Learning to Walk Structured Text Networks
ثبت نشده
چکیده
We propose representing a text corpus as a labeled directed graph, where nodes represent words and weighted edges represent the syntactic relations between them, as derived by dependency parsing. Given this graph, we adopt a graph-based similarity measure based on random walks to derive a similarity measure between words, and also use supervised learning to improve the derived similarity measure for a particular task. Empirical evaluation of the approach on the task of coordinate term extraction shows that the suggested framework improves on a state-of-theart distributional similarity measure.
منابع مشابه
Learning to Walk Structured Text Networks
We propose representing a text corpus as a labeled directed graph, where nodes represent words and weighted edges represent the syntactic relations between them, as derived by dependency parsing. Given this graph, we adopt a graph-based similarity measure based on random walks to derive a similarity measure between words, and also use supervised learning to improve the derived similarity measur...
متن کاملImprovement of generative adversarial networks for automatic text-to-image generation
This research is related to the use of deep learning tools and image processing technology in the automatic generation of images from text. Previous researches have used one sentence to produce images. In this research, a memory-based hierarchical model is presented that uses three different descriptions that are presented in the form of sentences to produce and improve the image. The proposed ...
متن کاملارائه مدلی برای استخراج اطلاعات از مستندات متنی، مبتنی بر متنکاوی در حوزه یادگیری الکترونیکی
As computer networks become the backbones of science and economy, enormous quantities documents become available. So, for extracting useful information from textual data, text mining techniques have been used. Text Mining has become an important research area that discoveries unknown information, facts or new hypotheses by automatically extracting information from different written documents. T...
متن کاملBibliometric Networks on Analyze Flipped Learning Research
Aim: The purpose is to provide a comprehensive overview of the current state of research in the field of flipped learning and classroom. It is a science metrics attempt to extract and analyze bibliographic networks based on the international scientific indexing (ISI) Methodology: Systematic search technique was applied: A set of scientific productions indexed in the field of flipped learning an...
متن کاملL2 Learners’ Lexical Inferencing: Perceptual Learning Style Preferences, Strategy Use, Density of Text, and Parts of Speech as Possible Predictors
This study was intended first to categorize the L2 learners in terms of their learning style preferences and second to investigate if their learning preferences are related to lexical inferencing. Moreover, strategies used for lexical inferencing and text related issues of text density and parts of speech were studied to determine their moderating effects and the best predictors of lexical infe...
متن کامل