Social Media Popularity Dataset

该数据集名为SMPD，包含了38,307名用户发布的305,595篇帖子，并记录了这些帖子的观看次数。数据集还包含了三级层次分类信息：类别、子类别和概念。此外，为了测试模型性能，该数据集被随机抽样为三种规模：100K、200K和300K。每种规模的数据集都按照8:1:1的比例分为训练集、验证集和测试集。规模上，数据集涵盖了38,307名用户的305,595篇帖子。任务方面，旨在进行人气预测。

This dataset is named SMPD. It contains 305,595 posts published by 38,307 users, with the view counts of each post recorded. It also includes three-level hierarchical classification information covering category, subcategory and concept. To evaluate model performance, this dataset was randomly sampled into three scaled variants with sizes of 100K, 200K and 300K respectively. Each scaled variant is split into training, validation and test sets at an 8:1:1 ratio. The core task of this dataset is popularity prediction.

搜集汇总

背景与挑战

背景概述

Social Media Popularity Dataset（SMPD）是一个用于社交媒体人气预测的数据集，包含38,307名用户的305,595篇帖子及其观看次数，并提供了类别、子类别和概念的三级层次分类。该数据集被抽样为100K、200K和300K三种规模，每种规模均按8:1:1比例划分为训练集、验证集和测试集，以支持模型性能测试。