instafake-dataset
instafake-dataset copied to clipboard
Dataset for Intagram Fake and Automated Account Detection
InstaFake Dataset
Dataset of the Intagram Fake and Automated Account Detection paper
Installation
Install miniconda
https://conda.io/en/latest/miniconda.html
Setup a CONDA environment
We create a new vitrual environment named instafake
.
conda create --name instafake python=3.6
Activating the virtual environment
If you are inside the virtual environment, your shell prompt should look like: (instafake) user@computer:~$
If that is not the case, you can enable the virtual environment using:
conda activate instafake
To deactivate the virtual environment, use:
conda deactivate
Install required packages
To install the required packages, run the following command in your instafake
virtual environment:
pip install -r requirements.txt
Import Datasets as Dataframes
To import the fake and automated datasets as pandas dataframes, simply define the dataset folder path data
, and dataset version dataset_version
and call import_data
from utils
:
from utils import import_data
dataset_path = "data"
dataset_version = "fake-v1.0"
fake_dataset = import_data(dataset_path, dataset_version)
dataset_path = "data"
dataset_version = "automated-v1.0"
automated_dataset = import_data(dataset_path, dataset_version)
Dataset Structures
The dataset contains of 2 set of json files with given features:
Fake Account Detection
-
user_media_count
- Total number of posts, an account has. -
user_follower_count
- Total number of followers, an account has. -
user_following_count
- Total number of followings, an account has. -
user_has_profil_pic
- Whether an account has a profil picture, or not. -
user_is_private
- Whether an account is a private profile, or not. -
user_biography_length
- Number of characters present in account biography. -
username_length
- Number of characters present in account username. -
username_digit_count
- Number of digits present in account username. -
is_fake
- True, if account is a spam/fake account, False otherwise
Automated Account Detection
-
user_media_count
- Total number of posts, an account has. -
user_follower_count
- Total number of followers, an account has. -
user_following_count
- Total number of followings, an account has. -
user_has_highlight_reels
- Whether an account has at least one highlight reel present, or not. -
user_has_url
- Whether an account has an url present in biography, or not. -
user_biography_length
- Number of characters present in account biography. -
username_length
- Number of characters present in account username. -
username_digit_count
- Number of digits present in account username. -
media_comment_numbers
- Total number of comments for a given media. -
media_comments_are_disabled
- Whether given media is closed for comments, or not. -
media_has_location_info
- Whether given media includes location, or not. -
media_hashtag_numbers
- Total number of hashtags, given media has. -
media_upload_times
- Media upload timastamps. -
automated_behaviour
- True, if account is an automated account, False otherwise
Dataset Metadata
The following table is necessary for this dataset to be indexed by search engines such as Google Dataset Search.
property | value | ||||||
---|---|---|---|---|---|---|---|
name | InstaFake Dataset: An Instagram fake and automated account detection dataset |
||||||
alternateName | InstaFake Dataset |
||||||
url | https://github.com/fcakyon/instafake-dataset |
||||||
sameAs | https://github.com/fcakyon/instafake-dataset |
||||||
sameAs | https://github.com/fcakyon/instafake-dataset |
||||||
description | The InstaFake Dataset is comprised of anonymized Instagram user data collected by Fatih Cagatay Akyon and Esat Kalfaoglu over the second half of 2018. We’re releasing this dataset publicly to aid the research community in making advancements in machine learning based social media analysis. |
||||||
provider |
|
||||||
license |
|