Asosiy kontentga oʻtish
proceeding

Automating Digital Advertising: A Study on AI-Driven Bidding and Budget Allocation Efficiency

Nilufar YuldashevaDepartment of Management and Marketing Fergana State Technical University, Fergana, UzbekistanIlkhom Abbosovich NosirovDepartment of Management and Marketing Fergana State Technical University, Fergana, UzbekistanMakhfuza RustamovaDepartment of Management and Marketing Fergana State Technical University, Fergana, UzbekistanIlmidin Toshmatovich YormatovDepartment of Management and Marketing Fergana State Technical University, Fergana, UzbekistanNodirakhon MakhmudovaDepartment of Management and Marketing Fergana State Technical University, Fergana, UzbekistanFeruza Dzhurakuzievna AvulchayevaDepartment of Management and Marketing Fergana State Technical University, Fergana, Uzbekistan
2026en
Ochiq kirishOltin

Litsenziya: cc-by

ABI

Annotatsiya

Programmatic advertising's real-time bidding (RTB) ecosystem processes over 8 trillion bid requests per day across global supply-side platforms, demand-side platforms, and ad exchanges-generating a data-rich environment whose optimisation complexity exceeds human analytical capacity by orders of magnitude. This paper proposes an AI-Integrated Bidding and Budget Allocation Framework (AIBBA) combining a real-time bidding intelligence and dynamic floor price engine, a cross-channel budget allocation and campaign pacing module, and an audience segmentation, targeting precision, and compliance gateway, evaluated across 42 advertising companies (e-commerce$\mathbf{n}=\mathbf{1 4}$, financial services$\mathbf{n}=\mathbf{1 2}$, consumer goods$\mathbf{n}=\mathbf{1 0}$, technology$\mathbf{n}=\mathbf{6})$processing 18,480,000 impression and click records (12,936,000 training / 5,544,000 test, 70/30 stratified split) across 6 advertising platforms (Google Search, Meta Social, TikTok Video, LinkedIn B2B, Amazon DSP, Open Programmatic DSPs) and 284,800 ad campaigns over a 34-month evaluation horizon. The AIBBA achieves a composite advertising effectiveness score of$\mathbf{r}=\mathbf{0. 9 2 0}$against expert-validated campaign outcome benchmarks, outperforming MLonly advertising models$(\mathbf{r}=\mathbf{0. 7 8 8}$; +16.8%$[(0.920-0.788) / 0.788])$, rule-based bidding systems$(\mathbf{r}=\mathbf{0. 6 1 4}$; +49.8% [(0.920-0.614)/0.614]), and manual campaign management$(\mathbf{r}=\mathbf{0. 4 6 9}; \quad+\mathbf{9 6. 2 \%} \quad[(\mathbf{0. 9 2 0}-\mathbf{0. 4 6 9}) / \mathbf{0. 4 6 9}])$. Deployment outcomes: +34.0% Return on Ad Spend (ROAS), -27.0% Cost-per-Click (CPC), +52.4% Click-Through Rate (CTR), -38.0% budget waste reduction, +44.0% audience match rate, and 94.8% GDPR / CCPA regulatory compliance rate. Sub-domain performance: Real-Time Bidding Optimisation$(\mathbf{r}=\mathbf{0. 9 4 4})$, Budget Allocation Efficiency (0.926), Audience Targeting Precision (0.898), and Cross-Platform Campaign Integration (0.868). Ad fraud and bot traffic contamination degrade AIBBA accuracy to 0.861 (-6.4% [(0.920-0.861)/0.920]), and third-party cookie deprecation to 0.850 (-7.6% [(0.920-0.850)/0.920]), with deployed mitigation protocols for both conditions.

Hali tarjima qilinmagan

Identifikatorlar

Iqtiboslar va manbalar