User-Submitted Congestion Posts

该数据集包含了通过一款交通应用程序在六个月内收集的实际用户提交的拥堵报告。为确保数据质量，已筛选排除任何非服务时间内的记录，并且数据集中不包含任何可识别个人身份的信息。具体规模上，数据集共有2034,779条记录，其中标注记录有10,373条，未标注记录则有2024,406条。该数据集的任务是对火车拥堵情况进行预测。

This dataset contains real user-submitted congestion reports collected over a six-month period via a traffic application. To ensure data quality, any records submitted outside service hours have been filtered out, and the dataset does not contain any personally identifiable information. In terms of scale, the dataset has a total of 2,034,779 records, including 10,373 labeled records and 2,024,406 unlabeled records. The task of this dataset is to predict train congestion.