commit1

app.py

@@ -0,0 +1,10 @@
+import streamlit as st
+import eda
+import predict
+
+navigation = st.sidebar.selectbox('pilih halaman :', ('EDA',"Prediction"))
+
+if navigation =='EDA':
+    eda.run()
+else:
+    predict.run()

formalisasi.txt

@@ -0,0 +1,7 @@
+def formalize_words(text, word_formalization_dict):
+    pattern = re.compile('\\b(' + '|'.join(kamus.keys()) + ')\\b')
+
+    def replace_words(match):
+        return kamus[match.group(0)]
+    formalized_text = pattern.sub(replace_words, text)
+    return formalized_text

predict.py

@@ -0,0 +1,50 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+import tensorflow as tf
+import pickle
+import tensorflow_hub as hub
+
+
+hub_layer = hub.KerasLayer("https://tfhub.dev/google/nnlm-id-dim128-with-normalization/2",
+                           input_shape=[], dtype=tf.string,)
+tf.keras.utils.get_custom_objects()['KerasLayer'] = hub_layer
+nlp = tf.keras.models.load_model('sentiment_PPKM_NLP_Model.h5')
+
+with open('formalisasi.txt', 'r') as file:
+        function_text = file.read()
+with open("proses text.txt",'r') as file1:
+        proc = file1.read()
+with open('kamus.txt','rb') as file2:
+        kamus = pickle.load(file2)
+
+def create_function_from_text(function_tex):
+    # Gunakan pustaka exec() untuk mengevaluasi teks menjadi objek fungsi
+    exec(function_tex, globals())
+    # Ambil fungsi yang telah dibuat dan kembalikan sebagai output
+    return list(globals().values())[0]
+
+formalize_words = create_function_from_text(function_text)
+text_process = create_function_from_text(proc)
+
+def run():
+    with st.form('key=sentiment prediction'):
+        sentiment = st.text_input('ketikkan tweet anda:', '')
+
+        submitted = st.form_submit_button('Predict')
+
+    data_inf = {
+        'sentiment': sentiment
+    }
+    data_inf = pd.DataFrame([data_inf])
+
+    # Menerapkan fungsi klasifikasi sentimen menggunakan apply pada DataFrame
+    data_inf['proc_sentiment'] = data_inf['sentiment'].apply(text_process)
+
+    if submitted:
+        y_pred_inf = [np.argmax(pred) for pred in nlp.predict(data_inf['proc_sentiment'])]
+        st.write('sentiment : ', str(y_pred_inf))
+
+
+if __name__ == '__main__':
+    run()

proses text.txt

@@ -0,0 +1,18 @@
+def text_proses(teks):
+    teks = teks.lower()
+    teks = re.sub('@[A-Za-z0-9_]+', ' ', teks)
+    teks = re.sub('#[A-Za-z0-9_]+', ' ', teks)
+    teks = re.sub('\\\\n', ' ', teks)
+    teks = teks.strip()
+    teks = re.sub('http\\S+', ' ', teks)
+    teks = re.sub('www.\\S+', ' ', teks)
+    teks = re.sub("[^A-Za-z\\s']", ' ', teks)
+    teks = re.sub('rt', ' ', teks)
+    teks = re.sub('\\b\\w{1,2}\\b', '', teks)
+    teks = re.sub('(\\w)\\1{2,}', '\\1', teks)
+    teks = re.sub('\\b(wk+|aok+|ha+h[aeiou]+|hi+h[aeiou]+|ho+h[aeiou]+)\\b', 'haha', teks)
+    tokens = word_tokenize(teks)
+    teks = ' '.join([word for word in tokens if word not in stpwds_id])
+    teks = formalize_words(teks, formalize_words)
+    teks = stemmer.stem(teks)
+    return teks

requirements.txt

@@ -0,0 +1,8 @@
+streamlit
+pandas
+seaborn
+matplotlib
+plotly
+Pillow
+numpy
+scikit-learn==1.2.2

sentiment_PPKM_NLP_Model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1575ff10a1ae6f5e060d4913bc0819fdcb452aa67db9b86c22f26a26bc13cfe2
+size 450240560