Data Analysis Project on Cryptocurrency Market Based on Machine Learning Algorithm

Hi! In this project, we are going to analyze the Dataset of Cryptocurrency Market

In this project, you have to Apply Regression to predict the closing price. And after that, you will convert the closing price into True and False. if CLOSE > 2.1

Download the CryptoCurrency Dataset

Now Import some Library to analyze and predict the Closing Price

import pandas as pd
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score

In this Model, we are using Logistic Regression for analysis

Now we are going to read the Cryptocurrency dataset for analysis

df = pd.read_csv("crypto-markets.csv")

After Reading this CSV file we get our dataset

  slug    symbol name    date       ranknow	 open   high   low   close      volume		
0 bitcoin BTC	 Bitcoin 2013-04-28	  1	    135.30 135.98 132.10 134.21	    0.0
1 bitcoin BTC	 Bitcoin 2013-04-29	  1	    134.44 147.49 134.00 144.54	    0.0
2 bitcoin BTC    Bitcoin 2013-04-30   1	    144.00 146.93 134.05 139.00	    0.0
3 bitcoin BTC    Bitcoin 2013-05-01	  1	    139.00 139.89 107.72 116.99	    0.0
4 bitcoin BTC    Bitcoin 2013-05-02   1     116.38 125.60 92.28	 105.21	    0.0              

To find the Total Number of Rows and Columns

df.shape

To find the Name of all Column

df.keys()

As we don't need all the column for calculating the Closing Price so we include only 'open', 'high', 'low', 'close' column in our data frame

df = df[[ "open", "high", "low", "close"]]

There are lot's of the row which has a close value less than 1 so we don't include those rows

df = df[df.close > 1]

Now all the values of close in my data frame are greater than 1

   open	     high	 low	 close
0  135.30   135.98	132.10	134.21
1  134.44   147.49	134.00	144.54
2  144.00	146.93	134.05	139.00
3  139.00	139.89	107.72	116.99
4  116.38	125.60	92.28	105.21

Now we want to analyze the close price which is greater than 2.1 so we give this condition if it is greater than 2.1 then it returns 'True' otherwise 'false'

TF = df.close > 2.1

Now see how many values are greater and less than 2.1

TF.value_counts()

Now we get the output

True    86056
False   38972

Name: close, dtype: int64

Now we are going to predict the Closing Price by my Machine

df["TF"] = TF.values
x = df.drop(["close", "TF"], axis = 1).values
y = df.TF.values

Now Train my machine using sklearn model

trainX, testX, trainY, testY = train_test_split(x,y, test_size= 0.2)
Lr = LogisticRegression()
Lr = Lr.fit(trainX, trainY)
pred = Lr.predict(testX)

After prediction, we get the predicted output

So our Model is Trained now

array([False,  True, False, ...,  True,  True, False])

Now we are going to find out the accuracy score

accuracy_score(pred, trainY)
accuracy_score(pred, testY)

Therefore we get our Accuracy of test data

Accuracy_Score for Traing Data is: 0.9830237347783488
Accuracy_Score for Test Data   is: 0.9830840598256418

So the Accuracy_Score for Training and Testing Data is Nearly equal

Therefore my machine is Ready to analyze our Data.

Thank You for Visiting my Article Hope, you learned how to analyze and train our model.