Збережено в:
| Автори: | , , , , , |
|---|---|
| Формат: | Recurso digital |
| Мова: | |
| Опубліковано: |
Zenodo
2026
|
| Предмети: | |
| Онлайн доступ: | https://doi.org/10.5281/zenodo.20084729 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
Зміст:
- Digital marketing requires product understanding, customer insight, competitive awareness, creative writing, and platform-specific communication. For small businesses, independent sellers, student entrepreneurs, freelancers, and influencers, producing effective campaigns is difficult because it demands both creativity and continuous market research. Existing AI copywriting tools can generate promotional text, but many depend mainly on text prompts, produce generic outputs, and do not fully incorporate visual product cues or real-time market context. This paper presents Ad Genie, a multimodal generative AI framework for automated marketing campaign creation. The proposed system accepts a product image and a campaign or product description as input, extracts visual and semantic features using vision-language models, generates search queries for market intelligence, retrieves trend and review-oriented information from online sources, and produces structured campaign assets using a large language model. The generated outputs include social media posts, a blog concept, a short promotional video script, target audience persona, market trends, sentiment summary, and structured intermediate results. The prototype demonstrates how multimodal AI, natural language processing, computer vision, retrieval-augmented generation, and web intelligence can be integrated into a unified workflow for context-aware marketing assistance. The work contributes a practical architecture for AI-driven campaign automation and identifies future directions such as multilingual generation, brand voice learning.