The proceedings contain 20 papers. The topics discussed include: bootstrapping large-scale fine-grained contextual advertising classifier from Wikipedia;modeling graph structure via relative position for text generati...
ISBN:
(纸本)9781954085381
The proceedings contain 20 papers. The topics discussed include: bootstrapping large-scale fine-grained contextual advertising classifier from Wikipedia;modeling graph structure via relative position for text generation from knowledge graphs;entity prediction in knowledge graphs with joint embeddings;hierarchical graph convolutional networks for jointly resolving cross-document coreference of entity and event mentions;learning clause representation from dependency-anchor graph for connective prediction;selective attention basedgraph convolutional networks for aspect-level sentiment classification;keyword extraction using unsupervised learning on the document’s adjacency matrix;improving human text simplification with sentence fusion;and on geodesic distances and contextual embedding compression for text classification.
Event extraction is an important part of naturallanguage information extraction,and it’s widely employed in other naturallanguageprocessing tasks including question answering and machine reading ***,there is a lac...
详细信息
Event extraction is an important part of naturallanguage information extraction,and it’s widely employed in other naturallanguageprocessing tasks including question answering and machine reading ***,there is a lack of recent comprehensive survey papers on event *** the past few years,numerous high-quality and innovative event extraction methods have been proposed,making it necessary to consolidate these new developments with previous work in order to provide a clear overview for researchers and serve as a reference for future *** addition,event detection is a fundamental sub-task in event extraction,previous survey papers have often overlooked the related work on event ***,this paper aims to bridge these gaps by presenting a comprehensive survey of event extraction,including recent advancements and an analysis of previous research on event *** resources for event extraction are first introduced in this research,and then the numerous neural network models currently employed in event extraction tasks are divided into four types:word sequence-basedmethods,graph-based neural network methods,external knowledge-based approaches,and prompt-based *** compare and contrast them in depth,pointing out the flaws and difficulties with existing ***,we discuss the future of event extraction development.
Knowledge graphs (KGs) are popularly used to develop several intelligent applications. Revealing valuable knowledge hidden in these graphs opened up a branch of research, known as KG reasoning, aiming at predicting th...
详细信息
Knowledge graphs (KGs) are popularly used to develop several intelligent applications. Revealing valuable knowledge hidden in these graphs opened up a branch of research, known as KG reasoning, aiming at predicting the missing links. Some methods take advantage of external information such as entity description but at the cost of more computational complexity. Besides, most of the current techniques focus solely on local information in the KG. However, the learning process can utilise valuable global information in the entire graph. In this paper, we propose a Pattern-based Knowledge graph Completion (PKGC) method that consists of three phases. The first phase utilizes multi-source information and expands the KG using entity description as external information with efficient naturallanguageprocessing (NLP) techniques. In the second phase, we mine frequent patterns from the expanded KG, extract connections between them and assign entities to the patterns that construct the abstraction layer. based on the extracted patterns, connections, and entity assignments, a flow network is constructed on the abstraction layer in the third phase. We use global internal information, namely patterns, by adapting the minimum-cost circulation problem to the flow network. This way the links in a larger neighborhood are involved in the inference. We conducted experiments on the link prediction task and evaluated the training time on two benchmark datasets, WordNet and Freebase. Experiments have demonstrated that the proposed method is superior to the state-of-the-art methods and that pattern extraction is effective for knowledge graph completion tasks.
Transformers have gained prominence in naturallanguageprocessing due to their representational capabilities and performances. Transformers process naturallanguage as a sequence on finite context windows;however, gl...
详细信息
Transformers have gained prominence in naturallanguageprocessing due to their representational capabilities and performances. Transformers process naturallanguage as a sequence on finite context windows;however, global relationships among words beyond these windows cannot be completely modeled via sequence processing only. graph neural network (GNN) based models have been proposed to alleviate this problem, as they provide geometric extensions to neural networks, enabling models to learn associations within a text. However, regular graph-basedmethods ignore the sequential nature of underlying texts. In this paper, we propose EmoVis, the first generic graph-based neural network that utilizes visibility graphs, which converts classical time-series information to graph representations. We cast the problem as an emotion classification task, enabling the proposed model to learn associations between the labels and words in a sentence. Moreover, EmoVis can be used as a highly modular graph-based extension to any transformer-based model, significantly improving their performance and learning capabilities in various languages. We experimentally show that EmoVis enables transformer-based models to outperform the state-of-the-art baselines across three diverse datasets in different languages in the SemEval2018 competition datasets and the GoEmotions dataset.
The proceedings contain 15 papers. The topics discussed include: a graphical framework for contextual search and name disambiguation in email;graphbased semi-supervised approach for information extraction;graph-based...
The proceedings contain 15 papers. The topics discussed include: a graphical framework for contextual search and name disambiguation in email;graphbased semi-supervised approach for information extraction;graph-based text representation for novelty detection;measuring aboutness of an entity in a text;a study of two graph algorithms in topic-driven summarization;similarity between pairs of co-indexed trees for textual entailment recognition;learning of graph-based question answering rules;seeing stars when there aren’t many stars: graph-based semi-supervised learning for sentiment categorization;and random-walk term weighting for improved text classification.
The proceedings contain 14 papers. The topics discussed include: a survey of embedding models of entities and relationships for knowledge graph completion;graph-based aspect representation learning for entity resoluti...
ISBN:
(纸本)9781952148422
The proceedings contain 14 papers. The topics discussed include: a survey of embedding models of entities and relationships for knowledge graph completion;graph-based aspect representation learning for entity resolution;merge and recognize: a geometry and 2D context aware graph model for named entity recognition from visual documents;joint learning of the graph and the data representation for graph-based semi-supervised learning;contextual BERT: conditioning the language model using a global state;semi-supervised word sense disambiguation using example similarity graph;incorporating temporal information in entailment graph mining;relation specific transformations for open world knowledge graph completion;and PGL at Textgraphs 2020 Shared Task: explanation regeneration using language and graph learning methods.
The proceedings contain 10 papers. The topics discussed include: adapting predominant and novel sense discovery algorithms for identifying corpus-specific sense differences;merging knowledge bases in different languag...
ISBN:
(纸本)9781945626609
The proceedings contain 10 papers. The topics discussed include: adapting predominant and novel sense discovery algorithms for identifying corpus-specific sense differences;merging knowledge bases in different languages;parameter free hierarchical graph-based clustering for analyzing continuous word embeddings;spectral graph-based method of multimodal word embedding;extract with order for coherent multi-document summarization;work hard, play hard: email classification on the avocado and Enron corpora;a graphbased semi-supervised approach for analysis of derivational nouns in Sanskrit;and evaluating text coherence based on semantic similarity graph.
This paper presents our LLM-based system designed for the MEDIQA-CORR @ NAACLClinicalNLP 2024 Shared Task 3, focusing on medical error detection and correction in medical records. Our approach consists of three key co...
详细信息
graph-based techniques have gained traction for representing and analyzing data in various naturallanguageprocessing (NLP) tasks. Knowledge graph-basedlanguage representation models have shown promising results in ...
详细信息
Dependency parsing is a syntactic analysis method to analyze the dependency relationships between words in a sentence. The interconnection between words through dependency relationships is typical graph data. Traditio...
详细信息
暂无评论