Frequency Count of Papers Most Cited in a Specific Authors Publications

papers_most_often_cited_in_specific_author_papers.py

from semanticscholar import SemanticScholar
from collections import defaultdict
from tqdm import tqdm
import os

# Read key from env instead of hard-coding
#s2_api_key = os.environ.get("SEMANTIC_SCHOLAR_API_KEY")
sch = SemanticScholar(timeout=10)
# alec
author_ids = [2242015445,145151177,2251097727,2241538537,2244829783,2199254215,2309006642,2275054179,2312381751,2242589355]
# If author_ids is a single int, wrap it in a list
if isinstance(author_ids, int):
    author_ids = [author_ids]

# Batch fetch all authors in one call
authors = sch.get_authors([str(aid) for aid in author_ids])

papers = []
names = []

for author in authors:
    names.append(author.name)
    if len(author.papers) < author.paperCount:
        print(
            f"Warning: {author.name} has {author.paperCount} papers, "
            f"but only {len(author.papers)} were retrieved"
        )
    papers.extend(author.papers)

names = ", ".join(names)

papers_by_id = {paper["paperId"]: paper for paper in papers}

# --- Batch fetch full paper records for these paper IDs ---
BATCH_SIZE = 1000  # API allows up to 1000 IDs per call

for i in range(0, len(paper_ids), BATCH_SIZE):
    batch_ids = paper_ids[i : i + BATCH_SIZE]

    # Ask specifically for references, title, year to minimize payload
    batch_papers = sch.get_papers(
        batch_ids,
        fields=["title", "year", "references.title", "references.year"],
    )

    for authors_paper_full_rec in batch_papers:
        # authors_paper_full_rec is a Paper object

        # Get its references (may be list of dicts or objects depending on version)
        references = getattr(authors_paper_full_rec, "references", []) or []

        for ref in references:
            # Handle both dict-style and object-style references
            if isinstance(ref, dict):
                ref_data = ref
            else:
                ref_data = getattr(ref, "raw_data", {}) or {}

            ref_id = ref_data.get("paperId")
            if not ref_id:
                continue

            paperId_counts[ref_id] += 1
            # Store the short record so we can print title/year later
            paperId_short_recs[ref_id] = ref_data

# --- Sort and print results ---
paperId_counts_sorted = sorted(
    paperId_counts.items(), key=lambda item: -item[1]
)

for paperId, count in paperId_counts_sorted:
    rec = paperId_short_recs.get(paperId, {})
    title = rec.get("title", "UNKNOWN TITLE")
    year = rec.get("year", "UNKNOWN YEAR")
    print(f'{count}\t"{title}," {year}')