Visit the GROBID documentation for more detailed information.
GROBID (or Grobid, but not GroBid nor GroBiD) means GeneRation Of BIbliographic Data.
GROBID is a machine learning library for extracting, parsing and re-structuring raw documents such as PDF into structured XML/TEI encoded documents with a particular focus on technical and scientific publications. First developments started in 2008 as a hobby. In 2011 the tool has been made available in open source. Work on GROBID has been steady as a side project since the beginning and is expected to continue as such.
The following functionalities are available:
- Header extraction and parsing from article in PDF format. The extraction here covers the usual bibliographical information (e.g. title, abstract, authors, affiliations, keywords, etc.).
- References extraction and parsing from articles in PDF format, around .87 F1-score against on an independent PubMed Central set of 1943 PDF containing 90,125 references, and around .90 on a similar bioRxiv set of 2000 PDF (using the Deep Learning citation model). All the usual publication metadata are covered (including DOI, PMID, etc.).
- Citation contexts recognition and resolution of the full bibliographical references of the article. The accuracy of citation contexts resolution is between .76 and .91 F1-score depending on the evaluation collection (this corresponds to both the correct identification of the citation callout and its correct association with a full bibliographical reference).
- Full text extraction and structuring from PDF articles, including a model for the overall document segmentation and models for the structuring of the text body (paragraph, section titles, reference and footnote callouts, figures, tables, etc.).
- PDF coordinates for extracted information, allowing to create "augmented" interactive PDF based on bounding boxes of the identified structures.
- Parsing of references in isolation (above .90 F1-score at instance-level, .95 F1-score at field level, using the Deep Learning model).
- Parsing of names (e.g. person title, forenames, middle name, etc.), in particular author names in header, and author names in references (two distinct models).
- Parsing of affiliation and address blocks.
- Parsing of dates, ISO normalized day, month, year.
- Consolidation/resolution of the extracted bibliographical references using the biblio-glutton service or the CrossRef REST API. In both cases, DOI/PMID resolution performance is higher than 0.95 F1-score from PDF extraction.
- Extraction and parsing of patent and non-patent references in patent publications.
In a complete PDF processing, GROBID manages 55 final labels used to build relatively fine-grained structures, from traditional publication metadata (title, author first/last/middle names, affiliation types, detailed address, journal, volume, issue, pages, DOI, PMID, etc.) to full text structures (section title, paragraph, reference markers, head/foot notes, figure captions, etc.).
GROBID includes a comprehensive web service API, Docker images, batch processing, a JAVA API, a generic training and evaluation framework (precision, recall, etc., n-fold cross-evaluation), systematic end-to-end benchmarking on thousand documents and the semi-automatic generation of training data.
GROBID can be considered as production ready. Deployments in production includes ResearchGate, Semantic Scholar, HAL Research Archive, scite.ai, Academia.edu, Internet Archive Scholar, INIST-CNRS, CERN (Invenio), and many more. The tool is designed for speed and high scalability in order to address the full scientific literature corpus.
GROBID should run properly "out of the box" on Linux (64 bits) and macOS (Intel and ARM). We cannot ensure currently support for Windows as we did before (help welcome!).
GROBID uses Deep Learning models relying on the DeLFT library, a task-agnostic Deep Learning framework for sequence labelling and text classification, via JEP. GROBID can run Deep Learning architectures (RNN or transformers with or without layout feature channels) or with feature engineered CRF (default), or any mixtures of CRF and DL to balance scalability and accuracy. These models use joint text and visual/layout information provided by pdfalto.
Note that by default the Deep Learning models are not used, only CRF are selected in the default configuration to accommodate "out of the box" hardware. For improved accuracy, you need to select the Deep Learning models to be used in the GROBID configuration file, according to your need and hardware capacities (in particular GPU availability and runtime requirements). Some GROBID Deep Learning models perform significantly better than default CRF, in particular for bibliographical reference parsing, so it is recommended to consider selecting them to use this tool appropriately.
For testing purposes, two public GROBID demo servers are available thanks to HuggingFace, hosted as spaces.
A GROBID demo server with a combination of Deep Learning models and CRF models is available at the following address: https://kermitt2-grobid.hf.space/ or at https://huggingface.co/spaces/kermitt2/grobid. This demo runs however on CPU only. If you have GPU for your own server deployment, it will be significantly faster.
A faster demo with CRF only is available at https://kermitt2-grobid-crf.hf.space/ or https://huggingface.co/spaces/kermitt2/grobid-crf. However, accuracy is lower.
The Web services are documented here.
Warning: Some quota and query limitation apply to the demo server! Please be courteous and do not overload the demo server. For any serious works, you will need to deploy and use your own Grobid server, see the GROBID and Docker containers documentation for doing that easily and activate some Deep Learning models.
![Try in PWD](https://raw.githubusercontent.com/play-with-docker/stacks/master/assets/images/button.png)
Wait for 30 seconds for Grobid container to be created before opening a browser tab on port 8080. This demo container runs only with CRF models. Note that there is an additional 60s needed when processing a PDF for the first time for the loading of the models on the "cold" container. Then this Grobid container is available just for you during 4 hours.
For facilitating the usage GROBID service at scale, we provide clients written in Python, Java, node.js using the web services for parallel batch processing:
- Python GROBID client (the most complete one in term of supported services and options)
- Java GROBID client
- Node.js GROBID client
All these clients will take advantage of the multi-threading for scaling large set of PDF processing. As a consequence, they will be much more efficient than the batch command lines (which use only one thread) and should be preferred.
For example, we have been able to run the complete full-text processing at around 10.6 PDF per second (around 915,000 PDF per day, around 20M pages per day) with the node.js client listed above during one week on one 16 CPU machine (16 threads, 32GB RAM, no SDD, articles from mainstream publishers), see here (11.3M PDF were processed in 6 days by 2 servers without interruption).
In addition, a Java example project is available to illustrate how to use GROBID as a Java library: https://github.com/kermitt2/grobid-example. The example project is using GROBID Java API for extracting header metadata and citations from a PDF and output the results in BibTeX format.
Finally, the following python utilities can be used to create structured full text corpora of scientific articles. The tool simply takes a list of strong identifiers like DOI or PMID, performing the identification of online Open Access PDF, full text harvesting, metadata aggregation and Grobid processing in one workflow at scale: article-dataset-builder
Visit the documentation page describing the system. To summarize, the key design principles of GROBID are:
-
GROBID uses a cascade of sequence labeling models to parse a document.
-
The different models do not work on text, but on Layout Tokens to exploit various visual/layout information available for every tokens.
-
GROBID does not use training data derived from existing publisher XML documents, but small, high quality sets of manually labeled training data.
-
Technical choices and default settings are driven by the ability to process PDF quickly, with commodity hardware and with good parallelization and scalability capacities.
Detailed end-to-end benchmarking are available GROBID documentation and continuously updated.
A series of additional modules have been developed for performing structure aware text mining directly on scholar PDF, reusing GROBID's PDF processing and sequence labelling weaponry:
- software-mention: recognition of software mentions and associated attributes in scientific literature
- datastet: identification of named and implicit research datasets and associated attributes in scientific articles
- grobid-quantities: recognition and normalization of physical quantities/measurements
- grobid-superconductors: recognition of superconductor material and properties in scientific literature
- entity-fishing, a tool for extracting Wikidata entities from text and document, which can also use Grobid to pre-process scientific articles in PDF, leading to more precise and relevant entity extraction and the capacity to annotate the PDF with interactive layout
- dataseer-ml: identification of sections and sentences introducing datasets in a scientific article, and classification of the type of these datasets
- grobid-ner: named entity recognition
- grobid-astro: recognition of astronomical entities in scientific papers
- grobid-bio: a toy bio-entity tagger using BioNLP/NLPBA 2004 dataset
- grobid-dictionaries: structuring dictionaries in raw PDF format
See the Changelog.
GROBID is distributed under Apache 2.0 license.
The documentation is distributed under CC-0 license and the annotated data under CC-BY license.
If you contribute to GROBID, you agree to share your contribution following these licenses.
Main author and contact: Patrice Lopez ([email protected])
ej-technologies provided us a free open-source license for its Java Profiler. Click the JProfiler logo below to learn more.
If you want to cite this work, please refer to the present GitHub project, together with the Software Heritage project-level permanent identifier. For example, with BibTeX:
@misc{GROBID,
title = {GROBID},
howpublished = {\url{https://github.com/kermitt2/grobid}},
publisher = {GitHub},
year = {2008--2023},
archivePrefix = {swh},
eprint = {1:dir:dab86b296e3c3216e2241968f0d63b68e8209d3c}
}
See the GROBID documentation for more related resources.