VLSP 2020 Dataset

该数据集包含了用于越南社交网络上的可靠情报识别的训练样本以及公共和私人测试集。数据集存在不平衡问题，其中不可靠类别在训练集中约占17%。数据集包括编码ID、文本内容、点赞数、分享数、评论数以及照片数量等特征。规模方面，训练样本有4,372个，公共和私人测试集各有1,642个样本。任务的目的是进行可靠情报的识别。

This dataset comprises training samples, as well as public and private test sets, for reliable information identification on Vietnamese social networks. The dataset exhibits class imbalance, where the unreliable category accounts for approximately 17% of the training set. It includes features such as encoded ID, text content, number of likes, shares, comments, and photos. In terms of scale, the training set contains 4,372 samples, while the public and private test sets each have 1,642 samples. The task objective is reliable information identification.