Construction of the Literature Graph in Semantic Scholar

By Bridger Waleed Ammar, Dirk Groeneveld, Chandra Bhagavatula, Iz Beltagy, Miles Crawford, Doug Downey, Jason Dunkelberger, Ahmed Elgohary, Sergey Feldman, Vu A. Ha and 13 more. 2018

442 citations recorded by the source corpus.

We describe a deployed scalable system for organizing published scientific literature into a heterogeneous graph to facilitate algorithmic manipulation and discovery. The resulting literature graph consists of more than 280M nodes, representing papers, authors, entities and various interactions between them (e.g., authorships, citations, entity mentions). We reduce literature graph construction into familiar NLP tasks (e.g., entity extraction and linking), point out research challenges due to differences from standard formulations of these tasks, and report empirical results for each task. The methods described in this paper are used to enable semantic features in www.semanticscholar.org.

Subfields: Artificial Intelligence, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Molecular Biology

Citations in the corpus

The Values Encoded in Machine Learning Research (2021, methodology): “, LWPR [8], standard GPR [1], sparse online Gaussian process regression (OGP) [5] and υ-support vector regression (υ-SVR) [11], respectively.”
S2ORC: The Semantic Scholar Open Research Corpus (2020, methodology): “…graphs like AMiner’s Open Academic Graph (Tang et al., 2008), the Microsoft Academic Graph (MAG) (Shen et al., 2018), and the Semantic Scholar literature graph (Ammar et al., 2018), have had widespread application in bibliomet-rics, science-of-science, information retrieval, and network analysis.”
SPECTER: Document-level Representation Learning using Citation-informed Transformers (2020, methodology): “Training Data To train our model, we use a subset of the Semantic Scholar corpus (Ammar et al., 2018) consisting of about 146K query papers (around 26.7M tokens) with their corresponding outgoing citations, and we use an additional 32K papers for validation.”
Masader: Metadata Sourcing for Arabic Text and Speech Data Resources (2021, unknown): “A prominent example of such effort is Ammar et al. (2018).”
Multi-Task Identification of Entities, Relations, and Coreference for Scientific Knowledge Graph Construction (2018, methodology): “Using these datasets, neural models (Ammar et al., 2017, 2018; Luan et al., 2017b; Augenstein and Søgaard, 2017) are introduced for extracting scientiﬁc information.”
GORC: A large contextual citation graph of academic papers (2019, methodology): “We release the Semantic Scholar GORC dataset, a large contextual citation graph consisting of 81.1M papers across broad domains of science.”
Learning Dynamic Author Representations with Temporal Language Models (2019, background): “- The Semantic Scholar [38] corpus (S2) is composed of titles from scientific papers published in machine learning conferences and journals from 1985 to 2017, split by year (33 timesteps).”
Large language models are few-shot clinical information extractors (2022, background): “More recently, following BERT, many clinical and biomedical variations swiftly followed including ClinicalBERT, SciBERT, BioBERT, and PubMedBERT (Devlin et al., 2018; Alsentzer et al., 2019; Ammar et al., 2018; Lee et al., 2020; Gu et al., 2021).”
OAG-BERT: Pre-train Heterogeneous Entity-augmented Academic Language Models (2021, unknown): “SciBERT uses the semantic scholar corpus [1].”
Time Waits for No One! Analysis and Challenges of Temporal Misalignment (2021, methodology): “For adaptation to this domain, we collect unlabeled data from science documents available in Semantic Scholar’s corpus, 5 which yields 650k documents, spread over a 30-year period (Ammar et al., 2018).”
Boosting Low-Resource Biomedical QA via Entity-Aware Masking Strategies (2021, methodology): “SciBERT (Beltagy et al., 2019) follows the BERT’s masking strategy to pre-train the model from scratch using a scientific corpus composed of papers from Semantic Scholar (Ammar et al., 2018).”