MDataEE:多因素时间序列数据的分析与可视化

MDataEE:Analysis and Visualization of Multifactor Time Series Data

多因素时间序列数据及异常数据的可视化对于提高决策分析效率等问题具有十分重要的意义.由于不同种类数据具有不同的特征,传统的可视化方法在绘制此类数据时会出现图像复杂、用户观察效率低的情况.为此,提出一种高效探索多因素时间序列数据及异常数据的可视化方法MDataEE.首先,使用可视化映射简化多种类数据的视图;其次,根据数据的密度和重要性以及视觉感知来优化坐标轴的绘制;最后,增加了一些交互操作,通过图像显隐及生成对比视图等操作,方便用户根据需求自由探索不同方面的数据.在真实PM2.5数据集上进行的实验结果表明,与传统可视化方法相比,所提出的方法能够生成简洁且易于分析的可视化视图,在分析异常数据变化的趋势及原因等方面更有优势,可提高用户理解并分析异常的多因素时间序列数据的效率.

The visualization of multifactor time series data and abnormal data is of great significance in improving the decision-making efficiency and other problems.Since different types of data have different characteristics,traditional visualization methods will face the challenges of complex graphics and low user observation efficiency when drawing such data.Therefore,an efficient visualization method MDataEE for exploring multifactor time series data and abnormal data is proposed.Firstly,the view of multiple kinds of data is simplified by visual mapping.Secondly,the rendering of coordinate axes is optimized according to the density and importance of data and visual perception.Finally,we added some interactive operations,such as displaying and hiding graphics and generating contrast views,so that users could explore different aspects of data freely according to their needs.The real-world PM_(2.5) data set is used for the experimental tests in this paper.The results show that the proposed method can generate a concise visual view,which has advantages in analyzing the trend and causes of abnormal data,and can improve the efficiency in understanding and analyzing abnormal multifactor time series data.