RV COLLEGE OF ENGINEERING®

(Autonomous Institution Affiliated to Visvesvaraya Technological University, Belagavi)

DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

Bengaluru – 560059

Internship Report
on
“ Stock Market Prediction using Support Vector Machine ”

CERTIFICATE
18CS74
Certified that the Internship work titled “Stock Market Prediction using Support Vector
Machine” carried out by Priyanshu Kumari (1RV19CS120), a bonafide student, submitted in
Submitted
partial fulfilment for the award of Bachelor by in Computer Science and Engineering
of Engineering
of RV College of Engineering , Bengaluru, affiliated to Visvesvaraya Technological University,
®

Belagavi, during thePriyanshu Kumari

year 2022-23. It is certified that USN: 1RV19CS120
all corrections/suggestions indicated for
internal assessment have been incorporated in the report deposited in the departmental library. The
Internship report has been approved as it satisfies the academic requirement in respect of
internship work prescribed for the said degree.

Submitted in partial fulfilment for the award of degree

of Dr. Ramakanth Kumar P
Head of Department
BACHELOR OF ENGINEERING
Department of CSE
in
RVCE, Bengaluru – 59
Computer Science and Engineering
Name of the DEPARTMENT
Examiners OF COMPUTER SCIENCE AND
Signature with Date
1. ____________________ ENGINEERING __________________

2. ____________________ __________________
2022-23
Certificate provided by Company

(i)
 ACKNOWLEDGEMENT

We are indebted to our guide, Prof Ramma A, Associate Professor, Dept of IEM for
her wholehearted support, suggestions and invaluable advice throughout our internship
and also helped in the preparation of this thesis.

Our sincere thanks to Dr. C K Nagendra Gupta., Professor and Head, Department of
Industrial Engineering and Management, RVCE for his support and encouragement.

Our sincere thanks to Dr. Ramakanth Kumar P., Professor and Head, Department of
Computer Science and Engineering, RVCE for his support and encouragement.

We express sincere gratitude to our beloved Principal, Dr. K. N. Subramanya for his
appreciation towards this project work.

We thank all the teaching staff and technical staff of Industrial Engineering and
Management department and Computer Science and Engineering department, RVCE for their
help.

Priyanshu Kumari
1RV19CS120

(ii)
 Executive Summary

The application of deep learning in the field of stock market prediction has attracted the
interest of many researchers. The work done in this area involves a combination of two ANN
model Support vector machine and Regression and Classification to achieve the regression
line and simultaneously generating context of those given data (prices) to predict the
upcoming price as a caption. This project aims to help the challenged in getting the context of
any image.
A combination of pretrained SVM and LSTM is used here to generate captions, trained on a
dataset of over 5,772 days prices with 5 attributes per day. The algorithm, SVM (Support
Vector Machine) which memorizes the history data for prediction. The paper focuses on the
US market that include companies like Yahoo, Microsoft, Apple, Intel etc. and uses stock
parameters like the date, open, high, low, close and volume. Specifically, Microsoft.
With a training time of over 24 hours, each algorithm was run 30 times to calculate the
prediction accuracy. SVM resulted in 65% mean accuracy and whereas the earlier developed
LSTM model achieved 60% accuracy. Ambiguities within the dataset have been highlighted
and ways to improve the accuracy of the model is also discussed. Challenges with the model,
and the methods of tuning have been explained further.

(iii)
 Table of Contents
Certificate from Industry/Organization (i)
Acknowledgement (ii)
Executive Summary (iii)

Table of Contents (iv)

List of Figures (v)

Chapter 1
Profile of the Organization
1.1 Organizational Structure 1
1.2 Products 1
1.3 Services 2
1.4 Financials 2

Chapter 2
Activities of the Department
2.1 About the department 3
2.2 Products 3
Chapter 3
Tasks Performed
3.1 Summary of the tasks performed during 5
Chapter 4
Reflections
4.1 Technical Knowledge acquired 9
4.2 Soft skills acquired 9

References and Annexures 10

(iv)
List of Figures
1.1.1. Organization structure 1
3.1.1. Installation of python libraries 5
3.1.2. Simple calculator program in python 5
3.1.3. Importing Libraries in Google collab and Collecting
Stock Data 6
3.1.4. Data Pre-processing 6
3.1.5. Using Matplotlib Libraries 6

3.1.6. Glimpse of Dataset 7

3.1.7. Architecture of SVM 7

3.1.8. Predicted Value form all three models and the root mean
square value 8

3.1.9. Support Vector Regression Graph 8

(v)
Chapter 1 Profile of the Organization

1.1. Organizational Structure

Internship is carried out under Centre of Competence in Centre of Logistic and Supply
Chain Management with Bhargawa Info Tech Solutions Pvt Ltd (BITS). Bhargawa Info Tech
Solutions Private Limited (BITS) is a private firm incorporated on 18 November 2019. Its
classified as Nongovernment company and registered at Registrar of Companies, Bangalore.
It is incorporated as a Type “A” Enterprise.
It provides various services in the sector of home automation systems for different industries
and Home purpose. Bhargawa Info Tech Solutions Private Limited's Annual General Meeting
(AGM) was last held on 31 December 2020 and as per records from Ministry of Corporate
Affairs (MCA), its balance sheet was last filed on 31 March 2020. Sudhanva Belle and
Shailaja are the directors of BITS who leads products and service sectors of the company

Figure 1.1.1. Organization structure

1.2. Products
R & D sector of BITS has started on smart home and industrial automation devices.
Industrial automation products refer to use of control systems such as information
technologies, robots and tools for handling different types of machinery and processes used in
several types of industries.
Industrial systems have seen large shifts in their wired and wireless infrastructure this past
decade with the shift from traditional fieldbus technologies to industrial Ethernet, with
protocols such as Ethernet/IP, ProfitNET, and EtherCAT overtaking legacy networks,
Industrial robot manipulators have been utilized in a wide array of industries to perform
repetitive, strenuous, and precise tasks.
Many products have been launched in the sector of home appliances automation devices
which includes wireless water controller for industry and home purposes. Automatic water
level controllers are a product that was created to automatically control a motor, which helps
to ensure a constant reserve of water in a storage tank. These automatic water level
controllers are used to automatically fill the over-head tank when it starts or has become
empty as well as monitor the water level in it.
The technique of water level monitoring and controlling system concentrated with some basic
parts which are softly aggregated together such as Water level Indicator, Level sensors,
microcontroller, pump controlling system etc. The company has successfully made 14 smart
homes and automated 11 industries/factories with custom devices based on client
requirements.

1.3. Services
BITS offer services for the clients which includes consultancies, maintenance of
websites or creation of multimedia presentations for other firms etc. The service provided by
BITS team is to provide the product to the client as per their requirement. Mainly it includes
the services like managing and development of website using several software tools available.
Web Managers are responsible for managing websites for organizations. Their duties include
managing website upgrades, testing for website performance, implement a strategy to
increase web traffic and subscribers, troubleshooting website issues, and responding to
security breaches. These IT specialists from BITS help other organizations by managing their
website. They oversee the functionality of the website, evaluate, and manage website
performance, facilitate hosting and server management, and develop, maintain, and update
website content.
Following duties are performed by IT specialists in BITS:

• Plan, implement, manage, monitor, and upgrade the organization's website.

• Respond to and troubleshoot all website issues.

• Identify and respond to all website security breaches.

• Ensure that the website is protected by enabling the appropriate security measures.

• Update HTML, CSS and JavaScript regularly.

• Conduct content audits to eliminate redundant and/or duplicate information.

• Create appropriate website content aligned with the organization’s strategy.

• Ensure website quality and efficiency by conducting regular test plans.

• Improve the User Experience of the website regularly.

• Create strategies to grow subscriber base and web traffic metrics.

• Ensure full compliance on the website with all laws and regulations.
• Keep up-to-date with industry best practices and monitor competitor websites.
1.4. Financials
BITS have authorized share capital as Rs. 5 lakh and its paid-up capital is Rs. 15
lakhs. The last balance sheet was filed on 31st March 2020.

Chapter 2 Activities of the Department

2.1. About the Department

Centre of Excellence in Logistic and Supply Chain Management at RVCE in
collaboration with BITS located in department of Industrial Engineering and management
provides computing facilities for students, research scholars and faculty of RV College of
Engineering.
RVCE, Bengaluru is established in 1963, offers 12 Undergraduate Engineering programs, 16
master’s degree programs and Doctoral Studies. There are various Centre of Excellence
(CoE) departments at RVCE for different emerging fields. The CoE is a initiative from
RVCE, Bengaluru, to set up an environment that would empower young minds through
knowledge and training for enhancing employability, research, and innovation in emerging
fields such as Visual Computing, IoT, Quantum Computing, mobile applications, Logistic
Management and Supply Chain etc. Internships at various CoE empower engineering students
with up skilling, enrichment of practical knowledge and career building.
The Centre of Competence in Logistic and Supply Chain Management located in department
of Industrial Engineering and management provides computing facilities for students,
research scholars and faculty of RV College of Engineering. The motive of the Supply Chain
Centre is to bring the students and faculty of various disciplines together to execute
interdisciplinary projects. The objective of CoE is to build technical capability and improve
employability among students and fresh graduates. It allows students work on state-of-the-art
multidisciplinary projects, sharpen experimental, analytical, and numerical skills, and learn
the applicability of technologies in diverse fields.
The mission of CoE is to develop employable human resource to meet the challenges in the
field of Logistic, marketing and supply chain at industries. The facility in CoE enables the
execution of computationally intensive projects and research works in the area of Virtual
Reality and Augmented Reality, Edge Computing, Parallel Programming, Artificial
Intelligence and Machine Learning, Stock Prediction and many more.

The Centre offers internships in collaboration with “Bhargawa Info Tech Solutions Private
Ltd” to facilitate skill enhancement in the domains like Audit, Marketing Group, Logistic,
Computer Vision, Mobile app development, Internet of Things, Natural Language Processing,
Human Computer Interaction etc. The available infrastructure of the Centre includes high end
workstations integrated with Titan X Pascal and Quadro RTX 8000 GPU cards and embedded
devices like Jetson Nano, Developer and Tx2 kits.

2.2. Products
R & D sector of BITS has started on smart home and industrial automation devices.
Industrial automation products refer to use of control systems such as information
technologies, robots and tools for handling different types of machinery and processes used in
several types of industries.
The company has successfully made 14 smart homes and automated 11 industries/factories
with custom devices based on client requirements. Many products have been launched in the
sector of home appliances automation devices which includes wireless water controller for
industry and home purposes. Automatic water level controllers are a product that was created
to automatically control a motor, which helps to ensure a constant reserve of water in a
storage tank. These automatic water level controllers are used to automatically fill the over-
head tank when it starts or has become empty as well as monitor the water level in it.
The technique of water level monitoring and controlling system concentrated with some basic
parts which are softly aggregated together such as Water level Indicator, Level sensors,
microcontroller, pump controlling system etc.
Chapter 3 Tasks Performed

3.1. Summary of tasks performed

Anaconda virtual environment created and installed with python 3.8 and all the
libraries required to perform data analysis and visualization.

Figure 3.1.1. Installation of python libraries

Installation of jupyter notebook on Virtual machines and exploration of Google colab

notebook. Learnt python fundamentals and programming such as loop constructs, module
creation, exception handling, creating sets, tuples, dictionaries, etc.

Figure 3.1.2. Simple calculator program in python

Training sessions attended in python basics, machine learning algorithms, and various
frameworks. It also included Data science introduction. Assignments were completed with
the use of python libraries. Exploration of various other python libraries was also involved.
The libraries were meant to be used for machine learning, data cleaning and analysis, and
data visualization techniques.
 Figure 3.1.3. Importing Libraries in Google collab and Collecting Stock Data

Figure 3.1.4. Data Pre-processing

Explored and understood the various python libraries such as matplotlib, numpy, scipy,
seaborn, plotly, pandas, scikit-learn, librosa and many more. Interns gained a sound
knowledge on machine learning models such as classification, regression. Classification
based machine learning algorithms such as Naïve bayes, K-Nearest neighbours, Decision tree,
SVM. Regression techniques such as Linear regression, Logistic regression, Polynomial and
Bayesian regressions.

Figure 3.1.5. Using Matplotlib Libraries

Gained good knowledge on design and development of prediction analysis models using
neural networks like Artificial neural networks, Support Vector Machine, etc. the accuracy is
evaluated of both the methods. The dataset contains the prices of ten years of NIFTY and SP
index. After
evaluating all the methods, the random forest algorithm performs better than other three
models.
Microsoft finance is a type of database which contains the prices of the stock. The stock
prices were scraped from the site using python request library. The data that was extracted
from the companies belonging to the technology sector. The data is ranges from January 2002
till December 2020. The stocks mentioned in the database are Oracle, Amazon, Intel, Lonovo,
Google etc. The most important factors were extracted like the date, volume, close, high, low.
Few other parameters taken into considerations are stock momentum, index volatility, index
momentum and sector momentum.

Figure 3.1.6. Glimpse of Dataset

ALGORITHM Support Vector Machine (SVM): SVM is generally considered to

implement a classification approach but I can be used for regression problems too. A
hyperplane is generated in multidimensional space to differentiate the classes. To minimize
the error, the optimal hyperplane is constructed in an iterative manner. The main feature of
SVM is to neatly segregate the corresponding data and the hyperplane with the maximum
margin is opted between the support vectors. Basically, the algorithm is generated using a
kernel. The kernel trick is a technique where the low dimensional input space is taken and it
is converted into a higher dimensional space. The data was not linear separable, so RBF
(Radial Basis Function) Kernel was used which to give better results for non-linear kernel.
So, firstly SVM is applied to the train dataset and the values are predicted for the test dataset.

Figure 3.1.7. Architecture of SVM

RESULT:
In the case for SVM, the mean accuracy for the 30 runs is 65.20% and the standard deviation
is 0.15. It is noticed that the performance is consistent due to its nature.

Figure 3.1.8. Predicted Value form all three models and the root mean square value

All the three models Radial Basis Model, Linear Model and Polynomial model is depicted in
the form of graph and the next price is also predicted.

Figure 3.1.9. Support Vector Regression Graph

Chapter 4 Reflections

4.1. Technical Knowledge acquired

• Learnt about ANN and SVM
• Learnt about Stock Atrributes
• Learnt about different python libraries
• Learnt how to use Google Collab

4.2. Soft skills acquired

• Worked together with teammates
• Learnt various skills and communicated well with guide
• Majorly focused on learning various skills with interest
 References and Annexures

[1.] Patel, J., Shah, S., Thakkar, P., Kotecha, K. (2015). Predicting stock and stock
price index movement using trend deterministic data preparation and machine
learning techniques. Expert systems with applications, 42(1), 259-268.
[2.] Nelson, D. M., Pereira, A. C., de Oliveira, R. A. (2017, May). Stock market’s
price movement prediction with LSTM neural networks. In 2017 International
joint conference on neural networks (IJCNN) (pp. 1419-1426). IEEE.
[3.] Lee, J. W. (2001, June). Stock price prediction using reinforcement learning. In
ISIE 2001. 2001 IEEE International Symposium on Indus- trial Electronics
Proceedings (Cat. No. 01TH8570) (Vol. 1, pp. 690-695). IEEE.
[4.] Usmani, M., Adil, S. H., Raza, K., Ali, S. S. A. (2016, August). Stock market
prediction using machine learning techniques. In 2016 3rd international
conference on computer and information sciences (ICCOINS) (pp. 322-327).
IEEE.
[5.] Shah, V. H. (2007). Machine learning techniques for stock prediction.
Foundations of Machine Learning— Spring, 1(1), 612.
[6.] Sonakshi Mehra , Dr.G.Manju , Aishwarya D , Yashaswi Bhavaraju Stock Price
Prediction using SVM and LSTM (July 2020), Student, CSE, SRM Institute of
Science and Technology, Kattankulathur, Chennai, India. e-ISSN: 2395-0056.
[7.] WEN Fenghua, XIAO Jihong, HE Zhifang, GONG Xu, ‘Stock Price Prediction
Based on SSA and SVM’, 2nd International Conference on Information
Technology and Quantitative Management, ITQM 2014 Business School of
Central South University, Changsha, 410081, China.
[8.] Ince H, Trafalis T. Short term forecasting with support vector machines and
application to stock price prediction. International Journal of General Systems;
2008, 37(6), p. 677-687.
[9.] Sai Ying, Zhang Fengting, Zhang Tao. China’s stock index futures regression
prediction research based on SVM. China Journal of Management Science;
2013, 03, p. 35-39
[10.] Beneki, C.Eeckels, B., and Leon L., Signal Extraction and Forecasting of the UK
Tourism Income Time Series: A Singular SpectrumAnalysis Approach. Journal
of Forecasting; 2012, 31(5), p. 351-400.