A Curve Binding Method for Parallel Coordinate Visualization Based on Class Attributes

Abstract

The invention discloses a parallel coordinate visual curve binding method based on class attributes, which comprises the following steps: if the original data set does not contain class attributes, the class attributes are obtained by using a clustering method; Center position; calculate the position of the data point on each attribute after it is offset to its class center according to the attraction coefficient; set the binding control point between adjacent attribute axes, and draw the binding curve of the sample in this interval; combine all relative The bound curves between adjacent attribute axes are connected to obtain a complete curve for the data points. The invention effectively reduces the visual clutter caused by the visualization of parallel coordinates when displaying a large amount of data, and provides users with the function of understanding and analyzing data more intuitively by binding and constraining samples of the same category.

Description

technical field [0001] The invention relates to the field of information visualization, in particular to a parallel coordinate visualization curve binding method based on class attributes. Background technique [0002] In the Internet era, the generation and dissemination speed of information has been rapidly developed, especially with the advent of the era of big data, data is increasingly showing a trend of high dimensionality and large capacity. When people are directly faced with a large amount of data, it is usually difficult to grasp the valuable information in it, so as to make new decisions to guide production and life. With the continuous development of visualization technology, people can analyze and explore data conveniently and intuitively by means of graphic images and human-computer interaction. At present, visualization technology has played an important role in many fields such as biology, medicine, physics, chemistry, etc., and has attracted more and more p...

AI Technical Summary

Problems solved by technology

However, as the sample size of the data set continues to increase, clutter such as crossing and overlapping between data axes will become more and more obvious, resulting in poor visualization effects. In severe cases, the data curve will completely cover the entire plane, and the data gap becomes Indivisible, unreadable, making visualization meaningless

Although with the help of human-computer interaction, people can filter the data and display local data operations, but users will not be able to obtain the overall change trend of the data and other valuable information from the original parallel coordinate visualization image of the data.

Images