IEEE Template 4 (1430)

You might also like

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 5

First Author#1, Second Author*2, Third Author#3

Department of Information Technology, Rajasthan


Technical University,India
1] First Author Name:-Devesh Sharma College Name:-Jaipur
Engineering College and Reseach Centre, Jaipur Address:- hri Ram
Ki Nangal,Via Vatika,Tonk Road, Jaipur, Rajasthan 302022 India
Email:-Deveshsharma.it23@gmail.com
Second Company
*

Address Including Country Name


2
second.author@second.com
approaches being the classification of
Abstract— Music is an integral part of our emotions that are existing are depicted as
lives. However, since the social media knowledge-based, statistical and hybrid
platforms like TikTok and Instagram have techniques. However there relies several
a huge influence on the music charts difficulties while retrieving the music
worldwide, users are exposed solely to information such as querying by singing,
mainstream music, therefore the genre classification, etc. The most
recommendations on music streaming possible implementation is by producing
platforms are not very personalized. The music suggestion that is based on the
proposed research work develops a content. The multiple efforts to detect and
personalized system, where the user's describe the feeling, to classify based on
current emotion is analyzed with the help features are all wanted to obtain
of the chatbot. The chatbot identifies the outstanding music recommendation
user's sentiment by asking some general system. For describing a music taxonomy,
questions. Based on the input provided by
a feeling descriptor is found to be useful.
the user, current emotion or mood is
The proposed research work develops a
analyzed by the chatbot and it will
personalized system, where the user's
generate the playlist. The proposed
current emotion is analyzed with the help
recommendation system utilizes the APIs
for the playlist generation and of the chatbot. The chatbot identifies the
recommendation user's sentiment by asking some general
questions. Based on the input provided by
the user, current emotion or mood is
Keywords— Interactive Chatbot,
analyzed by the chatbot and it will
Application Program Interface, Interactive
generate the playlist. The proposed
Systems, Recommender Systems, Playlist
recommendation system utilizes the APIs
generation, IBM Tone Analyzer API,
for the playlist generation and
Last.fm API, CakeChat server. Playlist
generation, IBM Tone Analyzer API, recommendation. recommendation, also
Last.fm API, CakeChat server other similar user opinions are
considered. The design of the proposed
I. INTRODUCTION system, its implementation along with its
evaluation procedure is also discussed. In
Everyday, each and every person
order to relate emotions to movies, the
undergoes lot of troubles and the reliever
users have to answer a questionnaire
of all the stress that are encountered is
about what movies or which categories of
Music. If it is so, the vital part of hearing
movies they liked to watch according to
the song has to be in a facilitated way,
each emotion. Furthermore, the system
that is player able to play the song in
captures user emotions by asking them to
accordance to the person’s mood. The
use 3 colours to decorate their avatar
paper proposes such a player and hence
named Emotion based music player.
Literature Survey: - Few of the key
Recognizing the human emotions is
features emphasized by the papers that
considered to be a global consistency but
have been surveyed are:
depict variability among the humans on
the basis of their abilities. The different
1. Nikhil et al. [1] use algorithms and
technologies which include Haar cascade, to decorate their avatar
Canny edge, Blob detection for the
process of emotion detection. The system 3. Jae Sik Lee et al. [3] have used the
captures pictures of the user and concept of context reasoning wherein the
according to that mood gets detected. context data is utilized to understand the
Inputs like face and emotions are taken user’s situation. They propose a music
from the picture, and the system also recommendation system that comprises
provides a chat box to give responses. the ability of context reasoning in this
The proposed system in the paper paper. Their proposed system contains
presents a new approach for building modules such as Intention Module, Mood
desktop application for chat bot using text Module and Recommendation Module
and gestures. The system is able to make each of which provide a unique
a conversation through the chatting functionality to the system and play a
application. The system will send some vital role for the system’s performance as
links, web pages or information a whole. Context reasoning is done by the
depending on the response from the user. Intention Module with the help of
The system detects smile and stress. environmental context data and concludes
When a smile is detected by the system, whether the user is interested in listening
jokes pop-ups will be shown on the to music or not. Next, the type of music
screen, and when stress is detected, that is deemed to be most appropriate to
inspirational quotes pop-ups will be the user’s context is determined by the
shown on the screen. Also, happy songs Mood Module. Lastly, the music is
are played when a smile is detected. And recommended to the user by the
similarly, inspirational songs are played Recommendation Module
when stress is detected. 2. Ai Thanh Ho et
al. in their paper 4. Renuka R. Londhe et al. in [4]have
studied the concept of recognizing facial
[2], introduce an Emotion-based Movie expressions by taking into account the
Recommender System (E-MRS) which is various properties that are associated with
intended to solve the problem that the a person’s face. Whenever there is a
conventional system of user profile does change in the facial expression, changes
not take into consideration how important can be noticed in the curvatures on the
user’s emotions are and how they affect face as well as features of the face such as
user’s choices, which the recommender nose, lips, eyebrows and mouth area. And
systems are unable to understand and accordingly, there will be changes in the
capture the constantly changing intensity of the corresponding pixels of
preferences of user. According to the the images. These features are then
paper, the objective of EMRS is to give classified into six expressions which
the users a list of suggestions that are include anger, disgust, fear, happy, sad
customized using a combination of and surprise with the help of artificial
collaborative filtering and content-based neural network. The Scaled Conjugate
techniques. Here the user’s emotions as Gradient back-propagation algorithm is
well as his preferences are taken into used to train and test the twolayered feed
account when providing a forward neural network. They acquired a
recommendation, also other similar user 92.2 % recognition rate. Here, they have
opinions are considered. The design of made use of the JAFFE database which
the proposed system, its implementation consists of seven expressions for analysis
along with its evaluation procedure is also through the computer. 5. Dolly Reney et
discussed. In order to relate emotions to al. in their paper
movies, the users have to answer a
questionnaire about what movies or [5] address the importance of face and
which categories of movies they liked to emotion identification in the field of
watch according to each emotion. security and how it helps give solutions to
Furthermore, the system captures user the different challenges faced. Database
emotions by asking them to use 3 colours plays a major role when comparing the
facial attributes and sound Mel frequency expected rating. Find p and q such it
components, when it comes to whichever minimizes the following:
face and emotion identification system.
The database is created for which facial *Nearest Neighborhood Model: Nearest
characteristics are computed and these are Neighborhood model involves collecting
then stored in the database. Various data from numerous clients at that point
algorithms are used in order to analyze making forecasts according to the
the face and emotion with the help of the similarity measures between users and
aforementioned database. The between items.
implementation of the process of This might be grouped into client-based
recognizing the person’s face and the and item based models. In item-based
emotion being expressed by him uses an model, it's expected that songs that are
effective method for the creation of a frequently listened together by certain
database comprising the facial clients will in general be indistinguishable
expressions and emotion. They have used and are bound to be listened together in
the Viola-Jones algorithm for the face future additionally by another client.
identification process and the face and According to client based similarity
emotion identification is evaluated by the model, clients who have similar listening
KNN classifier. histories, i.e., have listened in to similar
songs inside the past will in general have
Architecture:- comparative interests and may most likely
hear similar songs in future as well.

Advantages:-

 Chat-bots have 24/7 Availability: Chat-


bots are available to solve customer
problems 24/7 whether it is day or night!
They don’t need to sleep after all! This is
much more difficult to achieve using
human customer service as it would
require rotating teams that would be more
complicated to manage as well. This
means that chat-bots can answer customer
Algorithms:- queries whenever customers have queries
which help in increasing customer loyalty.
* SVD (Singular Value Decomposition) If companies don’t address customer
Algorithm: problems as soon as possible, then the
SVD could also be a matrix factorization customers may switch to another
technique that is typically wont to company that provides better customer
diminish the quantity of feature of a service.
dataset by reducing the matrix from N  Chat-bots cause an Increase in Sales:
space to K space where K < N. Customers can get instant support and
solutions for their problems using chat-
For the point of the guidance framework bots while using agents takes a longer
be that as it may, we are just interested time. When customers get such good
about by the matrix factorization part service, chances are high that they will be
keeping same dimensional. The matrix more loyal to the company which will
factorization is completed on the useritem drive up sales. Companies can also use
ratings matrix built. chat-bots to market their latest products
directly to their customers. When
Each item is often represented with a q customers get recommendations for a
vector. Similarly, each client are often product while chatting to a chat-bot, it is
represented by a p vector such the inner much more personal than a mail or a
product of these 2 vectors is that the random ad and they are more likely to
buy the product. So chat-bots can increase inquiries forever! The company products
sales in this manner as well. change with time and more advanced
 Chat-bots provide Long-term Financial Natural Language Processing capabilities
Savings: Companies that have human are also developed with time. All of these
customer support teams have to pay a changes need to be programmed into the
large amount of money as salaries to their chat-bot so that it has the most up to date
employees. And if the company is large information. In addition to that, chat-bots
then their expenditure will be equally also need to be periodically analyzed so
large as well! Here, a chat-bot can reduce that the most common questions that
this expenditure so that the company has customers have can be identified and then
long-term financial savings. While it is their answers updated for future
true that initially, a company has to invest customers.
a lot of time and money in creating a
chat-bot, after some time the chat-bot will
be able to answer all the basic customer
queries that would have taken a lot of
time for humans. In addition to that, the
chat-bot can give instantaneous replies
which saves time. And time is money in
business!

Disadvantages:-  Chat-bots can only


handle basic questions: Chatbots are still
a basic Artificial Intelligence technology
and so they can only answer the basic
questions of customers and provide
general information that is already Conclusion:- We have presented a survey
available to them. They cannot solve and methodology for building the chatbot
complicated queries or answer out of song recommender system. To perform
script questions and companies need to this, we first identified various
have human customer service employees approaches for building a chatbot known
that can manage these for them. However, to date. We then evaluated the considered
this is changing with time and currently, algorithms which are useful in building of
more and more advanced chat-bots are our system in terms of their ability to
entering the market. work on the recommendation process of
the system. We also gathered all the
 Chat-bots are difficult to create: Chat- requirements needed for building our
bots are created using Natural Language system and studied the overall process
Processing which is extremely popular for involved in chatbot's working. Lastly we
customer support applications. Natural summarized the deployment requirements
Language Processing is a part of Machine of our system. On the conclusion note our
Learning which can be used to interact ‘Chatbot Song Recommender System’ is
with the users in textual form and solve used to facilitate the use by physically
their queries. However, this requires challenged people to automate and give
complex programming and is not easy for them better music player experience. The
companies. This becomes especially application solves the basic needs of
difficult if companies have to create chat- music listeners without troubling them as
bots from scratch and that is why many existing applications do.
online platforms help companies to build
and manage chat-bots easily. Reference:-

 Chat-bots require constant [1] Anagha S.Dhavalikar, and Dr. R. K.


maintenance: Companies cannot just Kulkarni, “Face Detection and Facial
create a chat-bot and then leave it hoping Expression Recognition System,”
that it will correctly answer customer Interntional Conference on Electronics
and Communication System (ICECS - Information Systems, 2013.
2014), 2014. [3] Arto Lehtiniemi and Jukka Holm,
[2] Yong-Hwan Lee, Woori Han and “Using Animated Mood Pictures in Music
Youngseop Kim, “Emotional Recognition Recommendation,” 16th International
from Facial Expression Analysis using Conference on Information Visualisation,
Bezier Curve Fitting,” 16th International 2012.
Conference on Network-Based

You might also like