Map

Overview

The Map panel is the main visualization space in Mantis, displaying points, bags, and clusters as a spatial embedding. It provides an interactive, high-level view of your data landscape, letting you navigate, select, and explore conceptual relationships between items.

Purpose

• Visually explore the overall structure and relationships within your dataset.
• Identify clusters, patterns, or outliers through spatial proximity.
• Select and highlight points, bags, or groups for further analysis.
• Navigate large data spaces efficiently with panning and zooming.

Key Features

• Interactive Embedding: Points are displayed according to their semantic or feature-based similarity.
• Selection Tools: Click, lasso, or rectangle-select points directly on the map.
• Bags and Clusters: Visualize and differentiate groups with color-coding and labels.
• Real-Time Updates: The map updates instantly with changes in filters, selections, or analysis parameters.
• Zoom and Pan: Move across the data landscape for focused or global views.
• Hover Details: Hover over points to see key metadata or feature values.

Tips

• Use lasso or box select to quickly group and bag points of interest.
• Zoom in to inspect dense areas or outliers in detail.
• Click on points or clusters to highlight and synchronize with other panels.