Load the necessary libraries

  Load the data set

  Convert the categorical variables to equivalent numeric classes

  Split the data set as train set and test set

  Initialize the keras sequential model

  Build the model with input layers,hidden layers and output layer as per the data size along with the activation function(here 1658(vectorizer count) I/P layer with relu activation and 2 O/P layer with sigmoid activation)

  Compile the model with required loss,metrics and optimizer(here loss=binary_crossentropy,optimizer=adam,metrics=accuracy)

  Fit the model using the train set

  Predict using the test set

  Evaluate the metrics

#Load necessary libaries
library(“readtext”)
library(“caret”)
library(“text2vec”)
library(keras)
data data1=(strsplit(data$text,”\n”))
data2=unlist(data1[[1]])
data3=strsplit(data2,”\t”)
data4=unlist(data3)
i=0
j=0
k=0
text=c()
pol=c()
for (i in (1:length(data4)))
{
if(i%%2!=0)
{
j=j+1
text[j]=data4[i]
}else
{
k=k+1
pol[k]=data4[i]
}
}
df=data.frame(text=text,polarity=pol,stringsAsFactors = FALSE)
#Setting seed to reproduce results of random sampling data
set.seed(100)
#Train set
train_ind<-createDataPartition(df$polarity,p=0.8,list=FALSE)
train_data<-df[train_ind,]
xtrain=train_data$text
ytrain=train_data$polarity
#Test set
test=df[-train_ind,]
xtest=test$text
ytest=as.numeric(test[[‘polarity’]])
#Define preprocessing function and tokenization function
prep_fun = tolower
tok_fun = word_tokenizer
#create vector of tokens and build the vocabulary
it_train = itoken(train_data$text,
preprocessor = prep_fun,
tokenizer = tok_fun,
progressbar = FALSE)
vocab = create_vocabulary(it_train)
vectorizer = vocab_vectorizer(vocab)
#Create the document term matrix(DTM) for train set
dtm_train = create_dtm(it_train, vectorizer)
#Create vector of tokens for test set
it_test = itoken(xtest,
preprocessor = prep_fun,
tokenizer = tok_fun,
progressbar = FALSE)
#Create the document term matrix(DTM) for test set
dtm_test=create_dtm(it_test,vectorizer)
ytrain=factor(ytrain)
ytest=factor(ytest)
#converting the target variable to once hot encoded vectors using keras inbuilt function
train_y<-to_categorical(ytrain)
test_y #defining a keras sequential model
model %
layer_dense(units = 70, input_shape = 1658 )%>%
layer_activation(activation = ‘relu’) %>%
layer_dense(units = 2)%>%
layer_activation(activation= “sigmoid”)
#compiling the defined model with metric = accuracy and optimiser as adam.
model %>% compile(
loss = ‘binary_crossentropy’,
optimizer = ‘Adam’,
metrics = ‘accuracy’
)
#Summary of the model
summary(model)
#fitting the model on the training dataset
model %>%keras:: fit(dtm_train, train_y, epochs = 200,batch_size=20)
#Predict using the Test data
yt=predict_classes(model,dtm_test)
loss_and_metrics % evaluate(dtm_test, test_y)
print(loss_and_metrics)
cat(“The confusion matrix is \n”)
print(confusionMatrix(as.factor(yt),as.factor(ytest)))