Spaces:
Sleeping
Sleeping
| import streamlit as st | |
| from Bio import Entrez, Medline | |
| import pandas as pd | |
| from transformers import pipeline | |
| from wordcloud import WordCloud | |
| import matplotlib.pyplot as plt | |
| # Configure the Entrez API | |
| Entrez.email = "[email protected]" # Change this to your actual email | |
| # Load the summarization model from Hugging Face | |
| summarizer = pipeline("summarization", model="facebook/bart-large-cnn") | |
| # Function to fetch papers by author | |
| def fetch_papers(author_name): | |
| search_query = f"{author_name}[Author]" | |
| handle = Entrez.esearch(db="pubmed", term=search_query, retmax=50, sort="relevance") | |
| record = Entrez.read(handle) | |
| handle.close() | |
| id_list = record["IdList"] | |
| return id_list | |
| # Function to fetch details of PubMed articles | |
| def fetch_paper_details(pubmed_ids): | |
| handle = Entrez.efetch(db="pubmed", id=pubmed_ids, rettype="medline", retmode="text") | |
| records = list(Medline.parse(handle)) | |
| handle.close() | |
| return records | |
| # Function to extract abstracts | |
| def extract_abstracts(records): | |
| abstracts = [] | |
| for record in records: | |
| abstract = record.get("AB", "") | |
| if abstract: | |
| abstracts.append(abstract) | |
| return abstracts | |
| # Function to extract keywords | |
| def extract_keywords(records): | |
| keywords = [] | |
| for record in records: | |
| keywords.extend(record.get("OT", [])) | |
| return keywords | |
| # Function to extract journal names and links | |
| def extract_journals(records): | |
| journals = [] | |
| for record in records: | |
| journal = record.get("JT", "") | |
| pmid = record.get("PMID", "") | |
| if journal and pmid: | |
| link = f"https://pubmed.ncbi.nlm.nih.gov/{pmid}/" | |
| journals.append((journal, link)) | |
| return journals | |
| # Function to generate a summary using the Hugging Face model | |
| def generate_summary(abstract): | |
| max_length = min(1024, len(abstract)) # Adjust max_length based on the abstract length | |
| summary = summarizer(abstract[:max_length], max_length=80, min_length=40, do_sample=False) | |
| return summary[0]['summary_text'] | |
| # Streamlit app | |
| st.title("Researcher Profile") | |
| # Input from user | |
| email = st.text_input("Enter your email") | |
| author_name = st.text_input("Enter the researcher's name") | |
| if st.button("Fetch Data"): | |
| if email and author_name: | |
| # Fetch papers | |
| pubmed_ids = fetch_papers(author_name) | |
| records = fetch_paper_details(pubmed_ids) | |
| if records: | |
| # Extract abstracts | |
| abstracts = extract_abstracts(records) | |
| if abstracts: | |
| # Generate and display the summary | |
| st.subheader("Research Summary") | |
| st.write("Summary of the last abstracts of the researcher") | |
| for i, abstract in enumerate(abstracts[:5]): | |
| summary = generate_summary(abstract) | |
| st.write(f"{i+1}. {summary}") | |
| # Extract keywords | |
| keywords = extract_keywords(records) | |
| if keywords: | |
| # Generate and display the word cloud | |
| st.subheader("Key of this researcher's papers") | |
| wordcloud = WordCloud(width=800, height=400, background_color='white').generate(' '.join(keywords)) | |
| plt.figure(figsize=(10, 5)) | |
| plt.imshow(wordcloud, interpolation='bilinear') | |
| plt.axis('off') | |
| st.pyplot(plt) | |
| else: | |
| st.write("No keywords found.") | |
| # Extract and display the last three journals with links | |
| journals = extract_journals(records) | |
| if journals: | |
| st.subheader("Last Journals Published In") | |
| for journal, link in journals[:3]: | |
| st.markdown(f"[{journal}]({link})") | |
| else: | |
| st.write("No journal information found.") | |
| else: | |
| st.write("No abstracts found.") | |
| else: | |
| st.write("No records found.") | |
| else: | |
| st.write("Please enter both email and researcher name.") | |