Hate speech detection is one of the most important aspects of event identification during political events like invasions. In the case of hate speech detection, the event is the occurrence of hate speech, the entity is the target of the hate speech, and the relationship is the connection between the two. Since multimodal content is widely prevalent across the internet, the detection of hate speech in text-embedded images is very important. Given a text-embedded image, this task aims to automatically identify the hate speech and its targets. This task will have two subtasks.
The goal of this task is to identify whether the given text-embedded image contains hate speech or not. The text-embedded images, which are the dataset for this subtask, will have annotations for the prevalence of hate speech.
The goal of this subtask is to identify the targets of hate speech in a given hateful text-embedded image. The text-embedded images are annotated for "community", "individual" and "organization" targets.
@inproceedings{bhandari2023crisishatemm,
title={CrisisHateMM: Multimodal Analysis of Directed and Undirected Hate Speech in Text-Embedded Images from Russia-Ukraine Conflict},
author={Bhandari, Aashish and Shah, Siddhant Bikram and Thapa, Surendrabikram and Naseem, Usman and Nasim, Mehwish},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
year={2023}
}
All the images have a unique identifier called "index". The labels for training data are organized in the folder provided. For evaluation and testing, the submission format is mentioned below.
Subtask 1
The script takes one prediction file as the input. Your submission file must be a JSON file which is then zipped. We will only take the first file in the zip folder, so do not zip multiple files together. Make sure that your hate label is given as "1" and non-hate label is given as "0".
IMPORTANT: The index in json should be in ascending order.
{"index": 45805, "prediction": 0}
{"index": 20568, "prediction": 1}
{"index": 30987, "prediction": 0}
A sample file is available here. Also, make sure that the index order in the submission file in JSON should be in ascending order. The JSON file basically tells what image (with a unique index) is given what label.
The script takes one prediction file as the input. Your submission file must be a JSON file which is then zipped. We will only take the first file in the zip folder, so do not zip multiple files together. Make sure that your individual, community, and organization labels are given as "0", "1", and "2" respectively.
IMPORTANT: The index in json should be in ascending order. Also, make sure that the index order in the submission file in JSON should be in ascending order.
{"index": 45865, "prediction": 0}
{"index": 23568, "prediction": 1}
{"index": 36987, "prediction": 2}
A sample file is available here. The JSON file basically tells what image (with a unique index) is given what label. Also, make sure that the index order in the submission file in JSON should be in ascending order.
For both Subtasks, the performance will be ranked by F1 score.
More about evaluation can be found here.
To submit the files, name your submission as submission.json and zip it with the file name, ref.zip. Make sure that the zip does not have any sub-directories. Windows users can right-click on the JSON file and save it as a zip.
The "Evaluation Phase" is meant for competitors to familiarize themselves with the Codalab site. We provide a Train set for training, and Dev set for testing. Therefore, competitors are not allowed to use the dev set for training in this phase.
In the "Testing Phase", competitors may feel free to incorporate the dev set for training. However, the test set should not be incorporated into training.
The use of external dataset are permitted, but they should not violate any other terms and conditions. You should also mention your usage in your paper write-up.
Organizers of the competition might choose to publicize, analyze and change in any way any content sent as a part of this task. Whenever appropriate academic citation for the sending group would be added (e.g. in a paper summarizing the task).
The organizers are free to penalize or disqualify for any violation of the above rules or for misuse, unethical behavior, or other behaviors they agree are not accepted in a scientific competition in general and in the specific one at hand.
The participants should agree to the terms and conditions.
Training & Evaluation data available: Nov 1, 2023
Test data available: Nov 30, 2023
Test start: Nov 30, 2023
Test end: Jan 7, 2024
System Description Paper submissions due: Jan 13, 2024
Notification to authors after review: Jan 26, 2024
Camera ready: Jan 30, 2024
CASE Workshop: 21-22 Mar, 2024
If there are any questions related to the competition, please contact surendrabikram@vt.edu
Start: Nov. 1, 2023, midnight
Description: Develop and train your system, and try evaluating on development data.
Start: Nov. 30, 2023, midnight
Description: Run the trained system on test data and upload predictions for leaderboard scoring.
Start: Nov. 1, 2023, midnight
Description: Develop and train your system, and try evaluating on development data.
Start: Nov. 30, 2023, midnight
Description: Run the trained system on test data and upload predictions for leaderboard scoring.
Jan. 7, 2024, 11:59 p.m.
You must be logged in to participate in competitions.
Sign In# | Username | Score |
---|---|---|
1 | Yestin | 1.2500 |
2 | AhmedElSayed | 1.7500 |
3 | Sadiya_Puspo | 3.0000 |