CRYPTOCURRENCY PRICE

ANALYSIS USING
ARTIFICIAL INTELLIGENCE

Internal Guide: Submitted By:

Mrs. E.Sushma M . Vennela(20W91A05D8)
P . Divya(20W91A05F6)
(Assistant Professor) M . Akash(20W91A05D5)
Mohammed Sohail(20W91A0E6)
CONTENTS
 Abstract
 Introduction
 Existing system
 Proposed system
 Hardware & Software
Requirements
 System architecture
 Modules
 Source code
 Output Screens
 Test Cases
 Conclusion
 References
ABSTRACT
 Cryptocurrency is playing an increasingly important role in
reshaping the financial system due to its growing popular
appeal and merchant acceptance.
 while many people are making investments in
cryptocurrency, the dynamically features, uncertainty, the
predictability of cryptocurrency are still mostly known ,
which dramatically risk the investments.
 we use advanced artificial intelligence frameworks of
fully connected Artificial Neural Network (ANN) and long
short-term memory (LSTM) Recurrent Neural Network to
analyze the price dynamics of Bit coin, Ethereum, and
Ripple.
INTRODUCTION
 Crypto currency is the peer-to-peer digital money and
payment system that exist online via a controlled algorithm.
 Crypto currencies are digital or virtual tokens that use
cryptography to secure their transactions and control the
creation of new units. And crypto currencies are often bought
with "fiat" or traditional currency like US dollars or Euros.
However, they can also be bought with crypto currencies like
Bit coin or Ethereum.
 Bit coin is the first and one of the leading digital currencies
(its market capitalization had more than $ 7 billion in 2014,
and then it increased significantly to $ 29 billion in 2017)
which was first introduced by Satoshi Nakamoto in 2008.
 EXISTING SYSTEM
 A few researches have been working
to comprehend the cryptocurrency
time series and develop statistical
models to replicate and forecast
price movements, despite the fact
that there are currently little efforts
on cryptocurrency analysis and
prediction.
 Fortunately, a special class of
Neural Network models known as
Recurrent Neural Networks (RNN)
is just for this purpose.
LIMITATIONS

 Inability to Operate on Sequence

Data
 Dependency on Recurrent Neural
Networks (RNNs)
 Limited Efforts in Cryptocurrency
Analysis and Prediction
PROPOSED SYSTEM
 The proposed model will use an advanced artificial
intelligence frame that comprises of completely connected
Artificial Neural Networks (ANN) and Long Short Term
Memory (LSTM) to study the request dynamics of crypto
currencies.
 Among ANN and LSTM, ANN will analyse more
effectively for long term history, LSTM is for short term
dynamics. LSTM is more useful than ANN for hidden
historical memory data analysis. With this, we can more
accurately estimate the price of the cryptocurrency market.
ADVANTAGES

 Decentralization and Trustlessness

 Blockchain Technology
 Potential for Fraud Prevention
 Efficiency and Cost Savings
 Application of Advanced AI Models
HARDWARE & SOFTWARE REQUIREMENTS
Hardware Requirements:
 System : Pentium IV 2.4.

 Hard Disk : 40GB.

 Ram : 512MB.

Software Requirements:
 Operating system : Windows 7 Ultimate.

 Coding Language : Python.

 Data Base : My SQL

SYSTEM ARCHITECTURE
 CLASS DIAGRAM:
A class diagram in UML illustrates the static structure of a software system,
detailing classes, attributes , operations, and their relationships.
 USECASE DIAGRAM:
A UML use case diagram provides a visual representation of a system’s
functionality, showcasing actors, their goals (use cases), and dependencies,
with a focus on illustrating the system functions performed for each actor.
 SEQUENCE DIAGRAM:
A UML sequence diagram depicts the interactions and order of processes in
a system, derived from Message Sequence Charts and known as event diagrams
or timing diagrams.
MODULES

User
Registration of User details next admin activated by the user. go to user
page and login. User view some fields now open start trading page contains
sale available cryptocurrencies and user buy the currency. User views the
transaction history details and user view the prediction for available datasets.
Agent
First registration of the agent page. next activated by the admin. Login
agent page. agent view some fields. Here three types of digital currency’s are
there [ bitcoins, ripple, Ethereum] agent can buy digital currency (any one or
all three).Agent buying the crypto currency. agent transaction history also
available.
Admin
The aim of admin is to approve the users and agents .admin contains some fields
admin view user and agent registered details and in crypto field user update the
currency rate and view the recently crypto currency changes list. admin view the
current transaction details.
Artificial intelligence
The application of advanced digital, smart technologies, robotic systems, new
materials and design techniques, creation of large data processing systems,
computer-aided learning and artificial intelligence (AI) are relevant for various
branches of science and technology, including manned space programs. Some
technology concepts and pilot systems based on the AI were developed in the
industry over several decade.
 SOURCE CODE
Urls.py
from django. contrib import admin
from django.urls import path
from .views import index,users,agents,admins,usersignup,agentsignup,logout

from users.views import

bituserregister,userlogincheck,StartUserTrading,UserBuyQuantity,UserBuyingCoins,UserTransactions
History,UserPredictionTest,UserPredictTestProcess
from agents.views import
bitagentregister,agentlogincheck,AgentBuyCrypto,agentbuycurrency,AgentTransactions,AgentHadCoi
ns,AgentLedgerStatus,AgentPredectionTest,AgentredictTestProcess
from admins.views import
adminlogincheck,viewusers,viewagents,activatewaitedusers,activatewaitedagents,currentrate,upda
tecryptocurrency,AdminGetLedger
urlpatterns = [

path('admin/', admin.site.urls),
path('',index,name='index'),
path('index/', index, name='index'),
path('users/',users,name='users'),
path('agents/',agents,name='agents'),
path('admins/',admins,name='admins'),
path('usersignup/',usersignup,name='usersignup'),
path('agentsignup/',agentsignup,name='agentsignup'),
path('logout/',logout,name='logout'),
path('bituserregister/',bituserregister,name='bituserregister'),
path('userlogincheck/',userlogincheck,name='userlogincheck'),
path('UserTransactionsHistory/',UserTransactionsHistory,name='UserTransactionsHistory'),
path('UserPredictionTest/',UserPredictionTest,name='UserPredictionTest'),
path('bitagentregister/',bitagentregister,name='bitagentregister'),
path('agentlogincheck/',agentlogincheck,name='agentlogincheck'),
path('AgentBuyCrypto/',AgentBuyCrypto,name='AgentBuyCrypto'),
path('agentbuycurrency/<currencyname>',agentbuycurrency,name='agentbuycurrency'),
path('AgentTransactions/',AgentTransactions,name='AgentTransactions'),
path('AgentHadCoins/',AgentHadCoins,name='AgentHadCoins'),
path('currentrate/',currentrate,name='currentrate'),
path('updatecryptocurrency/<curr>',updatecryptocurrency,name='updatecryptocurrency'),
path('AdminGetLedger/',AdminGetLedger,name='AdminGetLedger'),
path('UserPredictTestProcess/<value>',UserPredictTestProcess,name='UserPredictTestProcess'),
]
OUTPUT SCREENS

Main Home Page

User Login Page
User Registration Form
Agent Login page
Agent Register page
Admin Login Page
Admin Activate Users
Admin Activate Agents
Current Price and Update
Crypto update history
Blockchain ledger maintance
Agent buying crypto coins
Agent buy Transactions
Agent Transactions History
Agent view Ledger balance
Dataset analysis
TEST CASES
Remarks (IF
S.no Test Case Excepted Result Result
fails)

If user registration If user is not

1 User REGISTERED Pass
successfully. registered.

If agent registration If agent is not

2 Agent REGISTERED Pass
successfully. registered.

user rights will be If user is not

3 ADMIN Pass
accepted here. registered.

agent rights will be If agent is not

4 ADMIN Pass
accepted here. registered.
 If user
If user name and
name or
password is correct
5 User LOGIN Pass password
then it will getting
is not
valid page.
correct.
If agent
If agent name and
name or
password is correct
6 agent LOGIN Pass password
then it will getting
is not
valid page.
correct.
If sale
Agent buying If agent is correct crypto
7 crypto currency then it will getting Pass currencies
from admin valid page. are not
available.
If sale
User buying crypto If user is correct crypto
8 currency from then it will getting Pass currencies
agent valid page are not
available
CONCLUSION
• Cryptocurrency, such as Bitcoin, has established itself as the leading
role of decentralization.
• There are a large number of cryptocurrencies sprang up after Bitcoin
such as Ethereum and Ripple.
• In this study, we use two distinct artificial intelligence frameworks,
namely, fully-connected Artificial Neural Network (ANN) and Long-
Short-Term-Memory (LSTM) to analyze and predict the price
dynamics of Bitcoin, Ethereum, and Ripple.
• This study provides a unique demonstration that Cryptocurrency
market price is predictable. However, the explanation of the
predictability could vary depending on the nature of the involved
machine-learning model.
 REFERENCES
• Greaves, A., & Au, B. (2015). Using the bitcoin transaction graph to
predict the price of bitcoin. No Data.
• Hayes, A. S. (2017). Cryptocurrency value formation: An empirical
study leading to a cost of production model for valuing bitcoin.
Telematics and Informatics, 34(7), 1308-1321.
• Ariyo AA, Adewumi AO, Ayo CK. Stock price prediction using the
ARIMA model. In UKSimAMSS 16th IEEE International Conference
on Computer Modelling and Simulation (UKSim), 2014, pp. 106-112
• Shah, D., & Zhang, K. (2014, September). Bayesian regression and
Bitcoin. In Communication, Control, and Computing (Allerton), 2014
52nd Annual Allerton Conference on (pp. 409-414). IEEE.
THANK
YOU