Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165

Analysing Tesla Stock Prices Using Machine

Learning Algorithm
Dr. Chaitanya Kishore Reddy.M1 K.Pradeep2
1 2
Professor & Dean-IT, NRI Institute of Technology, UG Scholar, Dept. Of IT, NRI Institute of Technology,
A.P, India-521212 A.P, India-521212

V.Luke Sundar3 A.Devi Priya4

3 4
UG Scholar, Dept. Of IT, NRI Institute of Technology, UG Scholar, Dept. Of IT, NRI Institute of Technology,
A.P, India-521212 A.P, India-521212

Abstract:- Long-term research has focused heavily on the firm offers. In order to keep the vehicles usable in the
the prediction of stock values. Predicting stock prices has market and also decline of the stock prices the accurate
been one of the biggest concerns in recent years. Trading prediction of used car prices was very important for the new
stocks is a significant economic activity that contributes and existing customers and the company itself. Since the
to society and enables individuals to increase their commencement of the stock market, investors have sought
income. Making a prediction about the stock market to accurately predict stock prices.[2]
involves attempting to anticipate the future value of a
company's stock or another financial instrument traded Few stocks have been as fascinating and contentious
on a stock exchange. Gains for investors will be over the past two years as Tesla. By upending a crucial
maximized if a stock's future price can be accurately industry, the business hopes to profoundly alter the course
predicted. The stock market's issue is that it occasionally of human history. Elon Musk became one of the most
displays an erratic pattern that could result in a crash. prominent CEOs in the world thanks to Tesla, which also
Stock price predictions frequently involve machine made a lot of money for its investors. Market investors,
learning techniques. We employ a variety of machine however, are understandably pessimistic about how high the
learning techniques, both supervised and unsupervised, Tesla stock may rise. Tesla just released their Q3 2020
to inform investors of stock price increases and manufacturing and delivery figures. Despite having
decreases. Data acquisition, dataset preprocessing, delivered 241,000 vehicles, the firm only produced 237,000.
feature extraction, feature-based stock price prediction, With these figures, it is very nearly producing 1 million
and result presentation were the five stages of the units annually. While the international market leader
process. Data is initially gathered from various social Volkswagen group produces 9.3 million vehicles annually,
media platforms and historical company information. domestic brands like Ford produce over 4 million.
Pre-processing is the second phase, where incorrect,
duplicate, and dirt data are removed. In the third phase,
data sets are reduced and meaningful data are chosen. In
the fourth phase, predictions are made utilizing various
machine learning methods, including supervised and
unsupervised learning techniques. Now, various methods
are used in the final phase to determine correctness.

Keywords:- Tesla, Elon Musk, Stock Market, Linear

Regression, Machine Learning

I. INTRODUCTION
Fig 1 Introduction to TESLA
Tesla Inc., an American automaker, was established in
2003. It is one of the businesses that produced electric II. TECHNOLOGY USED
vehicles at a level that allowed for their availability on the
consumer market[Fig:1]. The stock price of Tesla began to  Machine Learning:
decline in the beginning of 2019. This instilled dread in Computers can learn intuitively thanks to the scientific
Tesla owners, and the majority of them began selling their field of machine learning. Machine learning has emerged as
cars. The used automobile market has seen a significant one of the most fascinating technologies ever. The moniker
spike in demand for Tesla vehicles. However, the cost of a signifies that the computer becomes more human-like
Tesla vehicle was heavily influenced by its stock price, through its capacity to learn. There may be a lot more
which had an impact on the market for used Tesla cars. locations than one might assume where machine learning is
Tesla automobiles heavily rely on the post-sale services that currently being actively deployed.

IJISRT23APR1851 www.ijisrt.com 1842

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Two of the most popular machine learning techniques Given that correlation and regression are frequently
are supervised learning and unsupervised learning, which used methods for examining the relationship between
train algorithms using example input and output data that is quantitative variables, it may be helpful to define them first
labelled by humans. Unsupervised learning involves giving for those who have not studied statistics. A measure of
the algorithm no labelled data in order to allow it to find association between two variables that are not categorised as
structure within its input data, and supervised learning uses dependent or independent is called correlation. At its most
example input and output data that is unlabeled. Let's look basic, regression is used to look at the relationship between
into these strategies more closely. two independent variables. Regression allows for prediction
since statistics may be used to forecast the dependent
 Supervised Learning: variable when the independent variable is known.
Examples of inputs and intended outputs are given to
the model or algorithm, which then looks for patterns and Machine learning strategies are always being
relationships between the two. Learning a general rule that improved.For our purposes, we'll go through a couple of the
connects inputs and outputs is the objective. The training well-liked machine learning techniques that are now in use.
procedure is carried out repeatedly until the model's
accuracy on the training set reaches the target level. Here are  Predictive Modelling
a few actual examples: Predictive modelling, in summary, is a statistical
technique that uses machine learning and data mining to
You practise picture classification using labels and predict and anticipate likely future outcomes using historical
images. Then you present a new image in the future with the and existing data. The prediction model predicts the market
expectation that the computer will identify the new object. as Positive or Negative based on a variety of inputs [1].It
functions by analysing recent and old data and applying
Market Prediction/Regression: You teach the computer what it discovers to a model created to predict expected
using past market data and ask it to forecast a future price. outcomes. Almost everything can be predicted with
predictive modelling, including TV ratings, a customer's
 Unsupervised Learning: future purchase, credit risk, and company earnings
The learning system is not given any labels; instead, it
is left to its own devices to identify structure in the data. It is The majority of predictive models operate quickly and
employed to divide the population into various groups. frequently finish their computations in real time. Because of
Unsupervised learning may serve as a primary objective this, businesses like banks and merchants may, for instance,
(discovering hidden patterns in data). forecast the risk associated with an online mortgage or credit
card application and quickly approve or reject the request
Clustering is a crucial step in scientific study where based on that assessment. Although the computation of
you ask the computer to group together comparable pieces some more complex predictive models, such as those used
of data into clusters. in computational biology and quantum computing, takes
longer than that of a credit card application, it can now be
High Dimension Visualization: We can see high completed much more quickly than in the past due to
dimension data with the aid of the computer. advancements in technology, including computing power.

Generated Models: A model can produce more data III. ALGORITHM USED
once it has captured the probability distribution of your
input data.This can help to strengthen your classifier greatly.  Linear Regression:
The term "linear regression" refers to a machine
Unsupervised learning may have the simple objective learning algorithm based on supervised learning. Regression
of finding hidden patterns in a dataset, but it may also have processing is carried out. To model a goal prediction value,
the objective of feature learning, which enables the regression uses independent variables. The main purpose of
computing device to automatically find the representations it is to establish the relationships between variables and
required to categorise the raw data. forecasts. The amount of independent variables that a
regression model uses and the sort of relationship that it
 Approaches considers between the dependent and independent variables
Understanding and utilising machine learning are two factors that make a regression model different
algorithms requires a background in statistics because [Fig:2].
computational statistics and machine learning are closely
related fields. By employing an independent variable as a basis,
linear regression is used to forecast the value of a dependent
CSS separates the document content from document variable (y) (x). This regression method therefore discovers
presentation. There are three types of CSS. They are Inline a linear relationship between x (the input) and y (the output)
CSS, Internal CSS and external CSS. (output). Linear regression was thus first used.

IJISRT23APR1851 www.ijisrt.com 1843

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
portability, the System is dependable and the performance is
maintained.

 Product Charactersistics:

 Usability:
This refers to how easy it is for any type of stock
trader and other stock market participants to understand the
user interface of stock prediction software.

 Efficiency:
Keeping the closing stock prices as accurate as
possible while using the least amount of time and data
possible.

Fig 2 Linear Regression  Performance:

Performance is a feature of stock prediction software
IV. SOFTWARE REQUIREMENTS that refers to how responsive it is to different user
SPECIFICATION interactions.

 Functional Requirements: V. EXISTING SYSTEM

In the technical perspective, functional requirements
address the functionality of the software. It improves and  Tesla Stock Price Prediction Using Sentiment Analysis:
describes the component flow and structural flow of the A tremendously volatile element of the financial world
same. The functional statement works with categorising and is the stock market. A very active area of research and
learning from the same dataset's raw datasets. Then the analysis involves making precise predictions about
datasets are grouped into clusters, and any degradation of numerous stocks. This research attempts to improve the
the groupings is examined for efficiency. After the dataset accuracy of the prediction outcomes by extending the earlier
has been cleaned, the machine learns the data and identifies ML prediction strategies that used artificial neural networks
the pattern set for it. It then goes through a number of and fuzzy-based techniques. A blend of algorithmic trading
iterations and produces results. is at the core of the research since many qualitative elements
influence the decision to purchase or sell a stock. This study
Functional specifications specify what the software intends to investigate the particular connection between
should be able to do (the functions). Consider the Elon Musk's Tweets and the price of Tesla stock. Our
fundamental functions. Functional requirements should be dataset, which had been pre-processed to eliminate any stop
written in the future tense because the "functions" are words, was used as the basis for exploratory data analysis as
already established before development.In the technical the main analysis technique. Combining these approaches
perspective, functional requirements address the and components produced a clear correlation that led to a
functionality of the software. It improves and describes the certain conclusion: more tweets and engagements were
component flow and structural flow of the same. The correlated with higher closing prices for Tesla, and vice
functional statement works with categorising and learning versa.
from the same dataset's raw datasets. Then the datasets are
grouped into clusters, and any degradation of the groupings  Sentiment Analysis:
is examined for efficiency. After the dataset has been Sentiment analysis, commonly referred to as opinion
cleaned, the machine learns the data and identifies the mining, combines modern data science, artificial
pattern set for it. It then goes through a number of iterations intelligence (AI), and the humanities and languages. The
and produces results. fundamental premise of sentiment analysis is to identify the
emotions and subjective feelings hidden behind a set of
 Non Functional Requirements: linguistic data through sentiment analysis. Many methods
Nonfunctional requirements address external aspects exist for conducting sentiment analysis. For instance, by
that are inherently nonfunctional. It serves as a tool for studying big text datasets, a machine may learn to recognise
analysis. The same applies as it carries out the operations' human emotions. We used sentiment analysis because we
performance evaluation. Since stock is practical and thought that tweets' conveyed sentiments would be related to
constantly changing, these additional effects and Tesla stock prices.Almost all facets of our lives need
specifications enable it to receive the most recent updates sentiment analysis. Pathos is a component of all
and integrate in a single step where the technicians may advertisements, campaigns, and commercials, and sentiment
work on and address any bugs or draughts that may exist. analysis is employed in a variety of ways in these mediums.
The efficiency and hit gain ratio are the non-functional The subjective state behind Elon Musk's tweets and replies
parameters that are adhered to. The code's applicability for was applied to Tesla's stock as part of our study's broad use
increased efficacy, implementation, and searching for the of sentiment analysis, which revealed unexpected links
security console. With the help of integration and between the two.

IJISRT23APR1851 www.ijisrt.com 1844

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Based on the predicted emotional language in each B. Modules Involved:
tweet, Twitter categorised each post into three categories:
good, neutral, and negative. Using Python and the
Scikitlearn tool, this was accomplished. The pre-processed
tweets from the text column of the CSV file were used for
this machine learning.

 Disadvantages of Existing System:

 Main drawback is that a lot of training data is needed to

be effective and that they fail to encode the position and
orientation of objects.
 High computational cost is required in order to predict
the future stocks.
 Without GPU data it is quite slow to train the model
 Mostly it depends on previous information for
prediction

VI. PROPOSED SYSTEM

In this research, prediction modelling and regression Fig 3 Data Processing

techniques are combined in an effort to forecast the direction
of the stock market. The data sets acquired are used to feed As input to the models for training, attributes like price
this model the most recent stock prices. The collected data is of open, high, low, close, and modified closing price are
divided up into different subsets or data sets that are used to obtained from a large dataset. Data pre-processing
train and test the algorithm. The data is then modelled using techniques like normalisation and one hot encoding are then
a regression model in R or Python. This model applies a applied to the dataset. After that, the data is split into two
thorough search algorithm to the data sets and, depending on sets, training and testing, with a ratio of 80:20 each. Then, a
the results, generates a summary table. Regression and linear regression model is trained using this collection of
clustering approaches are used to determine if the stock's data. Finally, root mean square error is used to analyse each
price has increased or decreased once the numbers are of these modules.
plotted on a chart.This model extrapolates the current stock
prices based on calculations to produce a prediction after a  Collection:
certain amount of time. Algorithms for supervised machine The most important step in beginning with ML is to
learning are used to create and train the models. The output acquire accurate and high-quality data. Any verified source,
will take the form of a graph and update as the dataset does. such as the UCI dataset repository or data.gov.kaggle.in, can
This model uses supervised machine learning methods and be used to acquire data. For instance, when studying for a
predicts up to 69% in-sample accuracy and 37% out-of- competitive exam, students use the best study materials they
sample accuracy. have access to in order to learn the most effective content
and produce the best outcomes. Similar to this, reliable and
A. Advantages of Proposed System: high-quality data will help the model learn faster and more
Implementing linear regression is straightforward, and effectively, and when tested, the model will produce cutting-
it is simpler to understand the output coefficients. This edge findings. The process of gathering data uses up a
algorithm is the best to apply when you know the tremendous amount of money, time, and resources.For
relationship between the independent and dependent instance, multiple photos of people in a range of human
variable has a linear relationship because it is less difficult expressions are required when working on the Facial
than other algorithms. Over-fitting can occur with linear Expression Recognizer. Solid evidence guarantees that the
regression, however it can be prevented by employing cross- outcomes of the model are valid and can be trusted upon.
validation, regularisation (L1 and L2) techniques, and some
dimensionality reduction approaches. When the dataset can  Preparation:
be separated into linear components, linear regression works The information gathered can be in a raw form that
effectively. It can be used to determine the type of cannot be immediately delivered to the machine. In this
relationship between the variables. process, datasets are gathered from various sources,
examined, and then a new dataset is created for further
processing and research. This preparation can either be done
manually or automatically. Data can also be prepared
numerically, which would speed up the model's learning
process. An image can be converted into a matrix of N x N
dimensions, with each cell's value denoting a single picture
pixel. [Fig:3]

IJISRT23APR1851 www.ijisrt.com 1845

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
 Input: effectively, even though extracting meaning from raw data
The data may not yet be machine-readable once it has is an art in and of itself and necessitates solid feature
been produced, thus certain conversion techniques are engineering skills and domain expertise (in special
needed to turn the data into something that can be read by circumstances) how to divide the data for training [Fig:4]
machines. To finish this exercise, a high level of and testing is the main issue that ML/DL practitioners
computation and accuracy are necessary. Data can be encounter.Even though it initially appears to be a
acquired from a variety of sources, including Twitter straightforward issue, only by delving deeply into it can its
comments, audio recordings, video clips, and MNIST Digit complexity be determined. Inaccurate training and testing
data (images). sets may have unanticipated consequences on the model's
output. It might cause the data to be overfitted or
 Processing: underfitted, and our model might end up producing biassed
At this point, processing the instructions given over a results.
vast amount of data reliably and effectively requires the use
of algorithms and machine learning techniques.  How should the Data be Divided?
The data should preferably be split into three sets: a
 Output: train set, a test set, and a holdout cross-validation set (dev
At this point, the machine generates conclusions that set). Let's first briefly discuss the meaning of these sets and
are clear to the user and meaningful. Reports, charts, the kinds of data they ought to contain.[Fig:4]
movies, and other outputs are a few examples.
 Train Set:
 Storage: The data that would be fed into the model would be
This is the final step in which the obtained output and in the train set. Simply put, our model would gain
the data model data and all the useful information are saved knowledge from this data.
for future use.[Fig:3].
 Dev Set:
C. Datasets: The trained model is validated using the development
Working with various dataset types is essential for set. As the foundation for our model evaluation, this
success in the field of machine learning or to become a great setting is the most crucial.
data scientist. Finding a proper dataset for every type of
machine learning project, however, is a challenging  Test Set:
challenge. A dataset is a group of data that has been The data used to test the trained and approved model
organised in some way. Any type of data, including series, are in the test set. It reveals the effectiveness of our entire
arrays, and database tables, can be found in a dataset. A model and the likelihood that it will forecast an illogical
tabular dataset can be thought of as a database table or event.
matrix where each row correlates to a dataset's fields and
each column to a single variable. "Comma Separated File,"
or CSV, is the file type that is most frequently supported for
tabular datasets. However, in order to store "tree-like
data," we can use the JSON file more efficiently.
Regression-based models are typically used to forecast
continuous values using some provided independent data.[5]

 Kaggle Datasets:
One of the top places to find datasets for data scientists
and machine learners is Kaggle. It makes it simple for users
to locate, download, and publish datasets. Additionally, it
offers the chance to collaborate with other machine learning
professionals and complete challenging Data Science-related
tasks.

We can quickly locate and download a high-quality

dataset from Kaggle in a variety of formats.
Fig 4 Data Splitting
D. Data Splitting:
The core of every ML problem is data. ML models are E. Training Dataset:
like bodies without a soul if they are not fed with the right The training dataset, which is used to train or fit the
data. But gathering data is no longer a significant issue in machine learning model, is the largest (in terms of size)
the 'big data' world of today. Every day, we consciously or subset of the original dataset. In order for the ML algorithms
unconsciously produce enormous datasets. Having an to learn how to make predictions for the given task, training
abundance of data available, however, does not make the data is first fed into them.
issue go away. The quality data is useless unless it is used

IJISRT23APR1851 www.ijisrt.com 1846

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
Whether we are using supervised learning or be completed quickly using Python. Python, though, is
unsupervised learning algorithms, the training data changes. not without flaws. Python is a dynamically written
language, and its modules are well-known for their
In unsupervised learning, inputs are not tagged with duck writing. This could be annoying if a package
the appropriate outputs, hence the training data contains approach returns something that, for example,
unlabeled data points. In order to produce predictions, resembles an array but isn't one. Additionally, the return
models must extract patterns from the provided training type of a method was not explicitly stated in the
datasets. standard Python documentation, which would have
required extensive trial-and-error testing in a less
In contrast, labels are included in the training data for capable language. This issue makes it more challenging
supervised learning in order to help the model be trained and than necessary to learn how to utilise a replacement
predictions made. Python package or library.

The model's accuracy and predictive power are heavily  Numpy:

influenced by the kind of training data we give it. It implies Python's Numpy package offers higher-level scientific
that the model will perform better the higher the quality of and mathematical abstractions wrapped in python. It serves
the training data. A typical ML project's training data makes as the foundational library for scientific computing and
up more than or equal to 60% of the total data. includes facilities for integrating C, powerful n-dimensional
array objects, C++, etc. Additionally, it helps in linear
 Test Dataset: algebra and random number generation.
Once the model has been trained using the training
dataset, it is time to test it using the test dataset. This dataset The array type in Numpy adds an effective data
assesses the model's performance and guarantees that it can structure for numerical computations, such as manipulating
generalise well to new or unexplored datasets. The test matrices, to the Python language. Additionally, Numpy
dataset is a different subset of the original data from the offers fundamental numerical operations, such as tools for
training dataset. When the model training is finished, it identifying Eigenvectors.
utilises it as a benchmark because it has some similar
features and a similar class probability distribution. A well-  Scikit Learn:
organized dataset called test data provides information for A free machine learning library for Python might be
each type of scenario the model might encounter in the Scikit-learn. Numerous techniques for classification,
actual world. The test dataset typically represents 20 to 25 clustering, and regression are included, including random
percent of the total original data for an ML project. forests, k-neighbors, and support vector machines. It also
supports SciPy and NumPy, two Python scientific and
At this point, we may also examine and contrast the mathematical libraries. Scikit-learn is a programme that is
testing accuracy with the training accuracy, or, more specifically written in Python, with the main algorithms
specifically, the accuracy of our model when applied to the implemented in Cython for performance. A wrapper in
test dataset in comparison to the training dataset. The model Cython for LIBSVM enforces support vector machines.
is considered to have overfitted if its accuracy on training Specifically, logistic regression and linear support vector
data is higher than its accuracy on testing data. machines using a wrapper similar to LIBLINEAR.

VII. TOOLS USED  Matplotlib:

Matplotlib is an excellent Python visualisation package
 Python : for 2D array displays. To handle the larger SciPy stack, a
Python was the language we decided to use for our multi-platform data visualisation toolkit called Matplotlib
project. For many reasons, this was an easy call. was developed and is based on NumPy arrays. John Hunter
introduced it for the first time in 2002.
 There is a sizable community that supports Python as a
language. A simple visit to Stack Overflow can fix any One of visualization's main benefits is that it allows us
issues that may arise. The most often used language on visual access to enormous amounts of data in easily
the site, Python, provides the most straightforward understandable forms. In Matplotlib, there are many
answers to all queries. different types of plots, such as line, bar, scatter, and
 Python has several strong tools available for scientific histograms.
computing packages. Packages like NumPy, Pandas,
and SciPy are thoroughly documented and completely  Jupyter Notebook:
free. These packages will drastically reduce and vary An open-source web programme called Jupyter
the amount of code required to create a certain Notebook allows users to create and share documents with
programme. This speeds up repetition. equations, live code, and visualisations. Data can be used for
 Python is a language that is tolerant and allows a variety of purposes, such as data cleansing, data
programmes that appear to be pseudocode.When the visualisation, data transformation, statistical modelling,
pseudocode provided in tutorial papers is required and machine learning, and many more.
checked, this can be useful. This step can occasionally

IJISRT23APR1851 www.ijisrt.com 1847

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
VIII. SYSTEM ARCHITECTURE IX. CONCLUSION

In organisational theory and software engineering, a It appears that Tesla's stock prices will decline soon if
structure chart is a diagram that depicts the breakdown of a they don't come up with a new strategy for communicating
system to its most controllable components. The way a their goal. Given that other companies have begun making
system is utilised, how it communicates with other systems, electric cars at significantly lower prices than Tesla, this
and how it interacts with the outside world are all reflected may be viable.
in its design. It specifies how one component of the system
connects to the others and how data is transferred between A successful future price prediction for a stock will
them. Each module is symbolised by a box that carries its maximise profits for investors.
name. A system's architecture reveals how it is
conceptualised in terms of its structure, operations, and REFERENCES
connections. The term "system" in architecture typically
refers to the software's architecture rather than the actual [1]. M. Usmani, S. H. Adil, K. Raza and S. S. A. Ali,
physical structure of the machines or structures. A system's "Stock market prediction using machine learning
architecture reflects how it is utilised, and as a result, it techniques," 2016 3rd International Conference on
evolves through time. Computer and Information Sciences (ICCOINS),
Kuala Lumpur, 2016, pp. 322-327.
[2]. K. Raza, "Prediction of Stock Market performance by
using machine learning techniques," 2017
International Conference on Innovations in Electrical
Engineering and Computational Technologies
(ICIEECT), Karachi, 2017, pp. 1-1.
[3]. H. Gunduz, Z. Cataltepe and Y. Yaslan, "Stock
market direction prediction using deep neural
networks," 2017 25th Signal Processing and
Communications Applications Conference (SIU),
Antalya, 2017, pp. 1-4.
[4]. M. Billah, S. Waheed and A. Hanifa, "Stock market
prediction using an improved training algorithm of
neural network," 2016 2nd International Conference
on Electrical, Computer & Telecommunication
Engineering (ICECTE), Rajshahi, 2016, pp. 1-4.
[5]. H. L. Siew and M. J. Nordin, "Regression techniques
for the prediction of stock price trend," 2012
International Conference on Statistics in Science,
Business and Engineering (ICSSBE), Langkawi,
2012, pp. 1-5.
[6]. K. V. Sujatha and S. M. Sundaram, "Stock index
prediction using regression and neural network
models under non normal conditions," INTERACT-
2010, Chennai, 2010, pp. 59-63.
[7]. S. Liu, G. Liao and Y. Ding, "Stock transaction
prediction modeling and analysis based on LSTM,"
2018 13th IEEE Conference on Industrial Electronics
and Applications (ICIEA), Wuhan, 2018, pp. 2787-
2790.
[8]. T. Gao, Y. Chai and Y. Liu, "Applying long short
term momory neural networks for predicting stock
closing price," 2017 8th IEEE International
Conference on Software Engineering and Service
Science (ICSESS), Beijing, 2017, pp. 575-578.
[9]. K. A. Althelaya, E. M. El-Alfy and S. Mohammed,
"Evaluation of bidirectional LSTM for short-and
longterm stock market prediction," 2018 9th
International Conference on Information and
Communication Systems (ICICS), Irbid, 2018, pp.
151-156

Fig 5 System Architecture

IJISRT23APR1851 www.ijisrt.com 1848

Volume 8, Issue 4, April – 2023 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165
BIOGRAPHIES

Dr. Chaitanya Kishore Reddy. M is currently working

as a Professor and Dean in the Department of Information
technology at NRI Institute Of Technology,
Pothavarappadu, Agiripalli, Krishna(dist), India. He
received Ph.D. in Computer Science and Engineering and
M. Tech in Computer Science and Engineering at Jawaharlal
Nehru Technological university, Kakinada. He has
Published 40 research papers in various National and
International Journals and International Conferences. He is
a member in ISTE, CSI, and IAENG. His research areas are
Mobile Ad-hoc Networks, IoT, and Cloud Computing.

K. Pradeep is currently studying B.Tech with

specification of Information Technology in NRI Institute of
Technology. He had done a mini project on Tesla Stock
Price Prediction.He had completed 2 NPTEL certification
courses.

V.Luke Sundar is currently studying B.Tech with

specification of Information Technology in NRI Institute of
Technology. He had done a mini project on Tesla Stock
Price Prediction. He had completed 2 NPTEL certification
courses.

A.Devi Priya is currently studying B.Tech with

specification of Information Technology in NRI Institute of
Technology. She had done a mini project on Tesla Stock
Price Prediction. She had completed 2 NPTEL certification
courses.

IJISRT23APR1851 www.ijisrt.com 1849