Abstract:We introduce two novel visualization designs to support practitioners in performing identification and discrimination tasks on large value ranges (i.e., several orders of magnitude) in time-series data: (1) The order of magnitude horizon graph, which extends the classic horizon graph; and (2) the order of magnitude line chart, which adapts the log-line chart. These new visualization designs visualize large value ranges by explicitly splitting the mantissa m and exponent e of a value v = m * 10e . We evaluate our novel designs against the most relevant state-of-the-art visualizations in an empirical user study. It focuses on four main tasks commonly employed in the analysis of time-series and large value ranges visualization: identification, discrimination, estimation, and trend detection. For each task we analyse error, confidence, and response time. The new order of magnitude horizon graph performs better or equal to all other designs in identification, discrimination, and estimation tasks. Only for trend detection tasks, the more traditional horizon graphs reported better performance. Our results are domain-independent, only requiring time-series data with large value ranges.
What problem does this paper attempt to address?
This paper attempts to solve the visualization problem of dealing with a large range of values (i.e., multiple orders of magnitude) in time - series data. Specifically, existing visualization techniques have deficiencies when dealing with a large range of values, especially performing poorly in identifying, differentiating, and estimating numerical values as well as detecting trends. For example, traditional line charts and horizon graphs are difficult to accurately read and compare small - scale changes when presenting data spanning multiple orders of magnitude.
To solve these problems, the author introduced two new visualization designs:
1. **Order of Magnitude Horizon (OMH)**: It extends the classic horizon graph by separating the mantissa \(m\) and exponent \(e\) of each value and representing different orders of magnitude with different color bands.
2. **Order of Magnitude Line (OML)**: It improves the logarithmic line chart by using a linear mapping within each order of magnitude and combining color coding to support the perception of numerical changes.
These two new designs aim to better handle a large range of values in time - series data and improve user performance in low - level tasks such as identification, differentiation, estimation, and trend detection. Through experimental research, the author evaluated the effectiveness of these new designs compared to existing standard visualization methods.
### Formula Explanation
- A data value \(v\) can be expressed as \(v = m\times10^{e}\), where \(m\) is the mantissa and \(e\) is the exponent.
- In the OMH and OML designs, the mantissa \(m\) and exponent \(e\) are processed differently:
- The mantissa \(m\) is linearly scaled to the y - axis within each order of magnitude.
- The exponent \(e\) is used to determine different order - of - magnitude bands and is represented by color saturation or position.
### Experimental Results
Through a user study of 90 participants, the author found that:
- **OMH** performs excellently in identification, differentiation, and estimation tasks, with the lowest error rate and the highest user confidence.
- **OML** improves user confidence in some tasks, but is not as good as the traditional horizon graph in trend - detection tasks.
- For trend - detection tasks, the traditional horizon graph performs better, perhaps because its linearly - scaled y - axis is easier to understand.
Overall, these two new designs provide significant improvements in dealing with a large range of values in time - series data, especially in identification, differentiation, and estimation tasks.