Telegram_Create_crypto_sentiment_bot.md

Template request | Bug report | Generate Data Product

Tags: #telegram #sentiment #bot #naas_drivers #ai #investors

Author: Yaswanthkumar GOTHIREDDY

Description: This notebook provides instructions on how to create a Telegram bot that tracks and analyzes sentiment around cryptocurrencies.

Input

Install packages

!pip install python-telegram-bot --user

Import libraries

import logging
from telegram.ext import *
import numpy as np
from naas_drivers import newsapi
from naas_drivers import sentiment
from datetime import datetime, timedelta

Variables

Create a Telegram bot and get your API KEY

TELEGRAM_API_KEY = "***********"

Get Newsapi API KEY

NEWSAPI_API_KEY = "***********"

# Set up the logging
logging.basicConfig(
    format="%(asctime)s - %(name)s - %(levelname)s - %(message)s", level=logging.INFO
)
logging.info("Starting Bot...")

Get last 24 hours articles from NewsAPI on a CoinMarketCap top coins

# parameters for newsapi
FROM_DATE = (datetime.now() - timedelta(days=1)).replace(microsecond=0).isoformat()
TO_DATE = datetime.now().replace(microsecond=0).isoformat()

# top marketcap coins according to coinmarketcap
coins = [
    "bitcoin",
    "ethereum",
    "ripple",
    "dogecoin",
    "cardano",
    "polygon",
    "binance",
    "polkadot",
    "uniswap",
    "litecoin",
    "chainlink",
    "solana",
    "stellar",
]

# connect to Newsapi Naas drivers
newsapi.connect(NEWSAPI_API_KEY)
newsapi.connect()

Model

User need to ask for coins ticker and get responses

# user_response
def get_response(user_message):

    usr_msg_text = user_message.lower()

    if usr_msg_text in coins:  # check if returned message has the top coins
        data = newsapi.get(
            q=usr_msg_text, from_param=FROM_DATE, to=TO_DATE, language="en"
        )
        if len(data) == 0:  # extra check to returned data
            sentiment = "No news for this ticker in last 24 hours, try another"
            return sentiment
        else:
            sentiment_df = sentiment.get(
                data, column_name="title"
            )  # sentiment calculation
            sentiment = sentiment_df["Score"].mean()
    else:
        sentiment = "There is no data for this ticker\n use \help command"

    return sentiment

Here are the commands to communicate with Telegram

# commands
def start_command(update, context):
    update.message.reply_text(
        """Hello there! I\'m a crypto sentiment bot.
    Send me the coin name, I\'ll send you sentiment!"""
    )


def help_command(update, context):
    update.message.reply_text(
        """check the coin names from the list below!
    bitcoin\n etherium\n ripple\n dogecoin\n cardano\n polygon
    binance\n polkadot\n uniswap\n litecoin\n chainlink\n solana\n stellar"""
    )


# message handler
def handle_message(update, context):
    text = str(update.message.text).lower()
    logging.info(f"User ({update.message.chat.id}) says: {text}")

    # Bot response
    sentiment_naas = get_response(text)
    update.message.reply_text(sentiment_naas)


# error logs
def error(update, context):
    # Logs errors
    logging.error(f"Update {update} caused error {context.error}")

Output

Start the bot

# Run the programme

# create an object "bot"
updater = Updater(TELEGRAM_API_KEY, use_context=True)
dp = updater.dispatcher

#  bot's command handlers
dp.add_handler(CommandHandler("start", start_command))
dp.add_handler(CommandHandler("help", help_command))

# bot's text handlers
dp.add_handler(MessageHandler(Filters.text, handle_message))

# bot's error handler
dp.add_error_handler(error)


# Run the bot
updater.start_polling(1.0)
updater.idle()

Info: the cell above will stay alive until you press stop, the bot will not respond anymore once the cell is stoped.