The designed chatbot follows the guides for connecting the APIs of Telegram, Wolframalpha and Twitter.

1. The idea behind this project is to show the conversational skills of the Wolframalpha artificial intelligence by means of Telegram. It can be also used as a bot for data inquiries to search in both Wolframalpha and Wikipedia search engines.
2. This bot is also designed for predicting the real risk transmitted by a Twitter post with a logistic regression estimator.

This bot is currently available in telegram and mounted on AWS-EC2 free instance.

1. Main Diagram

2. Installation requirements

3. API Basic configuration

4. Conversational results

5. Emergency predictor sample

A diagram of the implementation is shown below:

Install or upgrade python-telegram-bot by the command:

$ pip install python-telegram-bot --upgrade

Wolframalpha libraries are required. Install them with:

$ pip install wolframalpha

Install twitter libraries by:

$ pip install tweepy

After creating a bot in telegram with the help of @BotFather. Insert your token in the line:

updater =  Updater('', use_context=True) # Insert here API code of telegram

Moreover, you should be registered as Wolframalpha developer, then insert your token in line:

app_id = '' # Insert here API code of wolframalpha 
client = wolframalpha.Client(app_id)

Finally, you should log in as Twitter developer, then insert your keys in lines:

consumer_key = ""
consumer_secret = ""
access_token = ""
access_token_secret = ""
# authorization of consumer key and consumer secret
auth = tweepy.OAuthHandler(consumer_key, consumer_secret)
# set access to user's access key and access secret 
auth.set_access_token(access_token, access_token_secret)
# calling the api 
api = tweepy.API(auth)