We combine modern ML toolchains (Python/ML libraries, MLOps best practices) with experimental platforms and partnerships. See our WU Research profile for team, publications, projects, and activities.

AIMA RESEARCH 

Generative AI & Content Optimization

(LLMs, multimodal AI, GEO)

We study how generative AI systems create, rank, and optimize marketing content. Using large language models (LLMs) and multimodal AI, we analyze how language, images, and structure influence persuasion and visibility in AI-driven environments.

Example paper:

• Creating Effective Marketing Messages Through Moderately Surprising Syntax - A. Selin Atalay, Siham El Kihal, Florian Ellsaesser, 2023

Personalization, Pricing & Algorithmic Decision-Making

(recommender systems, dynamic pricing, bargaining)

Our work includes designing AI systems that adapt decisions to individual customers, including recommendations, pricing, and negotiation. We combine machine learning with economic modeling to optimize who gets what offer, at what price, and when.

Example papers:

• It•s not only what they buy, it•s also what they keep: Linking marketing instruments to product returns

• Bots Bargaining with Humans: Building AI Super-Bargainers with Algorithmic Anthropomorphization

Customer & Unstructured Data Analytics

(NLP, computer vision, multimodal data)

Leveraging unstructured and multimodal data - text, images, and behavioral signals - to uncover customer preferences and market dynamics. Our research bridges modern AI methods (NLP, computer vision) with marketing decision-making.

Example papers:

• Unstructured data research in business: Toward a structured approach - ScienceDirect

• Pooled Wisdom, Unique Insights: Improving Customer Predictions with Cooperative Databases and Deep Transfer Learning ​​​​​​​

Human-AI Interaction

(behavioral responses to AI systems)

Understanding how AI agents reshape interactions between consumers and firms in marketplaces, with a focus on trust, fairness, and behavioral responses to automated decision-making.

Example paper:

• Bots Bargaining with Humans: Building AI Super-Bargainers with Algorithmic Anthropomorphization

Causal AI, Experimentation & Customer Lifecycle

(upfilt modeling, policy learning, field experiments)

We develop and apply causal machine learning and experimental methods to identify and optimize the impact of marketing interventions over the full customer lifecycle. Our work goes beyond short-term conversion to study how actions shape returns, retention, and long-term customer value.

Example papers:

• Customer return rate evolution - ScienceDirect

• It•s not only what they buy, it•s also what they keep: Linking marketing instruments to product returns