text and Voice format 90%
Python in AI-based HealthCare Chatbot
system[9].
9. Zhang, X.D. and A Matrix Algebra Text, Content, and Softmax 92.3%
Approach, The development of a chatbot Intent
[10, 11]
VI. Conclusion
The chatbot provides a user-friendly platform that is easily accessible to anyone with an internet connection. It adheres to the minimum standards of
respecting user privacy, being evidence-based, and ensuring user safety. Despite its potential benefits in treating depression, many people are still unaware
of this technology. The mental healthcare chatbot has the potential to encourage users to open up and discuss their problems. This project utilizes Python
and employs machine learning techniques such as Sequential Model, ML, and NLP to improve the chatbot's accuracy. In the future, this desktop
application-based chatbot could be developed further to incorporate video consultancy with a doctor for users who are identified as suffering from
depression and stress.
References
[1]. T. Kamita ,T. Ito ,A. Matsumoto, T. Munakata, T. Inoue , A Chatbot System for Mental Healthcare Based on SAT Counseling Method, 2019.
[2]. B. Martin , In-depth: cognitive behavioral therapy, 2019.
[3]. A.Todorova, G.Todorova, Milen ,Digital technologies for art therapy practices used in healthcare, 2019, Vol. 13 Issue 1, p43-47. 5p.
[4]. S. Price, New chatbot therapist to help combat depression unveiled 2019.
[5]. K.Matthews, Therapy chatbots are Transforming psuchology, 2018.
[6]. Organisation, W.H. Depression – Key facts. [cited 2020 October 13th]; Available from: https://www.who.int/en/news-room/fact-
sheets/detail/depression.
[7]. A Survey Paper on Chatbots, Aafiya Shaikh, Dipti More,Ruchika Puttoo, Sayli Shrivastav, Swati Shinde.
[8]. https://www.wordstream.com/blog/ws/2017/10/04/chatbots
[9]. https://www.irjet.net/archives/V6/i4/IRJETV6I4383
[10]. Zhang, X.-D. A Matrix Algebra Approach to Artificial Intelligence; Springer, 2020.
[11]. Manaswi, N.K.; Manaswi, N.K.; John, S. Deep Learning with Applications Using Python; Springer, 2018;
[12]. T. Ito, T. Kamita, A. Matsumoto, T. Munakata, and T. Inoue, “Virtual reality course based on the SAT counseling method for self-guided
mental healthcare,” in Proceedings of 2018 IEEE International Conference on healthcare Informatics (ICHI), pp. 299–306, New York City,
NY, USA, June 2018. View at: Google Scholar
[13]. T. Ito et al., “A self-guided mental healthcare digital content for smartphone VR based on the counseling technique SAT method,” IPSJ SIG
Technical Report, vol. 2018, no. 37, pp. 1–8, 2018.View at: Google Scholar
[14]. T. Munakata, SAT Therapy, Kanekoshobo, Tokyo, Japan, 2006.
[15]. A Survey Paper on Chatbots, Aafiya Shaikh, Dipti More,Ruchika Puttoo, Sayli Shrivastav, Swati Shinde.
[16]. Statista, Number of mHealth Apps Available in the Apple Apple Store from 1st Quarter 2015 to 2nd Quarter 2020, (2020) (accessed October
2020), https://www. statista.com/statistics/779910/health-apps-available-ios-worldwide/.
[17]. Statista, Number of mHealth Apps Available in the Google Play Store from 1st Quarter 2015 to 2nd Quarter 2020, (2020)
(accessed October 2020), https://www.statista.com/statistics/779919/health-apps-available-google- play-worldwide/.
[18]. AC Tricco, E Lillie, W Zarin, KK O’Brien, H Colquhoun, D Levac, et al., PRISMA EXten- sion for Scoping Reviews (PRISMA-ScR):
Checklist and EXplanation, Ann Intern Med 169 (7) (2018) 467–473 Oct 02[doi: 10.7326/M18-0850] [Medline: 30178033].
[19]. N. Drissi, S. Ouhbi, M.A.J. Idrissi, M Ghogho, An analysis on self-management.
[20]. Bengfort, B.; Bilbro, R.; Ojeda, T. Applied Text Analysis with Python: Enabling Language- Aware Data Products with Machine Learning;
O’Reilly Media, Inc., 2018; ISBN 978-1-4919- 6299-2.
[21]. Nadeau, D.; Sekine, S. A Survey of Named Entity Recognition and Classification. Lingvisticae Investigationes 2007, 30, 3–26.
[22]. K. Oh, D. Lee, B. Ko and H. Choi, "A Chatbot for Psychiatric Counseling in Mental Healthcare Service Based on Emotional Dialogue Analysis
and Sentence Generation," 2017 18th IEEE International Conference on Mobile Data Management (MDM), Daejeon, 2017, pp. 371-375. doi:
10.1109/MDM.2017.64
[23]. Du Preez, S.J. & Lall, Manoj & Sinha, S. (2009). An intelligent webbased voice chat bot. 386 - 391.10.1109/EURCON.2009.5167660
[24]. Bayu Setiaji, Ferry Wahyu Wibowo, "Chatbot Using a Knowledge in Database: Human-to- Machine Conversation Modeling", Intelligent
Systems Modelling and Simulation (ISMS) 2016 7th International
[25]. Dahiya, Menal. (2017). A Tool of Conversation: Chatbot. INTERNATIONAL JOURNAL OF COMPUTER SCIENCES AND
ENGINEERING. 5. 158-161.2017.
[26]. C.P. Shabariram, V. Srinath, C.S. Indhuja, Vidhya (2017). Ratatta: Chatbot Application Using Expert System, International Journal of
Advanced Research in Computer Science and Software Engineering,2017
[27]. https://ieeexplore.ieee.org/document/7962482

Chat D

Uploaded by

Chat D

Uploaded by

International Journal of Research Publication and Reviews, Vol 4, no 4, pp 1981-1993, April 2023

International Journal of Research Publication and Reviews

Chatbot for Mental Health Care

R. Shanmukha Priya1, Ch. Navyamani2, M. MohammedAeliya3, Dr. V Ravi Kishore4

II. Related Work

[2] In 2019, A Mental Health Care Chatbot using CBT Method:

[3][4]AI for Chatbots in Mental Health-2019:

[5] In 2018, Review on Analysis of Chatbot in Mental Healthcare :

[6] M. Househ, A Review of Their Self-Care Features:

[10] Zhang, X.-D. A Matrix Algebra Approach, The development of a chatbot:

III. Methodology Used

A Chatbot for Mental Health Care using SAT Counseling Method

Flowchart of system configuration

A Chatbot System For Mental Health Care using CBT Method

Flow of CBT mechanism

Flowchart for User and System Interaction

Flowchart for voice assist

Emotion Recognition and Depression Detection using Deep Learning

1. Dataset of Facial Expression :

3. Convolutional neural community architecture:

5. Real time testing:

Principal Model with new version model

The development of a chatbot using CNN method

Implementation Design for Chatbot.

Output Screen for Implementation.

V. Results and Discussions

Various techniques in Chatbot system

S.No Techniques Used Advantages Disadvantages Future Directions

Chatbot system implementation details

S.No Techniques Used Tools Used Dataset Layer

4. In 2018, Review on Analysis of Chatbot Reviews on chatbot Real-time Application Layer

Chatbot system using techniques and its accuracy

S.No Techniques Used Input Output Parameters Accuracy

[2]. B. Martin , In-depth: cognitive behavioral therapy, 2019.

[5]. K.Matthews, Therapy chatbots are Transforming psuchology, 2018.

[14]. T. Munakata, SAT Therapy, Kanekoshobo, Tokyo, Japan, 2006.

[19]. N. Drissi, S. Ouhbi, M.A.J. Idrissi, M Ghogho, An analysis on self-management.

You might also like