A heuristic multi-criteria classification approach incorporating data quality information for choropleth mapping

Area of Science:

• Cartography
• Geographic Information Science (GIS)
• Data Visualization

Background:

• Choropleth map classification faces challenges with statistical indifference, leading to misrepresentation.
• Existing methods like class separability often result in unbalanced classes.
• Advancements in cartography and technology have not fully resolved these classification issues.

Purpose of the Study:

• To propose a novel heuristic classification approach for choropleth maps.
• To balance class separability with other criteria like evenness and intra-class variability.
• To develop a geovisual-analytic tool for evaluating classification trade-offs.

Main Methods:

• Developed a heuristic classification approach considering class separability, evenness, and intra-class variability.
• Created a geovisual-analytic package to support the mapping process.
• Enabled adjustment of class break values to optimize classification performance.

Main Results:

• The heuristic approach produces more balanced classes compared to methods focusing solely on separability.
• The geovisual-analytic package effectively supports the evaluation of trade-offs between classification criteria.
• Adjustable class breaks allow for improved classification performance and map interpretability.

Conclusions:

• The proposed heuristic classification method offers a more robust solution for choropleth map design.
• Balancing multiple criteria leads to more informative and visually effective choropleth maps.
• Geovisual-analytic tools are crucial for interactive and optimized map classification.