بررسی و تحلیل مبتنی بر داده نوآوری‌های هوش مصنوعی در بازاریابی بر اساس رویکرد ثبت اختراع

Introduction

Rapid advances in AI, machine learning, and data analytics are transforming marketing from intuition-driven practices to data-driven, predictive paradigms. Despite a surge of patents at the AI–marketing intersection, the literature remains fragmented and application-centric, lacking a comprehensive, data-driven framework to identify and cluster the underlying technologies. This study leverages patent data to map the technological landscape of AI that underpins modern marketing.



Methodology

Patents were collected from Lens.org using a hybrid search strategy (AI and marketing keywords plus IPC/CPC codes). An initial 7329 documents were filtered via a domain relevance score threshold of 0.50, yielding 2108 patents. Title and abstract texts were merged and embedded into high-dimensional vectors. Dimensionality reduction followed a two-step pipeline: PCA for initial denoising and UMAP to preserve nonlinear neighborhood structures. Clustering was performed using K-Means and validated with HDBSCAN. Cluster quality was assessed by a cosine similarity–based topic coherence metric; clusters below 0.36 were merged or removed.

Discussion and Results

The final structure comprises 10 meaningful clusters across 2108 patents, spanning predictive customer analytics, intelligent content generation and management, smart valuation and dynamic pricing, data integration and data-centric architectures, multimodal machine learning, and risk/compliance management. The uneven patent distribution indicates heterogeneous technological maturity and industrial demand. Results reveal a shift from isolated analytical tools toward integrated, predictive, decision-support systems and uncover tight coupling between marketing and operations/supply chain.

Conclusion

AI has evolved from a “facilitator” to the “architect” of the marketing ecosystem, rewriting the 4Ps: product as intelligent service, dynamic pricing, phygital place, and algorithmic promotion. Sustainable advantage resides less in algorithms and more in data architecture and quality, with innovation trajectories moving toward responsible AI. This technology map offers a foundation for policy design, investment, and strategic planning in digital marketing.

Introduction

Rapid advances in AI, machine learning, and data analytics are transforming marketing from intuition-driven practices to data-driven, predictive paradigms. Despite a surge of patents at the AI–marketing intersection, the literature remains fragmented and application-centric, lacking a comprehensive, data-driven framework to identify and cluster the underlying technologies. This study leverages patent data to map the technological landscape of AI that underpins modern marketing.



Methodology

Patents were collected from Lens.org using a hybrid search strategy (AI and marketing keywords plus IPC/CPC codes). An initial 7329 documents were filtered via a domain relevance score threshold of 0.50, yielding 2108 patents. Title and abstract texts were merged and embedded into high-dimensional vectors. Dimensionality reduction followed a two-step pipeline: PCA for initial denoising and UMAP to preserve nonlinear neighborhood structures. Clustering was performed using K-Means and validated with HDBSCAN. Cluster quality was assessed by a cosine similarity–based topic coherence metric; clusters below 0.36 were merged or removed.

Discussion and Results

The final structure comprises 10 meaningful clusters across 2108 patents, spanning predictive customer analytics, intelligent content generation and management, smart valuation and dynamic pricing, data integration and data-centric architectures, multimodal machine learning, and risk/compliance management. The uneven patent distribution indicates heterogeneous technological maturity and industrial demand. Results reveal a shift from isolated analytical tools toward integrated, predictive, decision-support systems and uncover tight coupling between marketing and operations/supply chain.

Conclusion

AI has evolved from a “facilitator” to the “architect” of the marketing ecosystem, rewriting the 4Ps: product as intelligent service, dynamic pricing, phygital place, and algorithmic promotion. Sustainable advantage resides less in algorithms and more in data architecture and quality, with innovation trajectories moving toward responsible AI. This technology map offers a foundation for policy design, investment, and strategic planning in digital marketing.

Introduction

Rapid advances in AI, machine learning, and data analytics are transforming marketing from intuition-driven practices to data-driven, predictive paradigms. Despite a surge of patents at the AI–marketing intersection, the literature remains fragmented and application-centric, lacking a comprehensive, data-driven framework to identify and cluster the underlying technologies. This study leverages patent data to map the technological landscape of AI that underpins modern marketing.



Methodology

Patents were collected from Lens.org using a hybrid search strategy (AI and marketing keywords plus IPC/CPC codes). An initial 7329 documents were filtered via a domain relevance score threshold of 0.50, yielding 2108 patents. Title and abstract texts were merged and embedded into high-dimensional vectors. Dimensionality reduction followed a two-step pipeline: PCA for initial denoising and UMAP to preserve nonlinear neighborhood structures. Clustering was performed using K-Means and validated with HDBSCAN. Cluster quality was assessed by a cosine similarity–based topic coherence metric; clusters below 0.36 were merged or removed.

Discussion and Results

The final structure comprises 10 meaningful clusters across 2108 patents, spanning predictive customer analytics, intelligent content generation and management, smart valuation and dynamic pricing, data integration and data-centric architectures, multimodal machine learning, and risk/compliance management. The uneven patent distribution indicates heterogeneous technological maturity and industrial demand. Results reveal a shift from isolated analytical tools toward integrated, predictive, decision-support systems and uncover tight coupling between marketing and operations/supply chain.

Conclusion

AI has evolved from a “facilitator” to the “architect” of the marketing ecosystem, rewriting the 4Ps: product as intelligent service, dynamic pricing, phygital place, and algorithmic promotion. Sustainable advantage resides less in algorithms and more in data architecture and quality, with innovation trajectories moving toward responsible AI. This technology map offers a foundation for policy design, investment, and strategic planning in digital marketing.