A Comparison of Weak Supervision methods for Knowledge Base Construction
نویسندگان
چکیده
We present a comparison of weak and distant supervision methods for producing proxy examples for supervised relation extraction. We find that knowledge-based weak supervision tends to outperform popular distance supervision techniques, providing a higher yield of positive examples and more accurate models.
منابع مشابه
Learning Relational Dependency Networks for Relation Extraction
We consider the task of KBP slot filling – extracting relation information from newswire documents for knowledge base construction. We present our pipeline, which employs Relational Dependency Networks (RDNs) to learn linguistic patterns for relation extraction. Additionally, we demonstrate how several components such as weak supervision, word2vec features, joint learning and the use of human a...
متن کاملClinical Supervision System: A Method for Improving Educational Performance of Nursing Personnel
Introduction: Employing supervision methods which focuses on educational aspects may increase the effectiveness of nursing personnel's educational activities, especially patient education. The aim of this study was to investigate the effect of clinical supervision on the effectiveness of nurses' educational performance. Methods: This quasi-experimental study was conducted using static groups' ...
متن کاملA Comparison of the Metacognitive Knowledge of Weak and Strong Students in Problem Solving
This study sought to compare the metacognitive knowledge profile of weak and strong students in problem solving. For achieving this purpose, one hundred students of the second educational District of Tehran were chosen by cluster sampling. They responded individually to the questions of "the Metacognitive Questionnaire Interview" and in the next step their performance in the pendulum problem so...
متن کاملKnowledge-Based Weak Supervision for Information Extraction of Overlapping Relations
Information extraction (IE) holds the promise of generating a large-scale knowledge base from the Web’s natural language text. Knowledge-based weak supervision, using structured data to heuristically label a training corpus, works towards this goal by enabling the automated learning of a potentially unbounded number of relation extractors. Recently, researchers have developed multiinstance lear...
متن کاملPhysicians' and Clinical Coders' Knowledge about Cause of Death Rules for Death Certification in Hospitals Affiliated to Kashan University of Medical Sciences
Introduction: To clarify mortality trend, enhancing the knowledge of physicians and clinical coders who play their role in coding based on ICD book is crucial. This study was performed to investigate the knowledge of physicians and clinical coders on the rules and guidelines of determining mortality causes. Methods: This descriptive cross-sectional study was performed on 152 health care provid...
متن کامل