Aspect’s extraction is a critical task in aspect-based sentiment analysis,including explicit and implicit aspects identification.While extensive research has identified explicit aspects,little effort has been put for...Aspect’s extraction is a critical task in aspect-based sentiment analysis,including explicit and implicit aspects identification.While extensive research has identified explicit aspects,little effort has been put forward on implicit aspects extraction due to the complexity of the problem.Moreover,existing research on implicit aspect identification is widely carried out on product reviews targeting specific aspects while neglecting sentences’dependency problems.Therefore,in this paper,a multi-level knowledge engineering approach for identifying implicit movie aspects is proposed.The proposed method first identifies explicit aspects using a variant of BiLSTM and CRF(Bidirectional Long Short Memory-Conditional Random Field),which serve as a memory to process dependent sentences to infer implicit aspects.It can identify implicit aspects from four types of sentences,including independent and three types of dependent sentences.The study is evaluated on a largemovie reviews dataset with 50k examples.The experimental results showed that the explicit aspect identification method achieved 89%F1-score and implicit aspect extraction methods achieved 76%F1-score.In addition,the proposed approach also performs better than the state-of-the-art techniques(NMFIAD andML-KB+)on the product review dataset,where it achieved 93%precision,92%recall,and 93%F1-score.展开更多
It is a debated topic if there are any observable precursor anomalies prior to the earthquake(EQ hereafter)and if the stronger EQ can be successfully predicted.During last few decades quite a lot of observable electro...It is a debated topic if there are any observable precursor anomalies prior to the earthquake(EQ hereafter)and if the stronger EQ can be successfully predicted.During last few decades quite a lot of observable electromagnetic(EM)precursors were published by using techniques equipped in either satellites or on ground-based stations.But there are only a few cases that the shortterm precursor anomalies of EM field before earthquakes were observed by using alternate EM fields on ground.This paper will present a new EM observation network built in recent years and show a new finding of EM field with the variation of a one-year cycle observed using the network.As an example,the short-term precursor anomalies of apparent resistivity before the Yangbi EQ(Ms 5.1)occurred on March 27,2017 in Yunnan Province will be studied.The observed anomalous phenomena indicate that the anomaly before the EQ can be captured only if reasonable effective methods including sophisticated analytical techniques are used,and it is believed that continuously observed data on the fixed observation network for a long time is an effective means for studying anomalies that appeared before earthquakes.This network can also play an important role in studying the EM environment from space.展开更多
The alternating electromagnetic(EM) field is one of the most sensitive physical fields related to earthquakes. There have been a number of publications reporting EM anomalies associated with earthquakes. With increasi...The alternating electromagnetic(EM) field is one of the most sensitive physical fields related to earthquakes. There have been a number of publications reporting EM anomalies associated with earthquakes. With increasing applications and research of artificial-source extremely low frequency EM and satellite EM technologies in earthquake studies, the amount of observed data from the alternating EM method increases rapidly and exponentially, so it is imperative to develop suitable and effective methods for processing and analyzing the influx of big data. This paper presents research on the self-adaptive filter and wavelet techniques and their applications to analyzing EM data obtained from ground measurements and satellite observations, respectively. Analysis results show that the self-adaptive filter method can identify both natural- and artificial-source EM signals, and enhance the ratio between signal and noise of EM field spectra, apparent resistivity, and others. The wavelet analysis is capable of detecting possible correlation between EM anomalies and seismic events. These techniques are effective in processing and analyzing massive data obtained from EM observations.展开更多
文摘Aspect’s extraction is a critical task in aspect-based sentiment analysis,including explicit and implicit aspects identification.While extensive research has identified explicit aspects,little effort has been put forward on implicit aspects extraction due to the complexity of the problem.Moreover,existing research on implicit aspect identification is widely carried out on product reviews targeting specific aspects while neglecting sentences’dependency problems.Therefore,in this paper,a multi-level knowledge engineering approach for identifying implicit movie aspects is proposed.The proposed method first identifies explicit aspects using a variant of BiLSTM and CRF(Bidirectional Long Short Memory-Conditional Random Field),which serve as a memory to process dependent sentences to infer implicit aspects.It can identify implicit aspects from four types of sentences,including independent and three types of dependent sentences.The study is evaluated on a largemovie reviews dataset with 50k examples.The experimental results showed that the explicit aspect identification method achieved 89%F1-score and implicit aspect extraction methods achieved 76%F1-score.In addition,the proposed approach also performs better than the state-of-the-art techniques(NMFIAD andML-KB+)on the product review dataset,where it achieved 93%precision,92%recall,and 93%F1-score.
基金National Development and Reform Committee of China(No.15212Z0000001)National Science Foundation of China(No.41374077)。
文摘It is a debated topic if there are any observable precursor anomalies prior to the earthquake(EQ hereafter)and if the stronger EQ can be successfully predicted.During last few decades quite a lot of observable electromagnetic(EM)precursors were published by using techniques equipped in either satellites or on ground-based stations.But there are only a few cases that the shortterm precursor anomalies of EM field before earthquakes were observed by using alternate EM fields on ground.This paper will present a new EM observation network built in recent years and show a new finding of EM field with the variation of a one-year cycle observed using the network.As an example,the short-term precursor anomalies of apparent resistivity before the Yangbi EQ(Ms 5.1)occurred on March 27,2017 in Yunnan Province will be studied.The observed anomalous phenomena indicate that the anomaly before the EQ can be captured only if reasonable effective methods including sophisticated analytical techniques are used,and it is believed that continuously observed data on the fixed observation network for a long time is an effective means for studying anomalies that appeared before earthquakes.This network can also play an important role in studying the EM environment from space.
基金supported by the National Natural Science Foundation of China(Grant Nos.41374077,41074047)CEA-NASCC Dragon Project Ⅲ(Grant No.10671)Special Public Benefit Program for Earthquake Study(Grant No.200808010)
文摘The alternating electromagnetic(EM) field is one of the most sensitive physical fields related to earthquakes. There have been a number of publications reporting EM anomalies associated with earthquakes. With increasing applications and research of artificial-source extremely low frequency EM and satellite EM technologies in earthquake studies, the amount of observed data from the alternating EM method increases rapidly and exponentially, so it is imperative to develop suitable and effective methods for processing and analyzing the influx of big data. This paper presents research on the self-adaptive filter and wavelet techniques and their applications to analyzing EM data obtained from ground measurements and satellite observations, respectively. Analysis results show that the self-adaptive filter method can identify both natural- and artificial-source EM signals, and enhance the ratio between signal and noise of EM field spectra, apparent resistivity, and others. The wavelet analysis is capable of detecting possible correlation between EM anomalies and seismic events. These techniques are effective in processing and analyzing massive data obtained from EM observations.
