This is a vtuber analysis project ~~~~

Hi every DD~~~

• Java 8(just because my friend only know java)
• python 3.5.6
• Linux 4.19
• mongodb

For contributer:

• update your twi account or google account meta info in meta.json, I will delete this after all finish, and mark this project as open
• plz use vscode editor for py repo, and IntelliJ for java repo
• plz make sure pylint and autopep8 is on, and follower it's code lint and name rule as possible, just for convience of merge code. any white list setting should be shared with us
• use maven not gradle for java user
• use npm not yarn for js dev
• it will be much easier if some one write a DockerFile especially when some one using tensorflow~~~

• /python : python source file
• meta.json: all configuration, Auth info.
• /data: folder for harvested data

Generally speaking, this project has 4 phase

A script can find virtual youtuber according to twitter friends of a known "seed" vtuber.

actualy I get all my vtuber list from third party data source

• vtuber-post
• daifuku data will be saved to /data

Some other related data in daifuku and socialblade still need to be collected (eg. comment of vtuber, income, ranking)

progress: >>>>>> PASSING

create a social network (undirected map with value edge) based on following things and the importance is from high to low

• live together(though youtube api)
• twitter interactive(only one month data will be used)

comment similarity of youtube will alse be collected, this will not be shows in graph but will be use in modeling

progress: >>>>>> FAILING

implememt k-means algorithm to cluster existing networkx-lib with skilearn lib

using java spring boot to just expose the data in mongodb, it is just used as a api-like too for data virtulize

using D3.js to implement a front end page

For mac:

$brew install pyenv
# after you intall set environment variable in your ~/.bash_profile
$pyenv install 3.5.6
$pyenv local 3.5.6
# enter root folder of project
$pip install -r requirements.txt

# if docker is not installed, plz install docker and docker-compose
$docker pull mongo
# make a workspace/data under '~', then back to project folder
# plz tell me if you need my data
$. ./run_mongo.sh

plz use Intellij Idea