Team Members
Logan Ankarberg, Nathanael Steven, Drew McMains
Abstract
The Hermes project, led by Team ThinFilm Thinkers and sponsored by Dr. Elton Graugnard, addresses the challenges researchers face when reading academic papers for their research. The goal of this project is extracting relevant information from Atomic Layer Deposition (ALD) research papers. ALD is an important and prevalent chemical process used in the construction of silicon wafers for computer chips among other things. The project aims to streamline the extraction of temperature-related data, enhancing researchers’ efficiency. The data being processed are pdfs of ALD research papers, and our strategy to extract temperatures involves defining declarative grammars for pattern matching as part of our parsing pipeline. This will allow researchers to easily identify papers that perform experiments with temperatures they cannot replicate. Hermes aims to enhance the experience of researchers by automatically providing them with relevant information regarding the paper’s experimental procedure, thereby expediting the process of literature review.
Project Description
Using python, we create a command line interface that performs temperature extraction on PDF files. This was designed for researchers in the field of atomic layer decomposition (ALD). It provides a user-friendly way to extract temperatures from pdfs so that the researcher can determine whether they can reproduce the experiment with the tools they have. We used a regular expression library in python to extract the temperatures, and we used pandas to store the results in a dataframe. We used pyinstaller to bundle the cli and its dependencies together for use on any popular os (macOS, Windows, and Linux).
In order to use the app, you simply run the executable from within the bundled pyinstaller folder. This will launch our cli, and it is an interactive application that will only close if you type exit or Ctrl+C. It gives you a help menu when you first run it that explains what the project is, and prompts you to either see more information about the project, or see all of the available commands.
Here are some screenshots of the cli application being used.
How to extract temperatures from a pdf - the main function of program
How to view the results
