Skip to content

RAHAMNIabdelkaderseifelislem/arabic_tweets_pretraitment

Repository files navigation

arabic tweets pretraitment

this website contains a pretraitment of arabic tweets to be used in sentiment analysis

it is a website that is built using php and javascript and it uses composer to manage the packages

structure

- tweets_pre/
    - assets/               // the assets of the website
        - css/             // the css files
            - style.css   // the main css file
        - js/              // the javascript files
            - script.js   // the main javascript file
    - Data/                 // the data that i used to test the website
        - 7-tweets.csv       // 7 arabic tweets
        - 50tweet.csv        // more than 50 arabic tweets exactly 54
        - 100tweets.csv      // more than 100 arabic tweets exactly 108
        - 1000tweets.csv     // more than 1000 arabic tweets exactly 1008
    - uploads/               // the uploaded files
        - tokens.csv        // the tokens of the tweets
        - tweets-ar.csv     // the arabic tweets after deleting the english tweets and every thing that is not
    - vendor/                // the composer packages
        - autoload.php      // the composer autoloader
        - composer/         // the composer packages
        - khaled.alshamaa/  // the arabic sentiment analysis API
    - index.php              // the main entry point for the application
    - history.php            // the latest treated tweets page
    - composer.json          // the composer configuration file
    - composer.lock          // the composer lock file

installation

to install the website you need to have composer installed on your machine

# clone the repository
git clone https://github.com/RAHAMNIabdelkaderseifelislem/arabic_tweets_pretraitment.git

# install the composer packages
composer install

usage

to use the website you need to run the following command

php -S localhost:8000

then you can access the website from your browser using the following url

http://localhost:8000

screenshots

first look : first look after upload : full page after upload

history : history page tables :

Original tweets table shown tweets after deleting the english tweets and every thing that is not table shown tokenization of each tweet table shows the sentiment analysis the arabic sentiment analysis table shown

graphs : the number of words per tweet bar chart shown the positive and negative tweets pie chart shown the number of positive and negative words bar chart shown the probability of each tweet to be positive or negative bar chart shown

demo video

credits

About

Arabic tweets pretreatment adding to it sentiment analysis

Resources

Stars

Watchers

Forks

Packages

No packages published