Emotion Visualization Comment Service is project for practice making a web service.
-
Renewal Front : It was refactoring version of 'ev-commentService'. It is based on React.
-
Renewal Backend : It was refactoring version of 'ev-commentService'. It is based on NodeJS.
-
ev-commentService : PHP Service for providing component service.
-
ev-backend : It is based on Django and includes an emotional analysis module based on konlpy and tensorflow.
-
ev-sentimentalAnalysis : This is a comment analysis module within the server.
All svg files used in project are downloaded at featherIcon
- Overall dependency
react
reactstrap
ramda
'Jekyll-Script.html' is Client's component file that allows users to easily add 'ev-comment-service' to their own Jekyll based blogs.
The procedure for registering on a Jekyll based blog is as follows.
1 - Signup and signin https://evcommentservice.ga, Register your blog's URL to service.
The form of the URL must be the same as right. (https:{domain}.ga)
2 - In your Jekyll blog, Add the following settings to the '_config.yml'.
User ID is the same one that joined above.
ev_username: {User ID}
3 - Download and include Jekyll-Script.html to your post layout or the page you want.
(Add {% include EV-Script.html %} or paste the script directly)
4 - Create div tag that iframe tag will be it's child, and set it's id 'EV-Start'
Bloggers can decide whether or not to apply an emotional analysis service to his blog comment service and how it will be styled.
The usage is as follows.
(below ev_mode is 'full' for default)
Set one of the follwing values in the 'ev_mode' in your _config.yml
ev_username: { full, binary, none, debug }
full : Depending on the negative-positive degree of the comments,
comment's color style is divided into 10 stages
binary : Depending on the negative-positive degree of the comments,
comment's color style is divided into 3 stages (Positive, Negative, Neutral)
none : Emotional analysis styles were not indicated.
debug : Do not use emotional analysis service