Cell-Based Line Charts for Time Series Visualization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for visualizing multiple time series data in charts often result in occlusion, making it difficult to read and interpret due to overlapping curves, especially when many time series are represented, causing some curves to be obscured and rendering the chart undecipherable.
Innovation Solution
The use of cell-based line charts with four dimensions – time, categories, attribute values represented by colors, and an ordering attribute – allows for clear visualization of multiple time series without occlusion, where each time series is represented as a row of cells, with missing data values indicated by gaps, and ordering by attribute values to highlight significant data points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If multiple time series are represented as curves in the same chart, then the chart can display many data series simultaneously, but the chart becomes difficult to read due to occlusion caused by multiple curves crossing over each other
Solution Approach 1:
The patent divides the visualization into multiple separate horizontal lines, with each line containing a subset of time series curves. This segmentation prevents occlusion by ensuring curves within each line do not overlap excessively, while still allowing comprehensive display of many time series across multiple lines.
Solution Approach 2:
The patent introduces a new vertical dimension (line position) in addition to the traditional time (horizontal) and attribute value (vertical within line) dimensions. This creates a four-dimensional visualization space where time series can be organized across multiple horizontal lines, effectively adding a categorical dimension to separate and organize the data series.
2Quantity of substance
If many curves are drawn in the same chart, then all time series can be visualized, but some curves become invisible due to occlusion and bunching near the bottom part of the chart
Solution Approach 1:
By segmenting time series across multiple horizontal lines, the patent ensures that curves are distributed more evenly across the visualization space. This prevents bunching near the bottom and ensures that curves representing time series with relatively low attribute values remain visible and distinguishable.
Solution Approach 2:
The introduction of the line dimension provides an additional spatial coordinate for organizing time series. This allows curves that would otherwise overlap and become invisible in a two-dimensional chart to be separated along the line dimension, maintaining visibility of all time series including those with low attribute values.
3Quantity of substance
If conventional chart techniques are used to visualize a large number of time series, then all data can be included, but the chart becomes undecipherable due to occlusion
Solution Approach 1:
The patent segments the large set of time series into multiple manageable groups displayed on separate horizontal lines. This segmentation reduces the visual complexity within each line, making individual curves easier to track and interpret, while still presenting the complete dataset across the entire visualization.
Solution Approach 2:
By adding the line dimension as an organizational structure, the patent transforms an undecipherable two-dimensional chart into a multi-level visualization where the line dimension serves as an additional index. This allows users to navigate and interpret large numbers of time series more easily by leveraging the extra dimensional organization.
Data Source
AI summary
A plurality of time series having corresponding sets of data over time are received. Visual indicators for display in a display region are presented for values of at least a first attribute of the plurality of time series in corresponding cell-based lines, where the cell-based lines have cells to display the corresponding visual indicators for the values of the at least first attribute over time. The plurality of time series in the display region are ordered according to an aggregate of at least one attribute of the plurality of time series.