sem_eval_2018_task_1

SemEval-2018 Task 1: Affect in Tweets 专注于从推文中识别情感，目前提供子任务5（E-c）的数据。它包含英文、阿拉伯文和西班牙文三种语言的推文，规模在1K到10K之间，并标注了推文中是否包含愤怒、期待、厌恶等11种情感。数据来源于Twitter用户，通过众包方式进行标注。该数据集支持文本分类任务，但授权许可未知。

SemEval-2018 Task 1: Affect in Tweets focuses on emotion recognition from tweets, and currently provides the dataset for Subtask 5 (E-c). It contains tweet corpora in three languages: English, Arabic, and Spanish, with the number of instances ranging from 1,000 to 10,000. The tweets are annotated with 11 emotion categories including anger, anticipation, disgust, and others. The dataset is sourced from Twitter users and annotated via crowdsourcing. This dataset supports text classification tasks, but its licensing permission is unknown.

搜集汇总

背景与挑战

背景概述

该数据集是一个多语言情感分析数据集，包含英语、阿拉伯语和西班牙语的推文，用于识别11种不同情感。数据通过众包标注，适用于文本分类任务，但目前仅提供子任务5（E-c）的数据。