Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,40 +1,425 @@
-import altair as alt
-import numpy as np
-import pandas as pd
+# === ClimatePulse: Chatbot Analisis Opini Publik ===
+import torch
 import streamlit as st
+import pandas as pd
+import pydeck as pdk
+import altair as alt
+from transformers import pipeline, AutoTokenizer, AutoModelForTokenClassification, AutoModelForSequenceClassification
+from geopy.geocoders import Nominatim
+from datetime import datetime
+import os
+
+# === Setup Halaman ===
+st.set_page_config(page_title="ClimatePulse", layout="centered")
+
+# === Load Model & Pipeline ===
+device = 0 if torch.cuda.is_available() else -1
+
+# Sentimen
+sent_tokenizer = AutoTokenizer.from_pretrained("mdhugol/indonesia-bert-sentiment-classification")
+sent_model = AutoModelForSequenceClassification.from_pretrained("mdhugol/indonesia-bert-sentiment-classification")
+pipe_sent = pipeline("sentiment-analysis", model=sent_model, tokenizer=sent_tokenizer)
+
+# Emosi
+pipe_emo = pipeline("sentiment-analysis", model="azizp128/prediksi-emosi-indobert", device=device)
+
+# NER
+ner_tokenizer = AutoTokenizer.from_pretrained("cahya/bert-base-indonesian-NER")
+ner_model = AutoModelForTokenClassification.from_pretrained("cahya/bert-base-indonesian-NER")
+pipe_ner = pipeline("ner", model=ner_model, tokenizer=ner_tokenizer, aggregation_strategy="simple")
+
+label_map = {'LABEL_0': 'Positif', 'LABEL_1': 'Netral', 'LABEL_2': 'Negatif'}
+
+# === Custom Dark Mode Style + Logo ===
+page_bg = '''
+<style>
+[data-testid="stAppViewContainer"] {
+    background-color: #0e1525;
+    color: white;
+}
+[data-testid="stHeader"] {
+    background-color: rgba(0,0,0,0);
+}
+[data-testid="stSidebar"] > div:first-child {
+    background-color: #1f2937;
+}
+.block-container {
+    padding-top: 2rem;
+    padding-bottom: 2rem;
+    font-family: "Segoe UI", sans-serif;
+}
+h1, h2, h3, h4, h5 {
+    font-family: 'Segoe UI', sans-serif;
+    color: #10B981;
+}
+.stButton>button {
+    background-color: #10B981;
+    color: white;
+    border-radius: 8px;
+    padding: 0.5rem 1rem;
+    font-size: 1rem;
+    border: none;
+}
+.stTextInput>div>div>input, .stTextArea>div>textarea {
+    background-color: #1f2937;
+    color: white;
+    border-radius: 6px;
+    border: 1px solid #374151;
+}
+</style>
+'''
+st.markdown(page_bg, unsafe_allow_html=True)
+
+# === Judul Halaman dengan Logo di Sebelah Teks ===
+col1, col2 = st.columns([1, 8])
+with col1:
+    st.image("logo.png", width=60)
+    
+with col2:
+    st.markdown("""
+    <div style='display: flex; flex-direction: column; justify-content: center;'>
+        <h4 style='color: #10B981; margin-bottom: 0;'>ClimatePulse - Analisis Opini SDG 13</h4>
+        <h1 style='color: white; margin-top: 0;'>Perubahan Iklim di Media Sosial</h1>
+        <p style='color: gray;'>Telusuri opini publik, sentimen, emosi, dan entitas terkait kebijakan dan bencana iklim</p>
+    </div>
+    """, unsafe_allow_html=True)
+
+# === Form Input User ===
+with st.form(key="input_form"):
+    text_input = st.text_area("Input Teks / Tweet", placeholder="Contoh: PLTN dibangun di Papua, saya takut dan kecewa", height=120)
+    submit = st.form_submit_button("🔍 ANALISIS")
+
+# === Analisis dan Visualisasi Lain Tetap ===
+# (seluruh isi kode berikutnya tetap seperti sebelumnya)
+# === Tidak ditampilkan ulang agar tidak duplikasi ===
+
+
+
+# === Analisis dan Visualisasi Lain Tetap ===
+# (seluruh isi kode berikutnya tetap seperti sebelumnya)
+# === Tidak ditampilkan ulang agar tidak duplikasi ===
+
+
+# === Proses Analisis Tunggal ===
+if submit and text_input.strip():
+    with st.spinner("Menganalisis opini publik..."):
+        sent = pipe_sent(text_input)[0]
+        sent_label = label_map.get(sent['label'], sent['label'])
+        emo = pipe_emo(text_input)[0]['label'].capitalize()
+        ner = pipe_ner(text_input)
+        ents = [e['word'] for e in ner]
+
+        lokasi_kunci = [
+    # === Wilayah Umum / Pulau ===
+    "sumatera", "jawa", "kalimantan", "sulawesi", "papua", "maluku", "nusa tenggara", "kepulauan seribu",
+
+    # === Nama Provinsi Lengkap (38) ===
+    "aceh", "sumatera utara", "sumatera barat", "riau", "kepulauan riau", "jambi", "bengkulu",
+    "sumatera selatan", "bangka belitung", "lampung",
+    "banten", "dki jakarta", "jawa barat", "jawa tengah", "daerah istimewa yogyakarta", "jawa timur",
+    "bali", "nusa tenggara barat", "nusa tenggara timur",
+    "kalimantan barat", "kalimantan tengah", "kalimantan selatan", "kalimantan timur", "kalimantan utara",
+    "sulawesi utara", "sulawesi tengah", "sulawesi selatan", "sulawesi tenggara", "gorontalo", "sulawesi barat",
+    "maluku", "maluku utara",
+    "papua", "papua barat", "papua selatan", "papua tengah", "papua pegunungan", "papua barat daya",
+
+    # === Ibu Kota Provinsi ===
+    "banda aceh", "medan", "padang", "pekanbaru", "tanjungpinang", "jambi", "bengkulu",
+    "palembang", "pangkalpinang", "bandar lampung",
+    "serang", "jakarta", "bandung", "semarang", "yogyakarta", "surabaya",
+    "denpasar", "mataram", "kupang",
+    "pontianak", "palangka raya", "banjarmasin", "samarinda", "tarakan",
+    "manado", "palu", "makassar", "kendari", "gorontalo", "mamuju",
+    "ambon", "ternate",
+    "jayapura", "manokwari", "merauke", "nabire", "wamena", "fakfak", "sorong", "timika",
+
+    # === Kota/Kabupaten Besar atau Strategis ===
+    "bekasi", "bogor", "depok", "tangerang", "cirebon", "tegal", "purwokerto", "solo", "magelang",
+    "malang", "kediri", "sidoarjo", "pasuruan", "probolinggo", "lumajang", "blitar", "jember",
+    "banyuwangi", "cilacap", "padangsidimpuan", "binjai", "sibolga", "lubuklinggau", "palopo",
+    "parepare", "bitung", "tomohon", "kotamobagu", "kotabaru", "pangkalan bun", "ketapang",
+    "palu", "baubau", "karangasem", "buleleng", "labuan bajo", "ende", "bima", "dompu",
+
+    # === Lokasi Baru / Khusus / Otorita ===
+    "nusantara",  # Ibu kota negara baru di Kaltim
+    "penajam paser utara", "balikpapan", "samarinda", "bontang",  # Kaltim area
+    "kepri", "ntb", "ntt", "kaltim", "kalteng", "kalsel", "kalbar", "kaltara",  # singkatan populer
+
+    # === Lokasi Adat/Kultural (yang sering disebut) ===
+    "minangkabau", "batak", "dayak", "asmat", "ambon", "bugis", "toraja", "sunda", "madura", "tapanuli"
+]
+
+        locs = []
+        for e in ner:
+            ent_text = e['word'].lower()
+            if e['entity_group'] == 'LOC':
+                locs.append(e['word'])
+            else:
+                for keyword in lokasi_kunci:
+                    if keyword in ent_text:
+                        locs.append(keyword.capitalize())
+        locs = list(set(locs))
+
+        geolocator = Nominatim(user_agent="climatepulse")
+        geo_locs = []
+        for loc in locs:
+            try:
+                location = geolocator.geocode(loc)
+                if location:
+                    geo_locs.append({
+                        'lokasi': loc,
+                        'lat': location.latitude,
+                        'lon': location.longitude,
+                        'jumlah': 1
+                    })
+            except:
+                continue
+
+        now = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+        df_log_single = pd.DataFrame([{
+            "timestamp": now,
+            "text": text_input,
+            "sentimen": sent_label,
+            "emosi": emo
+        }])
+
+        log_file = "log_tren.csv"
+        if os.path.exists(log_file):
+            pd.concat([pd.read_csv(log_file), df_log_single]).to_csv(log_file, index=False)
+        else:
+            df_log_single.to_csv(log_file, index=False)
+
+
+    emoji_map = {
+    "Senang": "😊",
+    "Sedih": "😢",
+    "Marah": "😡",
+    "Takut": "😨",
+    "Kecewa": "😞",
+    "Netral": "😐"
+}
+    # === Tampilkan Hasil ===
+    
+    st.markdown(f"""
+<div style='background-color: #1f2937; padding: 1rem; border-radius: 10px;'>
+    <h3 style='color: white;'>Hasil Analisis</h3>
+    <p><b>Sentimen:</b> <span style='color: red;'>{sent_label}</span>  |  
+       <b>Emosi:</b> <span style='color: #facc15;'>{emo} {emoji_map.get(emo, '')}</span></p>
+    <p><b>📍 Lokasi:</b> {', '.join(locs) or "Tidak ditemukan"}</p>
+    <p><b>🔖 Entitas:</b> {', '.join(ents) or "Tidak ditemukan"}</p>
+</div>
+""", unsafe_allow_html=True)
+
+   # === Tambahan: Peta Opini Publik berdasarkan Log ===
+if os.path.exists("log_tren.csv"):
+    df_log = pd.read_csv("log_tren.csv")
+    lokasi_kunci = [
+    # === Wilayah Umum / Pulau ===
+    "sumatera", "jawa", "kalimantan", "sulawesi", "papua", "maluku", "nusa tenggara", "kepulauan seribu",
+
+    # === Nama Provinsi Lengkap (38) ===
+    "aceh", "sumatera utara", "sumatera barat", "riau", "kepulauan riau", "jambi", "bengkulu",
+    "sumatera selatan", "bangka belitung", "lampung",
+    "banten", "dki jakarta", "jawa barat", "jawa tengah", "daerah istimewa yogyakarta", "jawa timur",
+    "bali", "nusa tenggara barat", "nusa tenggara timur",
+    "kalimantan barat", "kalimantan tengah", "kalimantan selatan", "kalimantan timur", "kalimantan utara",
+    "sulawesi utara", "sulawesi tengah", "sulawesi selatan", "sulawesi tenggara", "gorontalo", "sulawesi barat",
+    "maluku", "maluku utara",
+    "papua", "papua barat", "papua selatan", "papua tengah", "papua pegunungan", "papua barat daya",
+
+    # === Ibu Kota Provinsi ===
+    "banda aceh", "medan", "padang", "pekanbaru", "tanjungpinang", "jambi", "bengkulu",
+    "palembang", "pangkalpinang", "bandar lampung",
+    "serang", "jakarta", "bandung", "semarang", "yogyakarta", "surabaya",
+    "denpasar", "mataram", "kupang",
+    "pontianak", "palangka raya", "banjarmasin", "samarinda", "tarakan",
+    "manado", "palu", "makassar", "kendari", "gorontalo", "mamuju",
+    "ambon", "ternate",
+    "jayapura", "manokwari", "merauke", "nabire", "wamena", "fakfak", "sorong", "timika",
+
+    # === Kota/Kabupaten Besar atau Strategis ===
+    "bekasi", "bogor", "depok", "tangerang", "cirebon", "tegal", "purwokerto", "solo", "magelang",
+    "malang", "kediri", "sidoarjo", "pasuruan", "probolinggo", "lumajang", "blitar", "jember",
+    "banyuwangi", "cilacap", "padangsidimpuan", "binjai", "sibolga", "lubuklinggau", "palopo",
+    "parepare", "bitung", "tomohon", "kotamobagu", "kotabaru", "pangkalan bun", "ketapang",
+    "palu", "baubau", "karangasem", "buleleng", "labuan bajo", "ende", "bima", "dompu",
+
+    # === Lokasi Baru / Khusus / Otorita ===
+    "nusantara",  # Ibu kota negara baru di Kaltim
+    "penajam paser utara", "balikpapan", "samarinda", "bontang",  # Kaltim area
+    "kepri", "ntb", "ntt", "kaltim", "kalteng", "kalsel", "kalbar", "kaltara",  # singkatan populer
+
+    # === Lokasi Adat/Kultural (yang sering disebut) ===
+    "minangkabau", "batak", "dayak", "asmat", "ambon", "bugis", "toraja", "sunda", "madura", "tapanuli"
+]
+    lokasi_counter = {}
+    for text in df_log['text']:
+        for keyword in lokasi_kunci:
+            if keyword in text.lower():
+                lokasi = keyword.capitalize()
+                lokasi_counter[lokasi] = lokasi_counter.get(lokasi, 0) + 1
+
+    geo_locs = []
+    geolocator = Nominatim(user_agent="climatepulse-map")
+    for lokasi, jumlah in lokasi_counter.items():
+        try:
+            location = geolocator.geocode(lokasi)
+            if location:
+                geo_locs.append({
+                    'lokasi': lokasi,
+                    'lat': location.latitude,
+                    'lon': location.longitude,
+                    'jumlah': jumlah
+                })
+        except:
+            continue
+
+    if geo_locs:
+        map_df = pd.DataFrame(geo_locs)
+        st.markdown("### 🗺️ Peta Opini Publik")
+        st.pydeck_chart(pdk.Deck(
+            map_style=None,
+            initial_view_state=pdk.ViewState(latitude=-2.5, longitude=117.0, zoom=4, pitch=0),
+            layers=[
+                pdk.Layer(
+                    "ScatterplotLayer",
+                    data=map_df,
+                    get_position='[lon, lat]',
+                    get_color='[255, 100, 100, 160]',
+                    get_radius='jumlah * 10000',
+                    pickable=True,
+                    auto_highlight=True
+                )
+            ],
+            tooltip={"text": "{lokasi}: {jumlah} opini"}
+        ))
+    else:
+        st.info("❗ Tidak ada lokasi yang berhasil dipetakan dari histori log.")
+
+
+    st.markdown("### 📈 Tren Waktu Sentimen")
+    if os.path.exists("log_tren.csv"):
+        df_log = pd.read_csv("log_tren.csv")
+        df_log['timestamp'] = pd.to_datetime(df_log['timestamp'])
+        df_log['tanggal'] = df_log['timestamp'].dt.date
+        trend_all = df_log.groupby(['tanggal', 'sentimen']).size().reset_index(name='jumlah')
+        chart = alt.Chart(trend_all).mark_line(point=True).encode(
+            x='tanggal:T',
+            y='jumlah:Q',
+            color='sentimen:N'
+        ).properties(width=600)
+        st.altair_chart(chart, use_container_width=True)
+
+# === Upload CSV untuk Analisis Massal ===
+st.markdown("---")
+st.markdown("### 📥 Analisis CSV Massal")
+uploaded_file = st.file_uploader("Upload file CSV berisi kolom 'text'", type=["csv"])
+
+if uploaded_file is not None:
+    df_csv = pd.read_csv(uploaded_file)
+    st.write("Pratinjau Data:", df_csv.head())
+
+    if "text" in df_csv.columns:
+        result_data = []
+        geo_locs = []
+        log_rows = []
+        lokasi_kunci = [
+    # === Wilayah Umum / Pulau ===
+    "sumatera", "jawa", "kalimantan", "sulawesi", "papua", "maluku", "nusa tenggara", "kepulauan seribu",
+
+    # === Nama Provinsi Lengkap (38) ===
+    "aceh", "sumatera utara", "sumatera barat", "riau", "kepulauan riau", "jambi", "bengkulu",
+    "sumatera selatan", "bangka belitung", "lampung",
+    "banten", "dki jakarta", "jawa barat", "jawa tengah", "daerah istimewa yogyakarta", "jawa timur",
+    "bali", "nusa tenggara barat", "nusa tenggara timur",
+    "kalimantan barat", "kalimantan tengah", "kalimantan selatan", "kalimantan timur", "kalimantan utara",
+    "sulawesi utara", "sulawesi tengah", "sulawesi selatan", "sulawesi tenggara", "gorontalo", "sulawesi barat",
+    "maluku", "maluku utara",
+    "papua", "papua barat", "papua selatan", "papua tengah", "papua pegunungan", "papua barat daya",
+
+    # === Ibu Kota Provinsi ===
+    "banda aceh", "medan", "padang", "pekanbaru", "tanjungpinang", "jambi", "bengkulu",
+    "palembang", "pangkalpinang", "bandar lampung",
+    "serang", "jakarta", "bandung", "semarang", "yogyakarta", "surabaya",
+    "denpasar", "mataram", "kupang",
+    "pontianak", "palangka raya", "banjarmasin", "samarinda", "tarakan",
+    "manado", "palu", "makassar", "kendari", "gorontalo", "mamuju",
+    "ambon", "ternate",
+    "jayapura", "manokwari", "merauke", "nabire", "wamena", "fakfak", "sorong", "timika",
+
+    # === Kota/Kabupaten Besar atau Strategis ===
+    "bekasi", "bogor", "depok", "tangerang", "cirebon", "tegal", "purwokerto", "solo", "magelang",
+    "malang", "kediri", "sidoarjo", "pasuruan", "probolinggo", "lumajang", "blitar", "jember",
+    "banyuwangi", "cilacap", "padangsidimpuan", "binjai", "sibolga", "lubuklinggau", "palopo",
+    "parepare", "bitung", "tomohon", "kotamobagu", "kotabaru", "pangkalan bun", "ketapang",
+    "palu", "baubau", "karangasem", "buleleng", "labuan bajo", "ende", "bima", "dompu",
+
+    # === Lokasi Baru / Khusus / Otorita ===
+    "nusantara",  # Ibu kota negara baru di Kaltim
+    "penajam paser utara", "balikpapan", "samarinda", "bontang",  # Kaltim area
+    "kepri", "ntb", "ntt", "kaltim", "kalteng", "kalsel", "kalbar", "kaltara",  # singkatan populer
+
+    # === Lokasi Adat/Kultural (yang sering disebut) ===
+    "minangkabau", "batak", "dayak", "asmat", "ambon", "bugis", "toraja", "sunda", "madura", "tapanuli"
+]
+
+        geolocator = Nominatim(user_agent="climatepulse")
+
+        for i, row in df_csv.iterrows():
+            text = str(row["text"])
+            sent = pipe_sent(text)[0]
+            sent_label = label_map.get(sent['label'], sent['label'])
+            emo = pipe_emo(text)[0]['label'].capitalize()
+            ner = pipe_ner(text)
+            ents = [e['word'] for e in ner]
+
+            locs = []
+            for e in ner:
+                ent_text = e['word'].lower()
+                if e['entity_group'] == 'LOC':
+                    locs.append(e['word'])
+                else:
+                    for keyword in lokasi_kunci:
+                        if keyword in ent_text:
+                            locs.append(keyword.capitalize())
+            locs = list(set(locs))
+
+            for loc in locs:
+                try:
+                    location = geolocator.geocode(loc)
+                    if location:
+                        geo_locs.append({
+                            'lokasi': loc,
+                            'lat': location.latitude,
+                            'lon': location.longitude,
+                            'jumlah': 1
+                        })
+                except:
+                    continue
+
+            now = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+            log_rows.append({"timestamp": now, "text": text, "sentimen": sent_label, "emosi": emo})
+
+            result_data.append({
+                "text": text,
+                "sentimen": sent_label,
+                "emosi": emo,
+                "entitas": ", ".join(ents)
+            })
+
+        df_result = pd.DataFrame(result_data)
+        st.success("Analisis selesai!")
+        st.dataframe(df_result)
+
+        csv_download = df_result.to_csv(index=False).encode('utf-8')
+        st.download_button("📥 Download Hasil CSV", csv_download, "hasil_analisis.csv", "text/csv")
+
+        log_file = "log_tren.csv"
+        df_log_append = pd.DataFrame(log_rows)
+        if os.path.exists(log_file):
+            pd.concat([pd.read_csv(log_file), df_log_append]).to_csv(log_file, index=False)
+        else:
+            df_log_append.to_csv(log_file, index=False)
+
 
-"""
-# Welcome to Streamlit!
-
-Edit `/streamlit_app.py` to customize this app to your heart's desire :heart:.
-If you have any questions, checkout our [documentation](https://docs.streamlit.io) and [community
-forums](https://discuss.streamlit.io).
-
-In the meantime, below is an example of what you can do with just a few lines of code:
-"""
-
-num_points = st.slider("Number of points in spiral", 1, 10000, 1100)
-num_turns = st.slider("Number of turns in spiral", 1, 300, 31)
-
-indices = np.linspace(0, 1, num_points)
-theta = 2 * np.pi * num_turns * indices
-radius = indices
-
-x = radius * np.cos(theta)
-y = radius * np.sin(theta)
-
-df = pd.DataFrame({
-    "x": x,
-    "y": y,
-    "idx": indices,
-    "rand": np.random.randn(num_points),
-})
-
-st.altair_chart(alt.Chart(df, height=700, width=700)
-    .mark_point(filled=True)
-    .encode(
-        x=alt.X("x", axis=None),
-        y=alt.Y("y", axis=None),
-        color=alt.Color("idx", legend=None, scale=alt.Scale()),
-        size=alt.Size("rand", legend=None, scale=alt.Scale(range=[1, 150])),
-    ))
+