Relation extraction is an important semantic processing task in natu-ral language *** state-of-the-art systems usually rely on elaborately designed features,which are usually time-consuming and may lead to poor ***,mo...
详细信息
ISBN:
(纸本)9783030017156
Relation extraction is an important semantic processing task in natu-ral language *** state-of-the-art systems usually rely on elaborately designed features,which are usually time-consuming and may lead to poor ***,most existing systems adopt pipeline methods,which treat the task as two separated tasks,i.e.,named entity recognition and relation ***,the pipeline methods suffer two problems:(1)Pipeline mod-el over-simplifies the task to two independent parts.(2)The errors will be ac-cumulated from named entity recognition to relation ***,we present a novel joint model for entities and relations extraction based on multi-head attention,which avoids the problems in the pipeline methods and reduces the dependence on features *** experimental results show that our model achieves good performance without extra *** model reaches an F-score of 85.7%on SemEval-2010 relation extraction task 8,which has com-petitive performance without extra feature compared with previous joint *** publication,codes will be made publicly available.
In e-commerce websites,user-generated question-answering text pairs generally contain rich aspect information of *** this paper,we address a new task,namely Question-answering(QA)aspect classification,which aims to au...
详细信息
ISBN:
(纸本)9783030017156
In e-commerce websites,user-generated question-answering text pairs generally contain rich aspect information of *** this paper,we address a new task,namely Question-answering(QA)aspect classification,which aims to automatically classify the aspect category of a given QA text *** particular,we build a high-quality annotated corpus with specifically designed annotation guidelines for QA aspect *** this basis,we propose a hierarchical attention network to address the specific challenges in this new task in three ***,we firstly segment both question text and answer text into sentences,and then construct(sentence,sentence)units for each QA text ***,we leverage a QA matching attention layer to encode these(sentence,sentence)units in order to capture the aspect matching information between the sentence inside question text and the sentence inside answer ***,we leverage a self-matching attention layer to capture different importance degrees of different(sentence,sentence)units in each QA text *** results demonstrate that our proposed hierarchical attention network outperforms some strong baselines for QA aspect classification.
As an essential sub-task of frame-semantic parsing,Frame Identifica-tion(FI)is a fundamentally important research topic in shallow semantic ***,most existing work is based on sophisticated,hand-crafted fea-tures which...
详细信息
ISBN:
(纸本)9783030017156
As an essential sub-task of frame-semantic parsing,Frame Identifica-tion(FI)is a fundamentally important research topic in shallow semantic ***,most existing work is based on sophisticated,hand-crafted fea-tures which might not be compatible with FI *** that,they usually heavily rely on available natural language processing(NLP)toolkits and various lexical *** existing methods with hand-crafted features may not achieve satisfactory *** this paper,we propose a two-stage attention-based convolutional neural network(TSABCNN)to alleviate this problem and capture the most important context features for FI *** order to dynamically adjust the weight of each feature,we build two levels of attention over instances at input layer and pooling layer ***,the proposed model is an end-to-end learning framework which does not need any complicated NLP toolkits and feature engineering,and can be applied to any *** results on FrameNet and Chinese FrameNet(CFN)show the effectiveness of the proposed approach for the FI task.
Extracting term translation pairs is of great help for Chinese histori-cal classics translation since term translation is the most time-consuming and challenging part in the translation of historical ***,it is tough t...
详细信息
ISBN:
(纸本)9783030017156
Extracting term translation pairs is of great help for Chinese histori-cal classics translation since term translation is the most time-consuming and challenging part in the translation of historical ***,it is tough to recognize the terms directly from ancient Chinese due to the flexible syntactic of ancient Chinese and the word segmentation errors of ancient Chinese will lead to more errors in term translation *** most of the terms in ancient Chinese are still reserved in modern Chinese and the terms in modern Chinese are more easily to be identified,we propose a term translation extract-ing method using multi-features based on character-based model to extract his-torical term translation pairs from modern Chinese-English corpora instead of ancient Chinese-English ***,we first employ character-based BiLSTM-CRF model to identify historical terms in modern Chinese without word segmentation,which avoids word segmentation error spreading to the term *** we extract English terms according to initial capitaliza-tion *** last,we align the English and Chinese terms based on co-occurrence frequency and transliteration *** experiment on Shiji demonstrates that the performance of the proposed method is far superior to the traditional method,which confirms the effectiveness of using modern Chinese as a substitute.
暂无评论