Uploaded Project files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+airline_sentiment.csv filter=lfs diff=lfs merge=lfs -text
+nlp_model/variables/variables.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text

acronym.json

@@ -0,0 +1 @@
+{"aka":"also known as","asap":"as soon as possible","brb":"be right back","btw":"by the way","dob":"date of birth","faq":"frequently asked questions","fyi":"for your information","idk":"i don't know","idc":"i don't care","iirc":"if i recall correctly","imo":"in my opinion","irl":"in real life","lmk":"let me know","lol":"laugh out loud","ngl":"not gonna lie","noyb":"none of your business","nvm":"never mind","ofc":"of course","omg":"oh my god","pfa":"please find attached","rofl":"rolling on the floor laughing","stfu":"shut the fuck up","tba":"to be announced","tbc":"to be continued","tbd":"to be determined","tbh":"to be honest","ttyl":"talk to you later","wtf":"what the fuck","wth":"what the heck"}

airline_recommend.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c79a26d312394e66dc29dba164097854cf47fa66eb3c2ffb6309493e80d71ae
+size 8192271

airline_sentiment.csv

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0246fb482ad199fe0820d169dc2db34fd5be3c310819aa6a0994f0c0822919b
+size 46112842

app.py

@@ -0,0 +1,319 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import joblib
+import tensorflow as tf
+import re
+import string
+import nltk
+from nltk.tokenize import RegexpTokenizer
+from nltk.stem import WordNetLemmatizer 
+
+df = pd.read_csv("airline_reviews_cleaned.csv")
+st.set_page_config(layout="wide", page_title="AeroAdvisor", page_icon="✈️")
+
+css = f"""
+<style>
+[data-testid="stAppViewContainer"] > .main {{
+background-image: url("https://cdn.discordapp.com/attachments/1075699203046641687/1165351110828101673/PhotoReal_From_a_distance_the_plane_appears_as_a_tiny_speck_ag_1.jpg?ex=654688cb&is=653413cb&hm=096dcc994304b93afd210555607d563f725d74c1dddfd392176f11e15076bcfa&");
+background-size: 120%;
+background-position: top left;
+background-repeat: repeat;
+background-attachment: local;
+}}
+
+[data-testid="stHeader"] {{
+background: rgba(0,0,0,0);
+}}
+
+[data-testid="stExpander"] {{
+background: rgba(0,0,0,0.5);
+border: 2px solid #000071;
+border-radius: 10px;
+}}
+</style>
+"""
+st.markdown(css, unsafe_allow_html=True)
+
+#-------------------------------- Function to clean reviews -------------------------------------#
+
+# Check if wordnet is installed
+try:                                                                         
+    nltk.find("corpora/popular.zip")          
+except LookupError:
+    nltk.download('popular')
+
+
+# Defining acronyms
+acronyms_dict = pd.read_json('acronym.json', typ = 'series')
+
+# Defining contractions
+contractions_dict = pd.read_json('contractions.json', typ = 'series')
+
+# Defining stopwords
+
+alphabets = ["a", "b", "c", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "q", "r", "s", "t", "u", "v", "w", "x", "y", "z"]
+others = ["ã", "å", "ì", "û", "ûªm", "ûó", "ûò", "ìñ", "ûªre", "ûªve", "ûª", "ûªs", "ûówe", "ï", "ûï"]
+stops = alphabets + others
+stops = list(set(stops))
+
+# Defining tokenizer
+regexp = RegexpTokenizer("[\w']+")
+
+# Preprocessing
+def preprocess(text):
+    
+    text = text.lower()                                                                                        # lowercase
+    text = text.strip()                                                                                        # whitespaces
+    
+    # Removing html tags
+    html = re.compile(r'<.*?>')
+    text = html.sub(r'', text)                                                                                 # html tags
+    
+    # Removing emoji patterns
+    emoji_pattern = re.compile("["
+                           u"\U0001F600-\U0001F64F"  # emoticons
+                           u"\U0001F300-\U0001F5FF"  # symbols & pictographs
+                           u"\U0001F680-\U0001F6FF"  # transport & map symbols
+                           u"\U0001F1E0-\U0001F1FF"  # flags (iOS)
+                           u"\U00002702-\U000027B0"
+                           u"\U000024C2-\U0001F251"
+                           "]+", flags = re.UNICODE)
+    text = emoji_pattern.sub(r'', text)                                                                         # unicode char
+    
+    # Removing urls
+    http = "https?://\S+|www\.\S+" # matching strings beginning with http (but not just "http")
+    pattern = r"({})".format(http) # creating pattern
+    text = re.sub(pattern, "", text)                                                                            # remove urls
+    
+    # Removing twitter usernames
+    pattern = r'@[\w_]+'
+    text = re.sub(pattern, "", text)                                                                            # remove @twitter usernames
+    
+    # Removing punctuations and numbers
+    punct_str = string.punctuation + string.digits
+    punct_str = punct_str.replace("'", "")
+    punct_str = punct_str.replace("-", "")
+    text = text.translate(str.maketrans('', '', punct_str))                                                     # punctuation and numbers
+    
+    # Replacing "-" in text with empty space
+    text = text.replace("-", " ")                                                                               # "-"
+    
+    # Substituting acronyms
+    words = []
+    for word in regexp.tokenize(text):
+        if word in acronyms_dict.index:
+            words = words + acronyms_dict[word].split()
+        else:
+            words = words + word.split()
+    text = ' '.join(words)                                                                                       # acronyms
+    
+    # Substituting Contractions
+    words = []
+    for word in regexp.tokenize(text):
+        if word in contractions_dict.index:
+            words = words + contractions_dict[word].split()
+        else:
+            words = words + word.split()
+    text = " ".join(words)                                                                                       # contractions
+    
+    punct_str = string.punctuation
+    text = text.translate(str.maketrans('', '', punct_str))                                                     # punctuation again to remove "'"
+                                                                       
+    # lemmatization
+    lemmatizer = WordNetLemmatizer()
+    text = " ".join([lemmatizer.lemmatize(word) for word in regexp.tokenize(text)])                             # lemmatize
+    
+    # Stopwords Removal
+    text = ' '.join([word for word in regexp.tokenize(text) if word not in stops])                              # stopwords
+    
+    # Removing all characters except alphabets and " " (space)
+    filter = string.ascii_letters + " "
+    text = "".join([chr for chr in text if chr in filter])                                                      # remove all characters except alphabets and " " (space)
+    
+    # Removing words with one alphabet occuring more than 3 times continuously
+    pattern = r'\b\w*?(.)\1{2,}\w*\b'
+    text = re.sub(pattern, "", text).strip()                                                                    # remove words with one alphabet occuring more than 3 times continuously
+    
+    # return final output
+    return text
+
+#-------------------------------- Container 1 for Heading -------------------------------------#
+container_1 = st.container()
+with container_1:
+    empty1, head2, empty3 = st.columns(spec = [2,3,2], gap = 'medium')
+    with empty1:
+        st.empty()
+    with head2:
+        st.title("Welcome Aboard")
+        st.write("## Tell us about your Experience :airplane:")
+    with empty3:
+        st.empty()
+        
+#-------------------------------- Container 2 for main_content --------------------------------# 
+container_2 = st.container()
+with container_2:
+    col1, col2, col3, col4 = st.columns(spec = [1,3,3,1], gap = 'medium')
+    with col1:
+        st.empty()
+
+    with col2:
+        expander_1 = st.expander(label = "Your Trip Info", expanded = True)
+        with expander_1:
+            airline = st.selectbox(
+                label = "Enter your Airline",
+                options = tuple(sorted(df['airline'].unique())),
+                index = None,
+                placeholder = "Choose an option..."
+            )
+
+            traveller_type = st.selectbox(
+                label = "Enter your trip type",
+                options = ("Business", "Solo Leisure", "Couple Leisure", "Family Leisure"),
+                index = None,
+                placeholder = "Choose an option..."
+            )
+
+            cabin = st.selectbox(
+                label = "Enter your seat class",
+                options = ("Economy Class", "Premium Economy", "Business Class", "First Class"),
+                index = None,
+                placeholder = "Choose an option..."
+            )
+
+            type_of_flight = st.radio(
+                label = "Enter your flight type",
+                options = ("Direct", 'Indirect'),
+                index = 0,
+            )
+
+            frequency = st.radio(
+                label = "How often do you fly?",
+                options = ('Often', 'Occasionally', 'Rarely'),
+                index = 1,
+            )
+
+    with col3:
+        expander_2 = st.expander(label = "Your Ratings", expanded = True)
+        with expander_2:
+            seat_comfort = st.slider(
+                label = "How comfortable are you with your seat?",
+                min_value = 1,
+                max_value = 5,
+                value = 3
+            )
+
+            cabin_service = st.slider(
+                label = "Please Rate your Cabin Service",
+                min_value = 1,
+                max_value = 5,
+                value = 3
+            )
+
+            food_bev = st.slider(
+                label = "Please rate the quality of food/beverages",
+                min_value = 1,
+                max_value = 5,
+                value = 3
+            )
+
+            entertainment = st.slider(
+                label = "Please rate the Entertainment Service",
+                min_value = 1,
+                max_value = 5,
+                value = 3
+            )
+
+            ground_service = st.slider(
+                label = "Please rate the Ground Service",
+                min_value = 1,
+                max_value = 5,
+                value = 3
+            )
+
+            value_for_money = st.slider(
+                label = "Value for Money Rating",
+                min_value = 1,
+                max_value = 5,
+                value = 3
+            )
+        
+    with col4:
+        st.empty()
+
+#-------------------------------- Container 3 for Final Rating Slider --------------------------------#
+container_3 = st.container()
+with container_3:
+    empty1, head2, empty3 = st.columns(spec = [1,3,1], gap = 'medium')
+    with empty1:
+        st.empty()
+    with head2:
+        overall = st.slider(
+            label = "How was your overall experience with the Airline?",
+            min_value = 1,
+            max_value = 10,
+            value = 7
+        )
+
+        review = st.text_area("Enter your review")
+    with empty3:
+        st.empty()
+
+
+#-----------------------------------------------------------------------------------------------------#
+# Creating DataFrame using values input by user
+temp_df = pd.DataFrame(
+                data = [[airline, traveller_type, cabin, type_of_flight, frequency,
+                            seat_comfort, cabin_service, food_bev, entertainment,
+                            ground_service, value_for_money, overall]], 
+                columns = ['airline', 'traveller_type', 'cabin', 'type_of_flight', 'frequency',
+                            'seat_comfort', 'cabin_service', 'food_bev', 'entertainment',
+                            'ground_service', 'value_for_money', 'overall']
+)
+
+# Loading Model using joblib file
+model = joblib.load('airline_recommend.joblib')
+
+# Defining a function to store the nlp_model in streamlit cache memory
+@st.cache_resource
+def cache_model(model_name):
+    model = tf.keras.models.load_model(model_name)
+    return model
+
+# Loading the nlp_model
+nlp_model = cache_model("nlp_model")
+
+#-------------------------------- Container 4 for Final Predictions --------------------------------#
+
+container_4 = st.container()
+with container_4:
+    empty1, head2, empty3 = st.columns(spec = [1.2,2,1.5], gap = 'medium')
+    with empty1:
+        st.empty()
+        
+    with head2:
+        # Creating a button to get prediction
+        if st.button('Get Prediction'):
+            y_pred = model.predict(temp_df)
+            y_pred_prob = model.predict_proba(temp_df)
+
+            if review=="":
+                st.warning("Please enter your review")
+                st.stop()
+
+            clean_review = preprocess(review)
+            review_pred_proba = nlp_model.predict([clean_review])
+            review_pred = np.where(review_pred_proba > 0.5, 1, 0)[0][0]
+
+            if (y_pred[0] == 'yes') & (review_pred == 1):
+                st.success("Thank you for your positive feedback! \nWe're delighted to hear that you had a great experience with our service.")
+                st.balloons()
+            elif (y_pred[0] == 'yes') & (review_pred == 0):
+                st.warning("We appreciate your positive rating, but we're sorry to hear about your concerns in the review. \nPlease share more details so we can address them and enhance your experience.")
+            elif (y_pred[0] == 'no') & (review_pred == 0):
+                st.error("We apologize for falling short of your expectations. \nYour feedback is valuable, and we're committed to improving. \nPlease provide specific details about your experience for us to better understand and address the issues.")
+            elif (y_pred[0] == 'no') & (review_pred == 1):
+                st.error("We're sorry to hear about your negative rating, but we're glad to see your positive comments in the review. \nWe'd like to learn more about your concerns to ensure we address any issues and enhance your satisfaction.")
+    
+    with empty3:
+        st.empty()

contractions.json

@@ -0,0 +1 @@
+{"'aight":"alright","ain't":"are not","amn't":"am not","arencha":"are not you","aren't":"are not","'bout":"about","can't":"cannot","cap'n":"captain","'cause":"because","'cept":"except","could've":"could have","couldn't":"could not","couldn't've":"could not have","dammit":"damn it","daren't":"dare not","daresn't":"dare not","dasn't":"dare not","didn't":"did not","doesn't":"does not","doin'":"doing","don't":"do not","dunno":"do not know","d'ye":"do you","e'en":"even","e'er":"ever","'em":"them","everybody's":"everybody is","everyone's":"everyone is","fo'c'sle":"forecastle","finna":"fixing to","'gainst":"against","g'day":"good day","gimme":"give me","giv'n":"given","gonna":"going to","gon't":"go not","gotcha":"got you","gotta":"got to","gtg":"got to go","hadn't":"had not","had've":"had have","hasn't":"has not","haven't":"have not","he'd":"he had","he'll":"he shall","helluva":"hell of a","he's":"he is","here's":"here is","he've":"he have","how'd":"how would","howdy":"how do you do","how'll":"how will","how're":"how are","how's":"how is","i'd":"i had","i'd've":"i would have","i'll":"i shall","i'm":"i am","imma":"i am about to","i'm'a":"i am about to","i'm'o":"i am going to","innit":"is it not","ion":"i do not","i've":"i have","isn't":"is not","it'd":"it would","it'll":"it shall","it's":"it is","iunno":"i do not know","kinda":"kind of","let's":"let us","li'l":"little","ma'am":"madam","mayn't":"may not","may've":"may have","methinks":"me thinks","mightn't":"might not","might've":"might have","mustn't":"must not","mustn't've":"must not have","must've":"must have","'neath":"beneath","needn't":"need not","nal":"and all","ne'er":"never","o'clock":"of the clock","o'er":"over","ol'":"old","oughtn't":"ought not","'round":"around","'s":"is","shalln't":"shall not","shan't":"shall not","she'd":"she had","she'll":"she shall","she's":"she is","should've":"should have","shouldn't":"should not","shouldn't've":"should not have","somebody's":"somebody is","someone's":"someone is","something's":"something is","so're":"so are","so's":"so is","so've":"so have","that'll":"that shall","that're":"that are","that's":"that is","that'd":"that would","there'd":"there had","there'll":"there shall","there're":"there are","there's":"there is","these're":"these are","these've":"these have","they'd":"they had","they'll":"they shall","they're":"they are","they've":"they have","this's":"this is","those're":"those are","those've":"those have","'thout":"without","'til":"until","'tis":"it is","to've":"to have","'twas":"it was","'tween":"between","'twhere":"it were","wanna":"want to","wasn't":"was not","we'd":"we had","we'd've":"we would have","we'll":"we shall","we're":"we are","we've":"we have","weren't":"were not","whatcha":"what are you","what'd":"what did","what'll":"what shall","what're":"what are","what's":"what is","what've":"what have","when's":"when is","where'd":"where did","where'll":"where shall","where're":"where are","where's":"where is","where've":"where have","which'd":"which had","which'll":"which shall","which're":"which are","which's":"which is","which've":"which have","who'd":"who would","who'd've":"who would have","who'll":"who shall","who're":"who are","who's":"who is","who've":"who have","why'd":"why did","why're":"why are","why's":"why is","willn't":"will not","won't":"will not","wonnot":"will not","would've":"would have","wouldn't":"would not","wouldn't've":"would not have","y'all":"you all","y'all'd've":"you all would have","y'all'd'n't've":"you all would not have","y'all're":"you all are","y'all'ren't":"you all are not","y'at":"you at","yes'm":"yes madam","yessir":"yes sir","you'd":"you had","you'll":"you shall","you're":"you are","you've":"you have","aight":"alright","aint":"are not","amnt":"am not","arent":"are not","cant":"cannot","cause":"because","couldve":"could have","couldnt":"could not","couldntve":"could not have","darent":"dare not","daresnt":"dare not","dasnt":"dare not","didnt":"did not","doesnt":"does not","doin":"doing","dont":"do not","eer":"ever","everybodys":"everybody is","everyones":"everyone is","gday":"good day","givn":"given","gont":"go not","hadnt":"had not","hadve":"had have","hasnt":"has not","havent":"have not","hed":"he had","hell":"he shall","hes":"he is","heve":"he have","howd":"how did","howll":"how will","howre":"how are","hows":"how is","idve":"i would have","ill":"i shall","im":"i am","ima":"i am about to","imo":"i am going to","ive":"i have","isnt":"is not","itd":"it would","itll":"it shall","its":"it is","lets":"let us","lil":"little","maam":"madam","maynt":"may not","mayve":"may have","mightnt":"might not","mightve":"might have","mustnt":"must not","mustntve":"must not have","mustve":"must have","neednt":"need not","neer":"never","oclock":"of the clock","oer":"over","ol":"old","oughtnt":"ought not","shallnt":"shall not","shant":"shall not","shed":"she had","shell":"she shall","shes":"she is","shouldve":"should have","shouldnt":"should not","shouldntve":"should not have","somebodys":"somebody is","someones":"someone is","somethings":"something is","thatll":"that shall","thatre":"that are","thatd":"that would","thered":"there had","therell":"there shall","therere":"there are","theres":"there is","thesere":"these are","theseve":"these have","theyd":"they had","theyll":"they shall","theyre":"they are","theyve":"they have","thiss":"this is","thosere":"those are","thoseve":"those have","tis":"it is","tove":"to have","twas":"it was","wasnt":"was not","wed":"we had","wedve":"we would have","were":"we are","weve":"we have","werent":"were not","whatd":"what did","whatll":"what shall","whatre":"what are","whats":"what is","whatve":"what have","whens":"when is","whered":"where did","wherell":"where shall","wherere":"where are","wheres":"where is","whereve":"where have","whichd":"which had","whichll":"which shall","whichre":"which are","whichs":"which is","whichve":"which have","whod":"who would","whodve":"who would have","wholl":"who shall","whore":"who are","whos":"who is","whove":"who have","whyd":"why did","whyre":"why are","whys":"why is","wont":"will not","wouldve":"would have","wouldnt":"would not","wouldntve":"would not have","yall":"you all","yalldve":"you all would have","yallre":"you all are","youd":"you had","youll":"you shall","youre":"you are","youve":"you have","'re":"are","thats":"that is"}

nlp_model/fingerprint.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ff0637eaea9c0888a9a1a47f51eb29379df82b783d4a6d32e02374bc8b9e368
+size 57

nlp_model/keras_metadata.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5316ddb63ad7ec3c3f14a8ebc6e92857230dabfdba85d3b3bbf5bc1fb71ce7d6
+size 21115

nlp_model/saved_model.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0da8aec7aca4c5d0faf1ed830243e6fee7c6a38cc6d92e13fea82aeeb7e652e4
+size 10039014

nlp_model/variables/variables.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06d21a9640a08890dad97cfa91d3560a1f520ebc177af67f1d9a09499ea3c7cf
+size 1029457957

requirements.txt

@@ -0,0 +1,8 @@
+streamlit==1.3.0
+Pillow == 9.4.0
+nltk == 3.8.1
+numpy == 1.24.3
+pandas == 2.0.3
+tensorflow  == 2.14.0
+regex == 2022.7.9
+joblib