Designing a Retail Strategy Taxonomy for AI-Powered Document Intelligence

Format: Online
Number of Interns: 3-4 students
Duration of the Internship: Summer term, approximately 2-3 weeks (full-time)
Start Date: June 22, 2026
Finish Date: July 10, 2026
Application Deadline: May 25, 2026
Project Supervisor: Asst. Prof. Ayşe Çetinel

Project Description: This project focuses on defining the themes and data fields that an AI system will use to analyze retailer 10-K/10-Q/annual reports. The student will carefully read selected filings, identify recurring strategic topics (e.g., business model, omnichannel, AI/data, operations, sustainability), and translate them into a clear set of theme labels and extraction fields. The main tasks include drafting theme definitions, inclusion/exclusion rules, and per-theme extraction schemas, and testing them on a small sample of text. The expected outcome is a well-documented taxonomy (codebook + tables) that can be directly used as configuration for an AI-powered document intelligence pipeline.

Research Intern Responsibilities: The primary responsibility of the intern is to conduct a detailed review of retailer annual reports (10-K/10-Q) to identify and categorize recurring strategic patterns. Based on this analysis, the student will develop a comprehensive taxonomy by defining clear categories for strategic themes and designing structured extraction templates (such as Excel or JSON schemas) that define exactly what data points need to be captured. Additionally, the intern will be responsible for drafting a "Codebook" that establishes strict inclusion and exclusion rules to ensure consistency in data labeling. Finally, the intern will validate the taxonomy by applying it to a small sample of text to test for coverage and ambiguity, refining the definitions as necessary based on these pilot tests.

Required Skills and Qualifications: Undergraduate student with strong reading comprehension in English and interest in strategy, or AI. Comfortable working with structured tables (Excel/Google Sheets); basic familiarity with Python is helpful but not required. Attention to detail and systematizing information are important. Expected Learning Outcomes: By the end of this project, the student will have gained significant proficiency in navigating and interpreting complex corporate financial filings and annual reports. They will develop a deep understanding of current strategic drivers in the retail sector, such as omnichannel integration, ESG initiatives, and digital transformation. Furthermore, the student will acquire practical experience in data structuring for AI, specifically learning how to translate unstructured text into organized formats, while enhancing their analytical rigor in qualitative coding and systematic documentation.

Click for Google Form.