ENTERPRISE AI ANALYSIS
Explainable AI with Fine-Tuned Large Language Models for Sustainable Cultural Heritage Management
The preservation of cultural heritage areas, especially in historical urban environments, requires a nuanced understanding of public perceptions to balance preservation with modernization. This study introduces an advanced AI-driven framework for Aspect Sentiment Quadruple Prediction (ASQP) to assess public perceptions of Lijiang Ancient Town, a UNESCO World Heritage site in China. We fine-tuned the large language model Qwen-14B using LoRA-based methods to augment sentiment data, effectively uncovering implicit emotional cues in social media content. The model integrates BERT, multi-layer BiLSTM, self-attention, CNN, and CRF for enhanced entity recognition and sentiment classification. Experimental results show that the enhanced model (Qwen-14B + ASQP) improved F1-score by 0.97% (from 75.42% to 76.39%) and Precision by 4.48% (from 76.14% to 80.62%) compared to the baseline. Analyzing data from platforms such as Weibo, Dazhong Dianping, and Xiaohongshu (2018-2024), the research uncovers factors influencing public perception, offering insights for heritage site management, urban planning, and the sustainable preservation of cultural heritage.
Key Performance Indicators
The enhanced ASQP model, fine-tuned with Qwen-14B, significantly outperforms baselines, demonstrating its capacity to uncover subtle emotional cues in social media content for precise public perception analysis in cultural heritage management.
0.97%
F1-Score Improvement
4.48%
Precision Improvement
3
Data Sources Analyzed
32,074
Social Media Entries
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
ASQP Framework for Sentiment Analysis
Phase 1: LoRA-based Data Augmentation
→
Phase 2: Entity Recognition
→
Phase 3: Aspect Category & Sentiment Classification
76.39%
F1-Score with LLM Augmentation
The enhanced model (Qwen-14B + ASQP) improved F1-score to 76.39%.
| Model | F1-Score | Precision |
|---|---|---|
|
75.42% | 76.14% |
|
76.39% | 80.62% |
Lijiang Ancient Town Case Study
Analyzing data from platforms such as Weibo, Dazhong Dianping, and Xiaohongshu (2018-2024), the research uncovers factors influencing public perception, offering insights for heritage site management, urban planning, and the sustainable preservation of cultural heritage. Positive sentiments predominated in 'Visual and Photographic Experience' and 'Cultural Heritage Transmission'.
Calculate Your Potential ROI
Estimate the annual cost savings and hours reclaimed by implementing an AI-driven sentiment analysis framework in your enterprise operations.
Estimated Annual Savings
$0
Hours Reclaimed Annually
0
Your AI Implementation Roadmap
A phased approach to integrating AI-driven sentiment analysis for sustainable cultural heritage management, ensuring a smooth transition and measurable impact.
Phase 1: Data Integration & LLM Fine-tuning (1-2 Weeks)
Integrate your social media data, historical records, and visitor feedback. Fine-tune Qwen-14B with LoRA for domain-specific sentiment augmentation.
Phase 2: ASQP Model Deployment (2-3 Weeks)
Deploy the BERT-BiLSTM-CRF-CNN ASQP framework. Configure for real-time sentiment extraction and categorization across various heritage aspects.
Phase 3: SHAP Interpretability & Insights Dashboard (1-2 Weeks)
Implement SHAP for explainable AI. Develop an interactive dashboard to visualize sentiment trends, identify key influencing factors, and generate actionable insights for urban planners.
Ready to Transform Your Enterprise?
Leverage cutting-edge AI to gain unparalleled insights into public perception and drive sustainable cultural heritage management. Our experts are ready to help you implement a tailored solution.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat