Meeting 04

The importance of building “Big Tables”

I met with some of you about your research projects, and I noticed some confusion about how to start a new research topic. From my perspective, building “big tables” is always the first step.

These big tables include:

• primary sources
• related scholarship
• other relevant materials

Introduction of Today’s Task

There are two PDF files in the pdf/ folder. They are extracted pages from Wolfgang Frank’s famous book Introduction of Ming sources (see a newer edition on WorldCat: https://search.worldcat.org/title/779242963). For this class, we are using an edition published in the 1960s.

• pdf/wf_abbreviations.pdf is the abbreviations list.
• pdf/wf_five_pages.pdf contains five pages from the first chapter.

Your tasks:

1. Extract the abbreviations and their meanings from the list into a table.
2. Turn the abbreviations in the content into clickable links.

Google NotebookLM

Landing page: https://notebooklm.google/

OCR tool for Apple

The software is called glm-ocr-mlx. It is provided as ocr_software/glm-ocr-mlx-main.zip. Download the file and unzip it, then read the README.md file for instructions on how to use it.

Currently, it only works on Apple Silicon computers. If you have a Windows/Linux/Intel Mac, you can use the OCR tool for LM Studio/Ollama, which we will introduce in the next section.

Important

Why do we need JSON files for OCR output?

OCR tool for LM Studio/Ollama

Repository: https://github.com/kltng/ocr_batch_processor

Ollama: https://ollama.com/

Use Openwork or chatbots to clean the OCR output and turn them into clickable links.

If you cannot get OCR output, you can use the JSON files in the ocr_results/ folder. They are the output of glm-ocr-mlx. You can use Openwork or chatbots to clean the OCR output and extract the abbreviations and their meanings into a table.