JiaWu-Repository / CoTrRank_Trust_Evaluation

CoTrRank: Trust Evaluation of Users and Tweets (IJCAI 2019 DEMO)

CoTrRank: Trust Evaluation of Users and Tweets

Click the below video to view the Demo on YouTube.

Results analysis

Twitter_Tas_dataset

This is the dataset used for our research project: Trust Evaluation on Twitter. Anyone who are working on data minning, social network analysis, or any related topics can download this dataset for research purposes.

The proposed CoTrRank method and detailed experimental results can be found in our paper:

" CoTrRank: Trust Evaluation of Users and Tweets" (IJCAI 2019).

The data set is designed for research purpose only. There are four files storing the Twitter data crawled from Twitter using its official REST API.

'tas_users'

This file stores all the users in the Tas dataset. We have given each user a user index and the UserID is their original user ID collected from Twitter. These users are crawled based on the 'location' attribute in their profile. They are considered all from the Tasmania Twitter Community.

For Example:

User_index	UserID
"7232"	"1076277643"
"791"	"1000773896"

'twitter_tweets_all_user_posted'

This file stores all the tweets posted by the users in the Tas dataset. These tweets are crawled from Twitter platform using RestAPI. Please note that, as restricted by Twitter, only a maximum of 3,200 most recent tweets of each user are allowed to be downloaded.

for example:

TweetID	Tweet_index	User_index
"100000183710007296"	"0"	"12423"

Because this file is too big, you can download it from here (you don't need to create a dropbox account to download the file.)

'twitter_follower_friend_index'

This file stores the follow relations between users. Note that, all the numbers are indicating user_index numbers.

for example:

Follower_index	Friend_index
"0"	"180"

(User_index 0 is following User_index 180)

'twitter_tweets_relation_index'

This file stores the retweet and reply relations between tweets. Note that, all numbers are indicating tweet_index numbers.

first column: original_tweet_index | second column: retweet/reply tweet_index | third column: relation type (1 for reply, 2 for retweet)

for example:

original_tweet_index	retweet/reply tweet_index	relation type
"35"	"40"	"2"
"95"	"103"	"1"

Here in relation type, 1 is for reply, 2 is for retweet. (tweet_index 35 is retweeted by tweet_index 40, and tweet_index 95 is replied by tweet_index 103)

'twitter_user_mention_index'

This file stores the mention relation. User is mentioned by a tweet.

first column: mentioned User_index | second column: tweet_index

for example:

mentioned User_index	tweet_index
"3"	"1856312"

(user_index 3 is mentioned by tweet_index 1856312)

Note that, due to the request from Twitter, we can not publish the content of Twitter.

Ground Truth User Idx

In the paper, we have detailed the method used to derive the ground truth user lists, which is provided in the 'GT_UserIDs.csv' file. Each value is the corresponding user idx defined for the Tas user set.

References

If you use this data set for research, please cite our paper:

@inproceedings{ijcai2019-950,
  title     = {CoTrRank: Trust Evaluation of Users and Tweets},
  author    = {Li, Peiyao and Zhao, Weiliang and Yang, Jian and Wu, Jia},
  booktitle = {Proceedings of the Twenty-Eighth International Joint Conference on
               Artificial Intelligence, {IJCAI-19}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},             
  pages     = {6542--6544},
  year      = {2019},
  month     = {7},
  doi       = {10.24963/ijcai.2019/950},
  url       = {https://doi.org/10.24963/ijcai.2019/950},
}

In addition, if you are interested in the topic of Social Network trust, you may want to know another work of ours: "CoRank A Coupled Dual Networks Approach to Trust Evaluation on Twitter"

If you have any questions, please do not hesitate to contact me.