EquityMood

EquityMood is a real-time sentiment analysis dashboard for the Indonesian stock market (IDX/IHSG). It aggregates financial news from 14+ Indonesian media RSS feeds, runs NLP-based sentiment scoring, and overlays the results with live market data from Yahoo Finance — all in one interactive interface.

Built with Python, Streamlit, and Plotly. No subscriptions. No paid APIs. Just open data.

---

Features

• Real-Time News Aggregation — Pulls articles from CNBC Indonesia, Kontan, Bisnis Indonesia, Detik Finance, Kompas, Tempo, Republika, and more.
• Dual Sentiment Engine — Lexicon-based analysis for speed; optional IndoBERT deep learning model for accuracy.
• Live Market Data — Real-time price, daily change, high/low, and volume via Yahoo Finance.
• Candlestick Chart — 5-day OHLCV candlestick visualization for the searched ticker.
• Multi-Ticker Comparison — Side-by-side candlestick charts and price cards for up to 4 stocks simultaneously.
• Sentiment Visualizations — Donut chart, gauge meter, word cloud, treemap, scatter plot, and historical trend.
• Auto-Refresh — Configurable auto-refresh timer (1, 2, 5, or 10 minutes) with a live countdown overlay.
• News Log & CSV Export — Full article table with sentiment badges, scores, and one-click CSV download.

---

Tech Stack

Layer	Tools
Interface	Streamlit
Charting	Plotly
NLP (Lexicon)	Custom Indonesian financial lexicon
NLP (Deep Learning)	IndoBERT via Transformers
Market Data	yfinance
Feed Parsing	feedparser + BeautifulSoup
Word Cloud	wordcloud

---

Project Structure

equitymood/
├── app.py                  # Main Streamlit application
├── feed_parser.py          # RSS feed aggregator (14 sources)
├── sentiment_analyzer.py   # Lexicon + IndoBERT sentiment engine
├── market_data.py          # Yahoo Finance real-time price wrapper
├── templates.py            # HTML/SVG component templates
├── style.css               # Global dark theme stylesheet
└── requirements.txt

---

Quickstart

1. Clone the repository

git clone https://github.com/yourusername/equitymood.git
cd equitymood

2. Install dependencies

pip install -r requirements.txt

For IndoBERT support, PyTorch must be installed separately. See pytorch.org for platform-specific instructions.

3. Run the app

streamlit run app.py

The dashboard will be available at http://localhost:8501.

---

Usage

1. Enter a stock ticker (e.g. BBRI, TLKM, GOTO) or the keyword IHSG in the sidebar.
2. Use the quick-select buttons at the top for common tickers.
3. Switch between Dashboard, Perbandingan (Comparison), and Log Berita (News Log) from the sidebar.
4. Toggle IndoBERT for higher-accuracy sentiment at the cost of slightly longer load times.
5. Enable Auto-Refresh to keep the data current automatically.

---

News Sources

Portal	Category
CNBC Indonesia	Market, Investment, Economy
Kontan	News, Investment, Macro
Bisnis Indonesia	General Finance
Detik Finance	General Finance
Liputan6 Bisnis	Business News
Kompas Money	Economy
Okezone Finance	Finance
Sindonews Ekonomi	Economy
Republika Ekonomi	Economy
Tempo Bisnis	Business

---

Notes

• All news data is sourced from publicly available RSS feeds.
• Market data is fetched via the unofficial Yahoo Finance API (yfinance). Data may be delayed by 15 minutes.
• IndoBERT inference runs locally on your machine — no external API calls are made for the NLP model.
• The lexicon-based analyzer is tuned for Indonesian financial text and runs without GPU requirements.

---

License

MIT License. See LICENSE for details.