diff --git a/Sentiment Classifier API/app.py b/Sentiment Classifier API/app.py
new file mode 100644
index 0000000..42381ac
--- /dev/null
+++ b/Sentiment Classifier API/app.py	
@@ -0,0 +1,51 @@
+from flask import Flask
+from flask_restful import reqparse, abort, Api, Resource
+import pickle
+import numpy as np
+from model import SentimentClassifierModel
+
+app = Flask(__name__)
+api = Api(app)
+
+model = SentimentClassifierModel()
+
+classifier_path = 'lib/models/SentimentClassifier.pkl'
+with open(classifier_path, 'rb') as f:
+    model.classifier = pickle.load(f)
+
+vectorizer_path = 'lib/models/tfidf_vectorizer.pkl'
+with open(vectorizer_path, 'rb') as f:
+    model.vectorizer = pickle.load(f)
+
+#parse args
+parser = reqparse.RequestParser()
+parser.add_argument('query')
+
+class PredictUserSentiments(Resource):
+    def get(self):
+        args = parser.parse_args()
+        user_query  = args['query']
+
+        vectorize_userq = model.transform_vectorizer(np.array([user_query]))
+        prediction = model.predict_sentiment(vectorize_userq)
+        predict_probability = model.predict_proba(vectorize_userq)
+
+        if prediction == 0:
+            predicted_sentiment = 'Negative'
+        else:
+            predicted_sentiment = 'Positive'
+
+        #confidence level
+        c_level = round(predict_probability[0], 3)
+
+        #json obj 
+        output_as_json = {'sentiment': predicted_sentiment, 'confidence': c_level}
+
+        return output_as_json
+
+#add resource routes
+api.add_resource(PredictUserSentiments, '/')
+
+#run app
+if __name__ == '__main__':
+    app.run(debug=True)
diff --git a/Sentiment Classifier API/build_model.py b/Sentiment Classifier API/build_model.py
new file mode 100644
index 0000000..498aaf2
--- /dev/null
+++ b/Sentiment Classifier API/build_model.py	
@@ -0,0 +1,32 @@
+from model import SentimentClassifierModel
+import numpy as np
+import pandas as pd
+from sklearn.model_selection import train_test_split
+
+
+def build_model():
+    model = SentimentClassifierModel()
+
+    with open('D:/Projects/REST APIs Flask/NLP Sentiment Classifier API/lib/data/train.tsv') as f:
+        data = pd.read_csv(f, sep='\t')
+        data = data.replace(to_replace = 'None', value = np.nan).dropna()
+
+    positive_negative = data[(data['Sentiment'] == 0) | (data['Sentiment'] == 4)]
+
+    positive_negative['Binary'] = positive_negative.apply(lambda x : 0 if x['Sentiment'] == 0 else 1, axis=1)
+
+    model.fit_vectorizer(positive_negative.loc[:,'Phrase'])
+    
+    x = model.transform_vectorizer(positive_negative.loc[:,'Phrase'])
+    y = positive_negative.loc[:, 'Binary']
+
+    x_train, x_test, y_train, y_test = train_test_split(x,y)
+
+    model.train(x_train, y_train)
+    print("Model trained successfully!")
+
+    model.pickle_classifier()
+    model.pickle_tfidfvec()
+
+if __name__ == "__main__":
+    build_model()
diff --git a/Sentiment Classifier API/model.py b/Sentiment Classifier API/model.py
new file mode 100644
index 0000000..5b8dfc3
--- /dev/null
+++ b/Sentiment Classifier API/model.py	
@@ -0,0 +1,43 @@
+#importing dependencies
+from sklearn.naive_bayes import MultinomialNB
+from sklearn.feature_extraction.text import TfidfVectorizer
+import pickle
+
+class SentimentClassifierModel(object):
+
+    def __init__(self):
+        self.classifier = MultinomialNB()
+        self.vectorizer = TfidfVectorizer()
+
+    def fit_vectorizer(self, x):
+        #fit tfidf to corpus
+        self.vectorizer.fit(x)
+
+    def transform_vectorizer(self,x):
+        #transform corpus to vector matrix (sparse tfidf matx)
+        transform_x = self.vectorizer.transform(x)
+        return transform_x
+
+    def train(self, x, y):
+        #trains the text classifier to associate target labels(sentiment scores) with matrix labels
+        self.classifier.fit(x,y)
+
+    def predict_proba(self,x):
+        #returns conditional probability of class y (target) wrt x for bin class 1
+        proba_y = self.classifier.predict_proba(x)
+        return proba_y[:,1]
+
+    def predict_sentiment(self,x):
+        y_pred = self.classifier.predict(x)
+        return y_pred
+
+    def pickle_tfidfvec(self, path = 'lib/models/tfidf_vectorizer.pkl'):
+        with open(path, 'wb') as f:
+            pickle.dump(self.vectorizer, f)
+            print("Pickled tfidf-vectorizer at {}".format(path))
+
+    def pickle_classifier(self, path ='lib/models/SentimentClassifier.pkl'):
+        with open(path, 'wb') as f:
+            pickle.dump(self.classifier, f)
+            print("Pickled classifier at {}".format(path))
+