Project Lucy plus Twitter Sentiment Analysis

Description

Project Lucy is basically a collection of programs which interact with the Twitter API. We chose to particularly focus on creating a Twitter bot as well as performing sentiment analysis. All this achieved using Python library tweepy as well as textblob for the latter.

Project Lucy:

• User communicating with Lucy:

  • Lucy will automatically reply to every hashtag containing #kanye, #dict and #help when tweeted
  • Lucy will analyse tweets for sentiment sentiment analysis
  • Lucy will produce a plot to represent the result; a graph of th enumber of likes/ retweets a user gets in a span of one week.
  • Lucy is able to update a status
  • Lucy is able to upload a picture

Features

• Word definition
• Random Quote
• Summary plot of most liked tweets and most retweets
• Sentiment Analysis

Behavior Driven Development

Input	Behaviour	Output
Mentioning @Lucy with #help	Lucy replying the tweet with a set of instructions on how to interact with her.	Display the tweet tweeted by lucy in response to the sender with instructions.
Mentioning @Lucy #dict followed by any word	Lucy will reply to your tweet with a definition of the word	Display the tweet tweeted by lucy in response to the sender. If the word is misplet, Lucy will recommend a word with the closest match to the one typed.
Mentioning @Lucy with the hashtag #Kanye / #ye / #KanyeWest	Lucy will reply with a random quote from Kanye West	Display the tweet tweeted by lucy in response to the sender.
The username and tweet limit	A table af tweet summary will be displayed	A plot of most liked tweets and most retweets as well as sentiment analysis table will be shown

Built with

- Python 3
- Tweepy (Python wrapper for the Twitter API.)
- Pandas

Requirements

• This program requires python3.+ (and pip) installed, a guide on how to install python on various platforms can be found here
• Twitter Consumer Key, Consumer Secret, Access Token and Access Token Secret which can be got by applying at developer.twitter.com

Installation

• Step 1 : Clone this repository using git clone https://github.com/collinsmuriuki/blog.git, or downloading a ZIP file of the code.
• Step 2 : The repository, if downloaded as a .zip file will need to be extracted to your preferred location and opened
• Step 3 : Go to the project root directory and install the virtualenv library using pip an afterwards create a virtual environment. Run the following commands respectively:
  • pip install virtualenv
  • virtualenv venv
  • source venv/bin/activate
    • Note that you can exit the virtual environment by running the command deactivate
• Step 4: Install all the dependencies by running the commmand
  • pip install -r requirements.txt
• Step 5: Create a run.sh file in which you will export the Api key/ secret and Access token/ secret as environmental variables. Make the file executable by running the command chmod a+x run.sh
  • To use the bot, type the command python3 bot.py
    • Login to Twitter to interact with it. To use sentiment analysis, type the command python3 user_analysis.py
    • Interact with the termainal to obtain results

Sample Results

Twitter Sentiment Analysis Test

Testing most liked last 200 tweets of @realDonaldTrump

Testing most retweeted last 200 tweets of @realDonaldTrump

Testing both most liked and retweeted last 200 tweets of @realDonaldTrump

Testing testing sentiment analysis for @eminem last 10 tweets

Known bugs

• App crashes randomly while listening for tweets and hashtags, looking into this
• App crashes once it responds to a tweet constaining the hashtag meme, as a result we have disabled this feature temporarily

Author

• Mick Ganza
• Collins Muriuki
• Isaac Gichuru
• Titus Ouko

Contact

In any case do not hesitate to reach the team at:

• murerwacollins@gmail.com
• ganzamick@gmail.com
• titusouko@gmail.com
• Official1offmark@gmail.com
• 7248zack@gmail.com

Licence

This project is Licensed under GNU GPL-v3.0