STOCK PRICE

PREDECTION USING
MACHINE LEARNING
 TEAM
PRESENTATION
MEET BHANUSHALI  KINJAL VAGHASIYA 
 TE A EXTC TE B EXTC
ROLL NO 05 ROLL NO 60

NIYATI KARANI  MEET BHANSALI 

TE B EXTC  TE A EXTC 
 ROLL NO 01 ROLL NO 04

2
 INTRODUCTION
▹ Machine learning has significant applications in the stock
price prediction. In this machine learning project, we will be
talking about predicting the returns on stocks. This is a very
complex task and has uncertainties. We will develop this project
into two parts:
▸ First, we will learn how to predict stock price using
the LSTM neural network.
▸ Then we will build a dashboard using Plotly dash for
stock analysis.

3
 DATASETS
▹ To build the stock price prediction model, we will use the NSE
TATA GLOBAL dataset. This is a dataset of Tata Beverages from
Tata Global Beverages Limited, National Stock Exchange of
India: Tata Global Dataset
▹ To develop the dashboard for stock analysis we will use another
stock dataset with multiple stocks like Apple, Microsoft,
Facebook: Stocks Dataset

4
LSTM (LONG SHORT-
TERM MEMORY)
LSTMs are very powerful in sequence prediction problems
because they’re able to store past information. This is important
in our case because the previous price of a stock is crucial in
predicting its future price.

In order to build the LSTM, we need to import a couple

of modules from Keras:
Sequential for initializing the neural network
Dense for adding a densely connected neural network layer
LSTM for adding the Long Short-Term Memory layer
Dropout for adding dropout layers that prevent overfitting

6
 PREDICTING
FUTURE STOCK USING
THE TEST SET
In order to predict future stock prices we need to do a
couple of things after loading in the test set:
Merge the training set and the test set on the 0 axis.
Set the time step as 60
Use MinMaxScaler to transform the new dataset
Reshape the dataset.

8
 CONCLUSION “
There are a couple of other techniques of predicting stock prices
such as moving averages, linear regression, K-
Nearest Neighbours, ARIMA and Prophet. These are techniques that one
can test on their own and compare their performance with
the Keras LSTM.