{"title":"Locating and parsing bibliographic references in HTML medical articles.","authors":"Jie Zou, Daniel Le, George R Thoma","doi":"10.1007/s10032-009-0105-9","DOIUrl":null,"url":null,"abstract":"<p><p>The set of references that typically appear toward the end of journal articles is sometimes, though not always, a field in bibliographic (citation) databases. But even if references do not constitute such a field, they can be useful as a preprocessing step in the automated extraction of other bibliographic data from articles, as well as in computer-assisted indexing of articles. Automation in data extraction and indexing to minimize human labor is key to the affordable creation and maintenance of large bibliographic databases. Extracting the components of references, such as author names, article title, journal name, publication date and other entities, is therefore a valuable and sometimes necessary task. This paper describes a two-step process using statistical machine learning algorithms, to first locate the references in HTML medical articles and then to parse them. Reference locating identifies the reference section in an article and then decomposes it into individual references. We formulate this step as a two-class classification problem based on text and geometric features. An evaluation conducted on 500 articles drawn from 100 medical journals achieves near-perfect precision and recall rates for locating references. Reference parsing identifies the components of each reference. For this second step, we implement and compare two algorithms. One relies on sequence statistics and trains a Conditional Random Field. The other focuses on local feature statistics and trains a Support Vector Machine to classify each individual word, followed by a search algorithm that systematically corrects low confidence labels if the label sequence violates a set of predefined rules. The overall performance of these two reference-parsing algorithms is about the same: above 99% accuracy at the word level, and over 97% accuracy at the chunk level.</p>","PeriodicalId":50277,"journal":{"name":"International Journal on Document Analysis and Recognition","volume":null,"pages":null},"PeriodicalIF":1.8000,"publicationDate":"2010-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1007/s10032-009-0105-9","citationCount":"28","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal on Document Analysis and Recognition","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1007/s10032-009-0105-9","RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 28
Abstract
The set of references that typically appear toward the end of journal articles is sometimes, though not always, a field in bibliographic (citation) databases. But even if references do not constitute such a field, they can be useful as a preprocessing step in the automated extraction of other bibliographic data from articles, as well as in computer-assisted indexing of articles. Automation in data extraction and indexing to minimize human labor is key to the affordable creation and maintenance of large bibliographic databases. Extracting the components of references, such as author names, article title, journal name, publication date and other entities, is therefore a valuable and sometimes necessary task. This paper describes a two-step process using statistical machine learning algorithms, to first locate the references in HTML medical articles and then to parse them. Reference locating identifies the reference section in an article and then decomposes it into individual references. We formulate this step as a two-class classification problem based on text and geometric features. An evaluation conducted on 500 articles drawn from 100 medical journals achieves near-perfect precision and recall rates for locating references. Reference parsing identifies the components of each reference. For this second step, we implement and compare two algorithms. One relies on sequence statistics and trains a Conditional Random Field. The other focuses on local feature statistics and trains a Support Vector Machine to classify each individual word, followed by a search algorithm that systematically corrects low confidence labels if the label sequence violates a set of predefined rules. The overall performance of these two reference-parsing algorithms is about the same: above 99% accuracy at the word level, and over 97% accuracy at the chunk level.
期刊介绍:
The large number of existing documents and the production of a multitude of new ones every year raise important issues in efficient handling, retrieval and storage of these documents and the information which they contain. This has led to the emergence of new research domains dealing with the recognition by computers of the constituent elements of documents - including characters, symbols, text, lines, graphics, images, handwriting, signatures, etc. In addition, these new domains deal with automatic analyses of the overall physical and logical structures of documents, with the ultimate objective of a high-level understanding of their semantic content. We have also seen renewed interest in optical character recognition (OCR) and handwriting recognition during the last decade. Document analysis and recognition are obviously the next stage.
Automatic, intelligent processing of documents is at the intersections of many fields of research, especially of computer vision, image analysis, pattern recognition and artificial intelligence, as well as studies on reading, handwriting and linguistics. Although quality document related publications continue to appear in journals dedicated to these domains, the community will benefit from having this journal as a focal point for archival literature dedicated to document analysis and recognition.