Modelpublic

data-is-everything/journal-retrieval-code

sourceHugging Facemitupdated 2y agoView on Hugging Face

0likes

Model Card

Model Card: Journal Retrieval Code Model

Model Description

—Model Name: Journal Retrieval Code Model
—Model Architecture: Biopython-based data extraction and retrieval
—Intended Use: Extracts and retrieves journal data from PubMed articles using PM IDs.
—License: MIT License

Training Data

—Dataset: The model was used on PubMed articles data.
—Data Preprocessing: Data was preprocessed by extracting relevant fields such as abstract, title, journal, language, year, and month.

Training Procedure

—Hyperparameters: N/A
—Training Duration: N/A
—Compute Resources: N/A

Evaluation

—Metrics: Accuracy of data retrieval and parsing
—Results: N/A
—Evaluation Data: Various PM IDs provided by users.

Intended Use Cases

—Primary Use Case: Academic research, systematic reviews, and meta-analyses.
—Secondary Use Cases: General data extraction for biomedical research.
—Misuse: Not suitable for extracting data outside the PubMed database. May not handle edge cases well.

Limitations

—Known Limitations: Only works with PubMed data and requires valid PM IDs.
—Potential Biases: Relies on the availability and accuracy of PubMed's data.

How to Use

Installation

Install the necessary dependencies:

bash

pip install biopython pandas huggingface_hub