Abstract
The scientific literature is growing exponentially, and professionals are no more able to cope with the current amount of publications. Text mining provided in the past methods to retrieve and extract information from text; however, most of these approaches ignored tables and figures. The research done in mining table data still does not have an integrated approach for mining that would consider all complexities and challenges of a table. Our research is examining the methods for extracting numerical (number of patients, age, gender distribution) and textual (adverse reactions) information from tables in the clinical literature. We present a requirement analysis template and an integral methodology for information extraction from tables in clinical domain that contains 7 steps: (1) table detection, (2) functional processing, (3) structural processing, (4) semantic tagging, (5) pragmatic processing, (6) cell selection and (7) syntactic processing and extraction. Our approach performed with the F-measure ranged between 82 and 92%, depending on the variable, task and its complexity.
Highlights
The literature in the biomedical domain is growing exponentially
Fields of text mining and natural language processing provide tools and methodologies that can help with retrieving relevant information
In order to examine why our method extracted the number of patients only from 26% of documents, we examined a sample of 25 documents from which the number of patients was not extracted and found following reasons:
Summary
The literature in the biomedical domain is growing exponentially. Currently, there are over 26 million articles indexed in MEDLINE [35]. Most of the current approaches are limited to the textual body of articles, usually ignoring figures, tables and other semi-structured presentation formats of information. Textual content is usually dense, containing ambiguous short chunks of text with the use of acronyms and abbreviations. This is especially true in biomedical publications. In addition to natural language processing challenges, make it hard to understand the structure and the information that the table introduces. Information extraction from tables requires multilayered analysis that will include functional, structural, pragmatic, syntactic and semantic analysis. Our research is focusing on the task of extracting numerical and textual information from tables. We present a framework for information extraction from tables in biomedical documents. We compare a machine learning approach to a rule-based approach to identify cells with information of interest and evaluate how and where machine learning can help efficient information extraction from tables
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call