A Lightweight WebGIS Visualization Platform for Historical and Cultural Heritage Based on Multi-Source Data Fusion

Abstract

The digital preservation and dissemination of historical and cultural heritage is a pivotal area at the intersection of digital humanities and geographic information science. To address the challenges of multi-source heterogeneity, limited dimensionality, and inadequate public engagement, this study designed and implemented an interactive visualization platform using modern Web technologies. Taking the Leshan Confucian Temple (religious heritage) and the former site of Wuhan University (educational heritage) as case studies, the platform integrates four types of heterogeneous data (geospatial coordinates, architectural attributes, visitor behavioral records, and multimedia imagery) into a unified spatiotemporal information model. Core technical implementations are built upon a lightweight front-end stack including the Gaode Map JavaScript API for geographic visualization, ECharts for dynamic statistical charting, and the Tailwind CSS framework for a fully responsive front-end interface. Key interactive features encompass linked map markers with contextual information windows, user-driven chart filtering, and paginated loading of cultural relic cards. Evaluation results demonstrate that the platform achieves cross-device response delay ≤3 s, supports spatially grounded, dynamic, and presentation of cultural heritage information, and attains a System Usability Scale (SUS) score of 82.5. This work offers a lightweight, scalable technical solution for advancing digital recording and public communication of historical and cultural heritage, while contributing to the theoretical discourse on spatial narrative and multi-source data integration in digital humanities.

1. Introduction

The digital protection and dissemination of historical and cultural heritage have become a critical research topic in the interdisciplinary fields of Digital Humanities and geographic information science, with great potential for cultural heritage recording, public education, and tourism innovation [1,2]. With technological advancements, traditional display modes struggle to meet public demands for efficient access to historical and cultural spatial information and immersive experiences. Utilizing spatial information and digital technologies to convert diverse spatial information carriers into digital formats and to construct highly interactive, information-rich visualization platforms has emerged as an effective approach to enhance the communication effectiveness of historical and cultural resources and to promote public participation [3,4,5]. Furthermore, Web-based visualization technology has become a key medium for digital communication of cultural heritage due to its intuitiveness, accessibility, and strong interactivity [6,7,8].

The rapid development of WebGIS, data visualization, and front-end engineering has made the construction of integrated, multidimensional cultural heritage digital platforms both an urgent need and a research direction of significant theoretical and practical value. Current cultural heritage presentation approaches suffer from three notable shortcomings: (1) Fragmented Information and Linear Narratives: Cultural information (spatial location, historical context, architectural details, related events) is often presented in isolated segments, lacking effective spatiotemporal correlation and logical integration, which hinders the public from forming a systematic and contextualized understanding; (2) Limited Spatial Awareness and Immersion: Conventional media fail to convey the spatial layout, environmental relationships, and geographical evolution of heritage sites, preventing audiences from fully grasping site distribution, scale, or spatial dynamics and reducing the sense of place and historical presence; (3) Weak Interactivity and Passive Engagement: Most existing displays adopt a “one-way transmission” model, where users passively receive information without mechanisms for active exploration, filtering, comparison, or in-depth interaction, failing to meet the growing demand for personalized, immersive cultural experiences [9,10,11].

From a theoretical perspective, this study engages with two foundational concepts in digital humanities: spatial narrative and multi-source data fusion. Spatial narrative theory emphasizes the role of geographic space as a storytelling medium, where locations, trajectories, and spatial relationships convey historical and cultural meanings [12]. By integrating geospatial visualization with user behavior data, this platform enables a form of “spatial storytelling” that reveals how heritage sites are experienced and interpreted over time. Multi-source data fusion theory, originally developed in sensor networks and remote sensing, provides a framework for combining heterogeneous data types to generate more complete and accurate representations of complex phenomena [13]. In the context of cultural heritage, data fusion operates at three levels: data-level fusion (e.g., aligning images with coordinates), feature-level fusion (e.g., linking attributes across sources), and decision-level fusion (e.g., integrating all information to support cultural understanding). This study applies this tripartite framework to heritage data, illustrating how the integration of geospatial coordinates, attribute texts, imagery, and user behavior data can support a more contextualized and spatially grounded presentation of cultural heritage information. While a fully coherent cultural narrative that integrates temporal sequences, causal relationships, and user-driven storytelling requires further semantic integration, the platform provides a foundational step by enabling users to explore heritage sites through linked spatial and statistical views.

As heritage sites with considerable historical and educational value, the Leshan Confucian Temple (religious architectural heritage) and the former campus of Wuhan University (wartime educational legacy) embody diverse information with rich spatiotemporal relevance. However, current digital practices for these sites either focus on single data type display (e.g., pure maps or images) or rely on expensive, complex 3D modeling technologies [14,15]. Deficiencies remain in lightweight deployment, multi-source integration, real-time interaction, and cross-device adaptability. For public-oriented Web platforms, the urgent challenge is to organically integrate geospatial data, historical attribute descriptions, visitation behavior trajectories, and multimedia resources to provide an intuitive, smooth, interactive experience. Based on the above background and existing limitations, this study focuses on the following two core research questions:

• How can heterogeneous, multi-source heritage data be effectively fused and organized within a lightweight WebGIS architecture?
• How can interactive visualization design improve public usability and engagement for cultural heritage-oriented WebGIS platforms?

To address these issues, this study designs and implements a modern Web-based historical and cultural spatial information visualization platform. Focusing on the Leshan Confucian Temple and the former Wuhan University site, the platform employs the Gaode Maps JavaScript API for geospatial visualization, integrates ECharts for dynamic representation of user behavior data, and utilizes the Tailwind CSS framework to create a fully responsive front-end interface. These components enable seamless interaction across map displays, data charts, heritage item cards, and project descriptions. Through a modular, decoupled architecture, the platform enhances multi-source heritage data integration efficiency and overall user interaction experience, offering a lightweight, scalable technical pathway for the digital dissemination of similar cultural heritage. The main contributions of this paper are as follows:

• Proposes and implements an integrated visualization framework for multi-source historical and cultural spatial data, clarifying the integration mechanisms for geographic information, attribute text, behavioral data, and media resources.
• Designs and develops a complete Web platform with map interaction, dynamic charts, paginated loading, and responsive layout, providing the public with an intuitive, convenient entry point for cultural exploration.
• Verifies the platform’s feasibility and effectiveness in information integration, interaction fluency, and multi-terminal adaptation through functional testing and performance evaluation.

The rest of this paper is organized as follows. Section 2 reviews the theoretical foundations, technological evolution, and conceptual framework. Section 3 describes the case studies, data fusion, platform architecture, and functional modules. Section 4 presents the platform evaluation, including functional, usability, performance, and comparative analyses. Section 5 discusses the contributions, limitations, and future directions. Section 6 concludes the paper.

2. Related Work

Building a historical and cultural visualization platform requires integrating knowledge of geographic information systems [16], data fusion theory [17], front-end engineering [12], and collaborative development [18]. The core theories, key technologies, and conceptual framework supporting the design and implementation of the platform are systematically reviewed and elaborated.

2.1. Theoretical Foundations

Spatial narrative theory arises from the intersection of narratology, human geography, and digital media studies [12,19]. It challenges the traditional view of space as a neutral, passive backdrop for events, instead emphasizing that spatial structures, including paths, boundaries, landmarks, clusters, and relational proximity, carry inherent narrative potential. In cultural heritage research, this perspective has been increasingly adopted to interpret how people experience and make sense of historic sites. Studies have applied spatial narrative to exhibition design [20], digital storytelling platforms [18], and locative media experiences [21], demonstrating how spatial arrangements can shape the unfolding of historical meaning. Despite these advances, existing applications often rely on fixed narrative structures or immersive 3D environments, leaving a gap for lightweight, web-based systems that support dynamic, user-centered spatial narratives.

Multi-source data fusion provides a formal framework for organizing and integrating heterogeneous information from distinct sources [22]. The widely recognized three-level hierarchy (data-level, feature-level, decision-level fusion) has become a foundational paradigm in remote sensing, intelligent systems, and geospatial analysis. Within cultural heritage, fusion is frequently employed implicitly, for instance by overlaying historical maps on modern imagery or combining photographs with archival records. However, many implementations lack an explicit hierarchical structure, and more systematic approaches often depend on heavy server-side frameworks or complex semantic architectures [15,23]. Consequently, a practical gap persists for accessible, browser-based fusion strategies that do not require advanced GIS expertise or dedicated backend infrastructure.

These two theoretical strands collectively establish the intellectual basis for the present study. Spatial narrative theory informs how heritage meaning can be expressed geographically and experienced interactively, while multi-source data fusion provides a structured methodology for unifying diverse heritage data. Together, they motivate the design principles and technical architecture developed in this work, which aim to address the identified gaps through a lightweight, client-oriented WebGIS framework.

2.2. Evolution of Methods and Technologies in WebGIS for Cultural Heritage

The development of WebGIS for cultural heritage can be broadly categorized into three evolutionary phases, reflecting shifts in technological capability and application priorities. During the early phase (2000–2010), systems focused on geospatial query and basic map visualization, typically relying on proprietary server-side GIS platforms [24]. These tools established the feasibility of online heritage documentation but offered limited interactivity and accessibility. The middle phase (2010–2020) was marked by the widespread adoption of open-source mapping libraries such as Leaflet and OpenLayers [25,26]. This period emphasized improved user experience, multimedia integration, and cross-platform compatibility, enabling more engaging and publicly oriented heritage dissemination. Nevertheless, most systems remained tied to backend databases and lacked structured approaches to heterogeneous data integration. In recent years (2020–present), research has trended toward 3D visualization, semantic web integration, and linked open data [15,27]. While these approaches enhance realism and knowledge connectivity, they often introduce significant computational overhead, deployment complexity, and barriers to entry for small institutions or non-specialist users. In this trend, a parallel but underdeveloped direction emphasizes lightweight client-side execution, minimal reliance on the backend, and broad usability for public audiences.

Existing representative platforms, including the UNESCO World Heritage Journeys platform and Leaflet-based regional heritage systems, still depend on backend services or dedicated geospatial databases. Few studies have fully explored browser-only architectures that maintain systematic data fusion and spatial narrative capabilities without server dependency. This study situates itself within this burgeoning field, investigating the trade-offs and potential benefits of lightweight design for disseminating accessible cultural heritage by eliminating backend dependencies and using only browser-based technologies (Gaode Map API, ECharts, Tailwind CSS).

2.3. Conceptual Framework

Building on the preceding theoretical foundations and technological review, a three-layer conceptual framework is proposed to structure the overall research design (Figure 1). The framework establishes a clear logical chain from theoretical underpinnings to design decisions and eventual implementation.

The uppermost theory layer consists of spatial narrative theory and multi-source data fusion theory, which together define the disciplinary grounding and analytical perspective of the study. Spatial narrative directs the experiential and geographic expression of heritage content, while multi-source data fusion provides the formal structure for organizing heterogeneous information. The middle design principles layer translates abstract theory into actionable design guidelines. Three core principles are derived: (1) Spatial storytelling, which embeds cultural and historical meaning within geographic representations. (2) Behavioral–spatial linkage, which connects user interactions and movement patterns to spatial entities. (3) Multi-level fusion, which organizes heterogeneous data integration across structural levels. The bottom implementation layer operationalizes these principles into concrete functional modules, including map display, data visualization, heritage information cards, and project overview components.

This framework ensures that technology selection is theoretically justified rather than made arbitrarily through empirical or impromptu decisions. Furthermore, it offers a replicable and potential model for designing lightweight WebGIS applications within the realm of cultural heritage.

2.4. Responsive Web Development Based on Tailwind CSS

The selection of Web development technologies and tools critically impacts the implementation difficulty and presentation effect of historical and cultural spatial information visualization platforms [28,29,30]. This study adopts a mainstream front-end technology stack to build an integrated platform integrating map display, data visualization, and cultural relic browsing. Through responsive Web design, the website adapts to different screen sizes, providing a consistent, smooth cross-device experience. Tailwind CSS is selected as the core interface construction tool: adhering to a practicality-first principle, it offers rich, fine-grained functional classes and built-in responsive breakpoint modifiers, improving development efficiency by 35% compared to traditional CSS development [31,32]. Its style-structure separation principle enhances code maintainability and customization flexibility. Based on Tailwind CSS, the overall user interface is rapidly developed and optimized, ensuring a good browsing experience on both desktop and mobile devices, and laying a solid foundation for cross-terminal visualization of multi-source data.

3. Platform Design and Implementation

3.1. Case Study Sites

Two heritage sites with distinct characteristics were selected to validate the platform: the Leshan Confucian Temple (religious architectural heritage) and the former site of Wuhan University in Leshan (wartime educational heritage). Leshan Confucian Temple is located in Shizhong District, Leshan City, Sichuan Province. Established in the Tang Dynasty (618–907 AD), the existing structures predominantly reflect Ming and Qing architectural styles. It is recognized as one of the best-preserved Confucian temples in Sichuan, representing traditional Chinese religious and educational culture. The former Site of Wuhan University in Leshan comprises eight surviving structures scattered across Leshan, where Wuhan University relocated during the Second Sino-Japanese War (1938–1946). These buildings include dormitories, science and engineering colleges, and a commemorative monument. This site represents modern Chinese educational heritage and wartime resilience. These two cases were selected because they represent different heritage types (religious vs. educational) and spatial patterns (clustered vs. dispersed), allowing the platform to demonstrate its adaptability to diverse heritage contexts. Figure 2 shows the geographical locations and representative images of these sites.

3.2. Data Collection and Fusion Processing

Effective collection and fusion of multi-source data is the primary premise of platform construction. Data in this study come from field research records (verified by on-site investigators) and supplementary public data (sourced from authoritative platforms such as Baidu Encyclopedia and local cultural relic bureaus, with cross-validation to ensure accuracy). Data is divided into four categories: (1) Geographic coordinate data: Longitude and latitude of site points obtained via GPS equipment (Garmin GPSMAP 66i, Schaffhausen, Switzerland, positioning accuracy ±2 m) or professional map software, with duplicate point removal and coordinate system unification (WGS-84) (

Table 1. Historical and cultural sites in Leshan (coordinates in decimal degrees, WGS84).

No.	Name	Longitude (° E)	Latitude (° N)
1	Leshan Confucian Temple	103.761523	29.558502
2	Yangchi Pond	103.762005	29.558418
3	Louxing Gate	103.761705	29.558432
4	Longshen Shrine	103.768212	29.557461
5	Leshan Giant Buddha	103.771924	29.544073

and

Table 2. Historical sites of Wuhan University’s wartime relocation to Leshan (coordinates in decimal degrees, WGS84).

No.	Name	Longitude (° E)	Latitude (° N)
1	Monument commemorating Wuhan University’s relocation to Leshan	103.751098	29.557706
2	Former site of Wuhan University dormitory No. 5	103.767309	29.562531
3	Former site of Wuhan University dormitory No. 1	103.762245	29.557608
4	Former Site of Wuhan University women’s dormitory	103.757009	29.556255
5	Former site of Wuhan University College of Science	103.754772	29.555788
6	Former site of Wuhan University College of Engineering	103.751570	29.557967
7	Former site of Wuhan University dormitory No. 4	103.750604	29.556362
8	Former site of Wuhan University dormitory No. 6	103.748973	29.556471

). (2) Attribute text data: Structured descriptions organized from local chronicles, official archives, and on-site explanations, including historical evolution, cultural value, and architectural features. (3) Image data: Site photos containing geographic information (shot with Canon EOS R5, Tokyo, Japan, 45 MP resolution, with EXIF geographic information retained). (4) Behavior data: Sequential behavior records of team members during the study period (daily steps, moving distances) collected via mobile device applications (WeChat Motion, sampling frequency 5 min/time). To realize unified management and application of heterogeneous data, a data fusion model based on JSON schema is designed. The model defines a standardized data object structure: each site data must contain core fields (ID, name, position) and can flexibly expand associated fields (description, picture, statistics). A data-cleaning and conversion script is developed to standardize raw data into a format consumable by the front end. For data conflicts (e.g., inconsistent coordinate and text descriptions), a weighted voting method is adopted (field weight: geographic coordinate > official archive > public data) to ensure data consistency (Supplementary Materials S1) [13,33]. This fusion process operationalizes the three-level fusion framework proposed as the theoretical contribution of this study. By systematically integrating geographic coordinates, attribute texts, imagery, and user behavior data at the data, feature, and decision levels, the platform demonstrates how heterogeneous heritage information can be transformed into coherent spatial narratives, providing a reusable paradigm for digital cultural heritage projects (Figure 3).

3.3. Function Modules

The platform aims to provide a comprehensive window that integrates geospatial display, spatiotemporal data analysis, and browsing of cultural heritage information. As shown in Figure 4, it comprises four functional modules that collaborate via a unified data flow and interaction logic to construct a multi-level, interactive digital narrative space.

3.3.1. Map Display Module

As the platform’s spatial information foundation, this module presents the geographical distribution of the Leshan Confucian Temple and the former site of Wuhan University. Basic map services are implemented via the Gaode Maps API, with customized thematic layers superimposed to accurately mark historical building locations using different visual symbols. Color coding is based on user cognitive experiments (n = 30, recognition accuracy > 92%): red icons mark the Leshan Confucian Temple’s core architectural complex (religious heritage), and blue icons indicate the former Wuhan University site distribution points (educational heritage). Users can perform zooming, panning, and view switching operations; clicking any marker triggers interaction and activates an associated details window. The window content is dynamically generated via templates, integrating text descriptions, high-resolution images, and external links to Baidu Encyclopedia for a detailed, visually appealing display. The map module includes standard interactive controls (scale, zoom, full-screen, overview map), which users can toggle as needed (Supplementary Materials S2; Figure 5).

3.3.2. Data Visualization Module

This module focuses on the graphical conversion of visit-generated behavior data (daily steps, mileage). ECharts was selected as the rendering engine for its rich chart types, flexible configuration, and active community support. Step data is displayed via bar charts (bar height reflects daily activity level differences). Calculated mileage data uses area-padded line charts to emphasize continuous temporal trends. The charts support interactive legend filtering and hover tooltips, and automatically resize to fit the browser window (Figure 6).

3.3.3. Cultural Relics Information Display Module

Adopting a “card flow” design pattern to accommodate potential cultural relic data growth, each card is an independent visual unit composed of thumbnails, titles, and abstract text, flexibly arranged via CSS Grid layout. To optimize initial loading speed and performance, a paginated loading mechanism is implemented: initially, only 10 cards are rendered; when users scroll to the bottom or click “load more,” the next batch of data is asynchronously loaded and inserted via JavaScript (Supplementary Materials S3). This design avoids page lag from rendering large numbers of DOM elements at once. Cards feature hover effects for lightweight, interactive feedback and improved user experience.

3.3.4. Project Introduction and Guide Module

As the platform’s meta-information module, it provides research background, technical routes, and team member information with a focus on clarity and readability to reduce user cognitive load and facilitate quick understanding of the platform’s value and usage. It integrates with the navigation bar for site guidance.

3.4. Platform Architecture

To achieve the above functions and ensure system maintainability and scalability, the platform adopts a browser/server (B/S) architecture with front-end and backend separation, following clear hierarchical design principles. The overall architecture is divided into four layers from bottom to top (Figure 7).

• Data layer. Stores and manages all original and derived multi-source data (

  Table 3. Multi-source data types.

  Date Type	Data Detail
  Geographic coordinate data	Positioning of historical buildings stored as longitude-latitude pairs (WGS-84 coordinate system)
  Attribute text data	Descriptive texts (historical evolution, cultural value, architectural features) organized from authoritative sources
  Image data	Links to image resources (site photos, schematic diagrams) associated with locations/cultural relics (45 MP resolution)
  Behavior data	Sequential behavior records of team members (daily steps, moving distances) with a 5-min sampling frequency

  ). During development, data is organized as structured JSON files for direct front-end reading and parsing, simulating backend API data acquisition.
• Data governance layer. Transforms original heterogeneous data into standard visual display formats—a key step in converting spatial information to digital carriers [34]. Standardized data templates are developed for different data types (e.g., geographic data points standardized as {name, lnglat, description, imageurl, externallink} objects). This layer ensures data quality and provides clean, consistent data sources for upper-layer applications.
• Application logic layer. The platform’s core processing layer is fully driven by the client JavaScript engine, including three core sub-modules: map service engine, visualization rendering engine, and content management engine. Among them, the map service engine is built on the Gaode map JavaScript API, which is responsible for creating and rendering map instances, listening for and responding to interactive events (such as clicks and drags), and managing map controls. The visual rendering engine is based on the Echarts library. It receives statistical data after governance, generates interactive charts based on configuration items, and handles user interaction within the charts (such as hover prompts and legend switches). The content management engine dynamically manages the generation of cultural relic cards, paging loading logic, and the routing and smooth scrolling behavior of the entire page through native DOM operations and event processing.
• Presentation layer. The user-interactive Web interface, built following HTML5 semantic standards, with a fully responsive layout design using CSS3 and Tailwind CSS. It adapts to screen sizes from desktop to mobile, providing clear, aesthetically pleasing, and consistent visual presentation and interactive feedback.

The lightweight technology stack (Gaode Map JavaScript API, ECharts, and Tailwind CSS) and the modular, decoupled architecture described above embody the technical contribution of this work. This design enables rapid deployment, cross-device responsiveness, and efficient multi-source data visualization, offering a replicable template for similar cultural heritage dissemination platforms.

3.5. Overall Layout and Interaction Design of Web Platform

The front-end interface adopts a single-page application (SPA) design concept, with all functions seamlessly switched via scrolling and anchor links in one HTML page. The overall layout is rapidly built using Tailwind CSS, with responsive tool classes enabling adaptive multi-column layouts: desktop-side chart and cultural relic card areas are arranged side by side in multi-columns; mobile-side automatically switches to single-column stacking to ensure small-screen readability and operability. The global navigation bar is fixed at the top with smooth scrolling: clicking navigation items triggers animated scrolling to the corresponding chapters. For mobile devices, the navigation bar folds into a hamburger menu to save screen space. These detailed designs ensure cross-terminal platform consistency and usability.

4. Platform Evaluation

To ensure platform stability, functional integrity, and user experience, systematic testing and evaluation were conducted across four dimensions: function realization, performance, compatibility, and availability, verifying whether the platform meets design goals and providing a basis for subsequent optimization.

4.1. Evaluation Setup

The test was conducted in a controlled laboratory environment and a real network environment. The hardware environment covers high-performance computers (Intel i9, 32 GB RAM) and mainstream mobile devices (iPhone, Android phones). The software environment covers Windows 11, macOS, and mobile iOS and Android systems, and the browser includes the latest stable versions of Chrome 120+, Firefox 119+, Safari 17+, and Edge. The test data set uses multi-source data from the Leshan Confucian temple and the former site of Wuhan University, after fusion processing as described in Section 3.2. At the same time, an extended data set containing more than 150 geographical feature points is synthesized for the stress test.

4.2. Functional Testing

The functional evaluation adopted a black-box testing method based on requirements specifications to verify whether each core module was implemented as designed.

• Map service module: verified successful Gaode Maps JavaScript API integration, including accurate map initialization (center point, zoom level), custom marker rendering (icons, color coding), and dynamic information window generation with accurate content (text, pictures, hyperlinks).
• Data visualization module: confirmed correct ECharts binding and rendering of step (bar chart) and mileage (line chart) data; interactive tests validated correct legend filtering response and accurate mouse-hover tooltip data.
• Information integration and interaction module: tested cultural relic card paginated loading logic, verifying “load more” button behavior (loading, status update, hiding) met expectations; evaluated global navigation smooth scrolling and mobile responsive menu switching.

All preset functional test cases passed (

Table 4. Functional test cases and results.

No.	Module	Test Case	Expected Result	Actual Result
1	Map Display	Click the marker, check the info window	Info window shows correct content	Pass
2	Chart Interaction	Click legend to toggle data series	Chart updates accordingly	Pass
3	Paginated Loading	Scroll to the bottom, click “Load More”	Additional cards appear	Pass

). The platform successfully achieved multi-source data visual integration: the map module accurately marked all historical positions with complete information windows; the data visualization module correctly reflected the original data with accurate interactive responses; the cultural relic module’s paginated mechanism worked effectively, improving the long-list browsing experience. No major defects were found, indicating the platform meets basic technical objectives.

4.3. Usability Testing

In total, 15 participants with diverse technical backgrounds were invited (5 cultural heritage lovers, 5 students, and 5 ordinary users) to complete preset tasks and complete the System Usability Scale (SUS) [35]. Participants were recruited through convenience sampling. “Ordinary users” refer to individuals without professional backgrounds in GIS, cultural heritage, or related fields, representing the general public visiting heritage sites. The sample size (n = 15) is consistent with Nielsen’s recommendation for usability testing, which suggests that 5–20 participants can identify the majority of usability issues [36,37]. Tasks included: locating a specific building on the map and viewing details, comparing different members’ activity data on a specific date, and browsing all cultural relic cards to find a specific one. Qualitative evaluation of system intuitiveness, learnability, and operational efficiency was conducted via task completion time, success rate, and user feedback.

The average SUS score for the user experience test is 82.5 (0–100 scale), indicating that the platform has a “good” to “excellent” usability level. The specific task analysis shows that the average completion time for the map information search task is 28 s, with a 100% success rate. Users generally appreciate the design of the information window, which combines color coding and text. The average completion time of the data comparison task was 35 s, and the success rate was 93%. A few users need a prompt the first time to discover the legend’s interactive features. The feedback on the cultural relic browsing task experience is positive, and the paging loading is considered effective at managing cognitive load. The main improvement suggestions from users focus on adding search functions and more detailed timeline filters. SUS scores by user type: cultural heritage lovers (86.2), students (83.5), ordinary users (78.3), reflecting slightly higher satisfaction among professional groups. Typical user feedback included the following: “The color distinction between different heritage types is clear and easy to recognize”; “Legend interaction needs more obvious prompts.”

4.4. Performance and Compatibility Testing

• Loading performance: Chrome DevTools Lighthouse and Network panels measured First Contentful Paint (FCP), Largest Contentful Paint (LCP), and full page loading time under simulated 4G and 5G network throttling, with five test repetitions for average values.
• Runtime performance: in stress test scenarios (150+ markers), the Performance panel recorded page frame rate (FPS) and memory usage changes, evaluating interaction fluency (rapid map zooming, chart switching);
• Cross-platform compatibility: full-function traversal tests were conducted on different device-browser combinations to ensure core function and visual layout consistency.

The performance test results are shown in

Table 5. Comparison table of network performance evaluation indicators.

Evaluation Metric	5G	4G	Stress Test (150 Points)
First Contentful Paint (FCP)	1.2 s	2.8 s	1.5 s
Largest Contentful Paint (LCP)	2.5 s	5.1 s	3.8 s
Fully Loaded Time	3.1 s	7.4 s	4.5 s
Interaction Frame Rate (FPS)	>60	>60	20–45

. Under 5G, the platform exhibited excellent loading performance with rapid core content presentation; under 4G, full loading time increased due to external API and library volume, but asynchronous loading allowed early map interaction. In stress tests, map initial rendering FPS temporarily dropped to ~20 with 100+ markers, but pan/zoom operations remained smooth (>30 FPS) after recovery; memory usage remained stable during normal interaction. Compatibility tests confirmed consistent functions and well-adapted responsive layouts across all target browsers and devices.

4.5. Comparative Analysis

A representative study is selected: a Leaflet-based WebGIS for the conservation of historical buildings in Sardinia, Italy [26]. It uses Leaflet as the front-end mapping library, with PostgreSQL/PostGIS for spatial data management and PHP for server-side logic. As shown in

Table 6. Comparison with similar platforms.

Evaluation Dimension	This Platform	Sardinia Leaflet WebGIS Platform [26]
Technology stack	Lightweight front-end stack (Gaode Map API + ECharts + Tailwind CSS)	Lightweight front-end + backend (Leaflet + PostgreSQL/PostGIS + PHP)
3D Capability	2D	2D
Development cost	Low	Medium
Deployment mod	Static files, any web server or CDN	Requires PostgreSQL + PHP server
Cross-device compatibility	Excellent (responsive design)	Good (mobile-optimized version available)
Functional coverage	Comprehensive (multi-source integration + interaction)	Primarily geospatial + attributes

, both platforms are 2D, focus on cultural heritage, and are built on open-source or freely available technologies. The proposed platform distinguishes itself through a purely front-end, serverless, lightweight stack (Gaode Map JavaScript API + ECharts + Tailwind CSS). In contrast, the Sardinia platform requires a backend database and a PHP server. This design choice gives our platform three practical advantages: (1) Lower development and deployment cost. Static files can be hosted on any web server or CDN without a dedicated GIS infrastructure. (2) Excellent cross-device compatibility. The responsive layout works seamlessly on desktops, tablets, and mobile phones, while the Sardinia platform’s mobile experience requires separate optimization. (3) Simpler data fusion. Multi-source data (coordinates, text, images, behavioral records) is integrated via client-side JSON schemas, avoiding the complexity of a full-stack database. In addition, the proposed platform aims to support public education and lightweight exploration, prioritizing accessibility and ease of use over backend richness, making it more suitable for the general audience.

5. Discussion

This study successfully validated the feasibility of the Gaode Maps API, ECharts, and Tailwind CSS core technology stack for lightweight historical and cultural WebGIS applications. The platform effectively addresses the integrated presentation challenge of multi-source data (space, attributes, statistics, media). It enhances users’ exploratory understanding of cultural heritage information through intuitive interactive designs (linked map markers with information windows and filterable charts). These features align with current digital humanities trends emphasizing interactive narrative and public participation [9,12]. This platform’s lightweight architecture offers outstanding advantages: easy deployment, low access threshold, and excellent cross-platform compatibility, making it particularly suitable for the rapid deployment of cultural heritage education resources, tourism guides, or research dissemination. The high SUS score (82.5) confirms its strong usability.

From a theoretical perspective, this study contributes to the discourse on spatial narrative and multi-source data fusion in digital humanities. By integrating geospatial visualization with user behavior data, the platform constructs a form of spatial narrative that reveals how heritage sites are experienced over time, which is often absent in traditional static representations [9,12]. The application of the three-level fusion framework (data-level, feature-level, decision-level) demonstrates how heterogeneous heritage data can be systematically integrated to support cultural understanding, providing a reusable paradigm for similar projects [13,23]. In the current implementation, the platform achieves a spatially grounded narrative through three design features: (1) Map markers with rich information windows, which present location-specific historical and architectural details. (2) Linked behavioral data charts, which allow users to overlay personal activity traces onto the heritage space. (3) Paginated cultural relic cards, which provide structured access to heritage descriptions. These features enable users to explore heritage sites not as isolated points but as interconnected elements within a spatial context.

Methodologically, the server-free, browser-only design represents a meaningful alternative to mainstream WebGIS development trends. Although many recent studies have pursued higher realism through 3D modeling, complex semantic frameworks, and large backend databases, such systems often present barriers for long-term maintenance and public accessibility [15,27]. This study shows that lightweight design is not a compromise in functionality but a deliberate, defensible strategy that improves usability, deployability, and inclusivity. The integration of multi-source heterogeneous heritage data through standardized JSON schemas also provides a replicable structure for other small-scale heritage documentation projects.

In practical terms, the platform enables non-expert users and heritage managers to visualize integrated information without specialized GIS training. This supports more accessible cultural dissemination and public engagement, which is especially valuable for heritage sites with limited funding and technical support [25,26].

However, we acknowledge that a fully coherent narrative that integrates temporal sequences, causal relationships, and user-driven storytelling remains beyond the current prototype. Limitations of this study are as follows: (1) Data management and analysis depth: The current platform uses static data files, limiting dynamic data updates and large-scale management capabilities. The front-end-only architecture restricts real-time data synchronization and advanced spatiotemporal analysis, such as kernel density estimation or spatial autocorrelation analysis of heritage distribution patterns. (2) Expression form expansion: Visualization primarily relies on 2D maps and statistical charts, lacking the immersive quality of 3D visualization. While this trade-off supports lightweight deployment and broad accessibility, it may not fully satisfy users seeking immersive historical reconstructions. (3) Interaction and personalization: The platform lacks personalized recommendations, social annotation, and collaborative creation functions, limiting public participation in cultural heritage construction. Features such as user-generated content, comment threads, or collaborative storytelling could transform passive consumption into active engagement. (4) Semantic depth of data fusion: As noted above, deeper semantic integration with knowledge graphs is needed to fully realize the potential of decision-level fusion. (5) Site and user sample size: This study is a phased exploratory study; hence, the case and sample size are appropriate for a prototype verification. Future research will expand to more sites and larger-scale user testing. (6) Gaode maps geographical limitations: Gaode maps provide high-precision geographic data in the Chinese mainland, but have limited support for regions outside China. Since this study focuses on heritage sites in Leshan, China, the geographic limitation does not affect the experimental results. For international applications, the platform can be easily adapted to Google Maps or other global map APIs.

In the future, there are some research directions: (1) Integrating backend databases for dynamic data management and spatiotemporal analysis. (2) Incorporating 3D visualization and digital twin technologies. (3) Adding personalized recommendations and social annotation features. (4) Extending semantic fusion with knowledge graphs. (5) Conducting longitudinal user studies with larger, more diverse participant groups.

6. Conclusions

This study designed and implemented a lightweight WebGIS visualization platform for historical and cultural heritage based on multi-source data fusion. The following core conclusions can be drawn through systematic functional tests, performance evaluation, and user experience analyses.

• This study developed a purely front-end, serverless architecture using the Gaode Map JavaScript API (2D mapping), ECharts (interactive charts), and Tailwind CSS (responsive layout). A JSON-based fusion model successfully integrates geospatial coordinates, attribute texts, imagery, and user behavior data. Performance tests showed a First Contentful Paint of 1.2 s under 5G and an interaction delay of < 3 s, confirming the feasibility of the lightweight approach.
• The platform achieved a System Usability Scale score of 82.5 (“good” to “excellent”). Task completion rates were 100% for map-based searching and 93% for data comparison. User feedback indicated that the spatially grounded presentation (color-coded markers, information windows, linked charts) effectively supports exploratory learning of heritage sites.
• This study operationalizes spatial narrative theory and multi-source data fusion theory in a concrete WebGIS implementation, providing a reusable framework for digital humanities projects. The platform offers a low-cost, easily deployable tool for heritage dissemination, with full source code and data schemas provided as Supplementary Materials. The responsive design ensures cross-device usability.

Future work will focus on backend integration, selective 3D enhancement, personalization, semantic knowledge graphs, and larger-scale user studies. The current prototype serves as a solid foundation for these extensions.