Summarize the latest research papers abstract from Arxiv using chatGPT for staying up to date about the latest state of the art papers. This project was built because of a personal need, which is to stay up to date about the latest research papers in the field I am working on.
- Using a query, retrieve a desired number of latest papers from
Arxiv.com
. - Using ChatGPT, summarize the abstarct of the papers.
- Save the papers information:
- Paper id;
- Paper title;
- Paper abstrat;
- Paper authors;
- Paper publication date;
- Paper summary.
- Python: Python 3.8 or newer.
- GPT-3.5 API Access.
- Clone the repository: 'https://github.com/JaouadT/papers_abstract_summarizer.git'.
- cd into the repository:
cd papers_abstract_summarizer
. - Install requirements:
pip install -r requirements.txt
. - Save your GPT-3.5 API key in a file named
openai_api_key.txt
in the folderopenai_key
. - Run the script
main.py
with the desired query and number of papers to summarize. For example:python main.py --query "Ultrasound images segmentation" --max_results 5
.
Note:
- The script will save the papers information in an excel spreadsheet named
paper_information.xlsx
in the folderoutput
. - The script expects to retrieve at least two papers to summarize. If the number of papers retrieved is less than two, the script will raise an error.
If you have any suggestions on how to improve the code, please feel free to open an issue or a pull request. Many features can be added to this project, such as:
- Support multiple open access papers sources (e.g. IEEE, Springer, ...).
- Summarize a paper from its url.
- Summarize a paper from its id.
- Summarize the papers introduction and conclusion.
- Summarize the papers body.
- Extract the papers keywords.
- Extract the papers references.
- Build a web app for the project.
- ...