BEGIN:VCALENDAR VERSION:2.0 PRODID:-//pretalx//talks.osgeo.org//foss4g-2026//speaker//FTERWY BEGIN:VTIMEZONE TZID:JST BEGIN:STANDARD DTSTART:20000101T000000 RRULE:FREQ=YEARLY;BYMONTH=1 TZNAME:JST TZOFFSETFROM:+0900 TZOFFSETTO:+0900 END:STANDARD END:VTIMEZONE BEGIN:VEVENT UID:pretalx-foss4g-2026-XWZYTY@talks.osgeo.org DTSTART;TZID=JST:20260902T110000 DTEND;TZID=JST:20260902T113000 DESCRIPTION:Geographic Information System (GIS) data is inherently multidim ensional\, encompassing spatial extent\, temporal dynamics\, and diverse a ttributes. Understanding and presenting this complex data through cartogra phic visualization requires both comprehensive geospatial knowledge and so phisticated map design skills. However\, a significant portion of geograph ic platform users—including web developers\, data analysts\, and profess ionals without formal cartography training—frequently encounter substant ial challenges throughout the map visualization workflow. These difficulti es span from initial data interpretation and selection of appropriate visu alization methods to the configuration of critical cartographic elements s uch as color schemes\, symbol sizes\, opacity levels\, and overall composi tional layout.\n\nIn contemporary web mapping systems\, data visualization is predominantly controlled through Map Style JSON\, a structured format that governs data rendering in mapping libraries such as MapLibre GL JS\, which operates according to the MapLibre Style Specification standard. Whi le this specification is open and highly flexible\, creating appropriate s tyles from actual datasets remains a task demanding both technical experti se and design experience. Consequently\, many users invest considerable ti me manually experimenting with style adjustments\, often through trial and error\, which can be both frustrating and inefficient.\n\nThis presentati on proposes an innovative approach to simplifying the map design process t hrough the development of a Model Context Protocol (MCP) for automated Map Style JSON generation from users' spatial vector data. This data can be s ourced through database connections or API service integrations. The funda mental concept underlying MCP is the creation of a "context layer" that en ables language models to systematically understand the structure and seman tic meaning of GIS data before applying this understanding to map style ge neration.\n\nThe proposed architecture integrates MCP with open-source Lar ge Language Models (LLMs) capable of operating locally through Ollama. The system analyzes users' spatial data characteristics\, including geometry types\, attribute structure\, and data distribution patterns. Subsequently \, it automatically generates MapLibre-compliant Map Style JSON that can b e immediately deployed in web mapping applications. This local processing capability addresses both performance and data privacy concerns that often arise with cloud-based solutions.\n\nThe distinctive advantage of this ap proach lies in how MCP extends beyond merely ensuring structurally correct JSON generation. The protocol fundamentally incorporates cartographic des ign principles based on the "Perceptual Properties of Linear and Spatial S ystems" into the decision-making process. This integration manifests in se veral critical ways: the selection of color schemes aligned with data sema ntics\, the assignment of appropriate symbols corresponding to geometry ty pes\, and the strategic application of color tones and opacity levels to e nhance user perception and readability. The system also accommodates datas ets with multiple classification classes and leverages modern color palett es to ensure that spatial data visualization achieves both clarity and acc essibility.\n\nThe technical implementation combines several key component s working in concert. First\, the MCP server acts as an intermediary layer that processes incoming spatial data\, extracting relevant metadata and s tructural information. This includes analyzing coordinate reference system s\, identifying attribute data types\, detecting statistical distributions \, and recognizing spatial patterns that inform styling decisions.\n\nThe language model component\, running locally through Ollama\, receives this contextualized information and applies learned cartographic principles to generate appropriate styling rules. The model has been trained to understa nd the relationships between data characteristics and visual representatio n best practices. For instance\, when encountering categorical data with d istinct classes\, the system automatically selects qualitatively different colors that maximize perceptual distinction. For continuous numerical dat a\, it applies sequential or diverging color schemes appropriate to the da ta's semantic meaning.\n\nThe generated Map Style JSON adheres strictly to MapLibre specifications\, ensuring immediate compatibility with MapLibre GL JS and other compliant rendering engines. The output includes properly structured layers\, sources\, paint properties\, and layout configurations that reflect both the data's inherent characteristics and established car tographic conventions.\n\nA critical innovation of this approach involves embedding cartographic design expertise directly into the generation proce ss. Traditional automated styling systems often produce technically correc t but cartographically naive outputs. This MCP-based system incorporates s everal levels of design intelligence:\n\nPerceptual hierarchy: The system understands which data elements should be visually prominent and adjusts s tyling properties accordingly\, considering factors such as feature import ance\, scale-dependent visibility\, and visual contrast.\n\nColor theory a pplication: Beyond simple color assignment\, the system applies principles of color harmony\, considers color blindness accessibility\, and ensures adequate contrast ratios for legibility across different display condition s.\n\nSymbolic representation : The selection of point symbols\, line patt erns\, and fill styles reflects both the semantic meaning of the data and established cartographic conventions\, making maps intuitively interpretab le even for non-expert users.\n\nScale responsiveness: Generated styles in clude appropriate zoom-level dependencies\, ensuring that map elements app ear at suitable scales and with appropriate levels of detail.\n\nThe objec tives of this presentation extend beyond merely demonstrating automated Ma p Style JSON generation. Fundamentally\, this approach democratizes qualit y map production\, enabling individuals without GIS backgrounds to create professional-quality cartographic visualizations. This democratization has significant implications for data journalism\, civic participation\, educ ational applications\, and small organizations that lack dedicated GIS exp ertise.\n\nFurthermore\, the utilization of open-source models and archite cture capable of local execution through Ollama aligns perfectly with the principles of the Open Geospatial Ecosystem. This approach ensures data so vereignty\, eliminates dependency on proprietary cloud services\, and faci litates integration with other open-source tools prevalent in the geospati al community. The system can be extended and customized by users\, fosteri ng innovation and adaptation to specific domain requirements.\n\nThe proto col-based architecture also enables future enhancements and integrations. As language models continue to evolve\, the MCP layer provides a stable in terface that can leverage improved capabilities without requiring fundamen tal system redesign. Additionally\, the approach can be extended to incorp orate user feedback\, learning from styling preferences and iteratively im proving recommendations.\n\nIn this presentation\, our goal is not only to automatically generate Map Style JSON from your data but\, more important ly\, to empower individuals without GIS expertise to create high-quality m ap visualizations with ease. Additionally\, by utilizing an open model and architecture that can operate on users' devices via Ollama\, we align wit h the principles of an open geospatial ecosystem and enable integration wi th other open-source tools within the community. DTSTAMP:20260717T225739Z LOCATION:Cosmos2 SUMMARY:Towards Automated Map JSON Style from Spatial Vector Data Using MC P - Arissara Sompita URL:https://talks.osgeo.org/foss4g-2026/talk/XWZYTY/ END:VEVENT END:VCALENDAR