Digital Scholarship

Applied machine learning, computational text analysis, and dataset creation

Andrew Paul Janco

apjanco@princeton.edu

Datafication: From document images to structured research data

Handwritten documents

Printed tables

Form data

Mixed media

Visual Document Understanding (VDU)

Visual Reasoning

Geolocation of images

Work with large text datasets 

- Institutional Books 

- Common Pile

Natural language processing

- Named entity recognition

- entity reconciliation

- entity linking

Text classification

Text embeddings

Topic Modeling

Similarity search

Agentic workflows (Langchain, LangGraph)

Computational Text Analysis

{{slide}}

Applied Machine Learning + GenAI

01

Model fine-tuning with SFT and GRPO

02

Evaluation and responsible AI research practices

{{slide}}