Update README.md

README.md

@@ -40,42 +40,50 @@ tags:
 - tweet
 - collection
 ---
-
-# Financial Sentiment Analysis Dataset
+# Financial Trading Signals Sentiment Dataset
 
 ## Overview
-This dataset is a comprehensive collection of tweets focused on financial topics, meticulously curated to assist in sentiment analysis in the domain of finance and stock markets. It serves as a valuable resource for training machine learning models to understand and predict sentiment trends based on social media discourse, particularly within the financial sector.
+This dataset contains 4,664 trading signals extracted from Telegram group chats, focused on financial instruments such as Forex pairs, commodities (Gold, Silver), stocks, and indices. Each signal is labeled with a sentiment value for use in financial sentiment analysis and machine learning applications.
 
 ## Data Description
-The dataset comprises tweets related to financial markets, stocks, and economic discussions. Each tweet is labeled with a sentiment value, where '1' denotes a positive sentiment, '2' signifies a negative sentiment, and '0' indicates a neutral sentiment. The dataset has undergone thorough preprocessing, including sentiment mapping and the removal of duplicate entries, to ensure data quality and consistency.
+Each record represents a trading signal and includes fields for symbol, sentiment (both text and integer), action, entry price, target, stop loss, time frame, original text, photo filename, and timestamp. Sentiment is mapped as follows:
+- `0`: Neutral
+- `1`: Bullish
+- `2`: Bearish
 
 ### Dataset Structure
-- **Tweet**: The text of the tweet, providing insights into financial discussions.
-- **Sentiment**: A numerical label indicating the sentiment of the tweet (1 for bullish, 2 for bearish, and 0 for neutral).
+- **symbol**: Trading pair/instrument (e.g., "EURUSD", "GOLD", "GBPUSD")
+- **sentiment**: Market sentiment ("Bullish", "Bearish", "Neutral")
+- **sentiment_int**: Sentiment as integer (0=neutral, 1=bullish, 2=bearish)
+- **action**: Trading action ("BUY", "SELL")
+- **entry_price**: Entry price level
+- **target**: Target price level
+- **stop_loss**: Stop loss price level
+- **time_frame**: Trading timeframe (e.g., "1H", "4H", "12H")
+- **text**: Original signal text from Telegram
+- **photo**: Associated chart image filename
+- **timestamp**: Signal timestamp
 
 ## Dataset Size
-- **Bullish Sentiments**: 17,368
-- **Bearish Sentiments**: 8,542
-- **Neutral Sentiments**: 12,181
-
-## Sources
-This dataset is an amalgamation of data from various reputable sources, each contributing a unique perspective on financial sentiment:
-
-- [FIQA Sentiment Classification](https://huggingface.co/datasets/ChanceFocus/fiqa-sentiment-classification): A sentiment analysis dataset with 721 positive, 379 negative, and 11 neutral sentiments.
-- [Stock Market Tweets Data](https://ieee-dataport.org/open-access/stock-market-tweets-data): A collection of tweets with 523 positive, 420 neutral, and 341 negative sentiments.
-- [Stock Related Tweet Sentiment](https://www.kaggle.com/datasets/mattgilgo/stock-related-tweet-sentiment): A dataset featuring 5005 positive, 741 neutral, and 736 negative sentiments.
-- [Master Thesis Data](https://github.com/moritzwilksch/MasterThesis/tree/main): Includes 3711 positive, 2784 neutral, and 2167 negative sentiments.
-- [Twitter Stock Sentiment](https://github.com/poojathakoor/twitter-stock-sentiment): Comprises 702 positive, 595 negative, and 481 neutral sentiments.
-- [Crypto Sentiment](https://github.com/surge-ai/crypto-sentiment/tree/main): Sentiment data for cryptocurrency-related tweets with 296 positive and 256 negative sentiments.
-- [Stock Sentiment](https://github.com/surge-ai/stock-sentiment/tree/main): Sentiment analysis on stock-related tweets, including 327 positive and 173 negative sentiments.
-- [Stockmarket Sentiment Dataset](https://www.kaggle.com/datasets/yash612/stockmarket-sentiment-dataset): Features 3685 positive and 2106 negative sentiments.
-- [Twitter Financial News Sentiment](https://huggingface.co/datasets/zeroshot/twitter-financial-news-sentiment): Contains 2398 positive, 1789 negative, and 7744 neutral sentiments.
+- **Total Signals**: 4,664
+- **Bullish Sentiments**: 2,026
+- **Bearish Sentiments**: 2,298
+- **Neutral Sentiments**: (calculated from sentiment_int == 0)
 
 ## Usage
-This dataset is ideal for training and evaluating machine learning models for sentiment analysis, especially those focused on understanding market trends and investor sentiment. It can be used for academic research, financial market analysis, and developing AI tools for financial institutions.
-
-## Acknowledgments
-We extend our heartfelt gratitude to all the authors and contributors of the original datasets. Their efforts in data collection and curation have been pivotal in creating this comprehensive resource.
+This dataset is ideal for training and evaluating machine learning models for financial sentiment analysis, market trend prediction, and NLP tasks related to trading signals. It can be used for academic research, financial analytics, and developing AI tools for financial institutions.
 
 ## License
-This dataset is made available under the MIT license, adhering to the licensing terms of the original datasets.
+This dataset is made available under the MIT license.
+
+## Citation
+If you use this dataset in your research, please cite:
+
+```bibtex
+@dataset{telegram_financial_signals_2025,
+  title={Financial Trading Signals Sentiment Dataset},
+  author={JonusNattapong},
+  year={2025},
+  url={https://github.com/JonusNattapong/Telegram-Group-Financial}
+}
+```