Sentiment analysis aims to systematically study affective states and subjective information in digital text through computational methods. Aspect Sentiment Triplet Extraction (ASTE), a subtask of sentiment analysis, aims to extract aspect term, sentiment and opinion term triplets from sentences. However, some ASTE’s extracted triplets are not self-contained, as they reflect the sentence’s sentiment toward the aspect term, not the sentiment between the aspect and opinion terms. These triplets are not only unhelpful to people, but can also be detrimental to downstream tasks. In this paper, we introduce a more nuanced task, Aspect–Sentiment–Opinion Triplet Extraction (ASOTE), which also extracts aspect term, sentiment and opinion term triplets. However, the sentiment in a triplet extracted with ASOTE is the sentiment of the aspect term and opinion term pair. We build four datasets for ASOTE. A Position-aware BERT-based Framework (PBF) is proposed to address ASOTE. PBF first extracts aspect terms from sentences. For each extracted aspect term, PBF generates an aspect term-specific sentence representation, considering the aspect term’s position. It then extracts associated opinion terms and predicts the sentiments of the aspect–opinion term pairs based on the representation. In the experiments on the four datasets, PBF has set a benchmark performance on the novel ASOTE task.

In order to solve the problems of insufficient information extraction and poor classification effect of a single deep learning model, this paper proposes a hybrid multi-neural network CBOW-BiLSTM-Attention-PreambleCNN model(The CBA-PreambleCNN for short, PreambleCNN is the name of the improved TextCNN). The model uses Word2Vec as the word embedding layer to obtain the vector representation of the word, and then feeds the Bidirectional Long Short-Term Memory (BI-LSTM) network to capture the global information of the text, and then uses the Attention mechanism to make the word get different weights. Finally, it is fed into the improved Text Convolutional Neural Network (TextCNN fused with previous information, named PreambleCNN) to obtain the topic features of the Text, and the obtained feature vectors are fed into the softmax function for classification. After comparison with other models, this model has achieved good classification effect, and achieved 93.47% and 84.65% accuracy on Sina news dataset and Sohu news dataset, respectively. To some extent, this model solves the problem of insufficient information captured by a single model.

The aspect-level sentiment analysis study ignores the combination of syntactic distance, physical distance and lexicality, and only considers the syntactic structure of the sentence, lacking the mining of the complete syntactic structure and deep sentiment polarity knowledge, we propose the aspect-level sentiment analysis model FE-GCN based on feature enhancement and graph convolutional network. First the hybrid encoding part incorporates lexical features, physical distance features and syntactic distance features and feeds into the output of BiGRU, then the graph convolutional network layer performs aspect weighting enhancement on the tree structure and incorporates knowledge of sentiment polarity to obtain important information, and finally the sentiment category of aspect words is obtained after the classification layer. Experiments were conducted on three public datasets, Lap14, Rest14 and Twitter, with accuracy rates of 84.49%, 87.99% and 78.48%, respectively, indicating that FE-GCN can better integrate semantic, lexical, physical distance, syntactic distance and sentiment polarity knowledge and can store more complete syntactic structures, and has some advantages in the field of aspect-level sentiment analysis.

In this paper, we propose a graceful sentence recognition model based on lexical features and CNN-Att-BiGRU. Firstly, the lexical items in the sentences and the lexical features corresponding to each lexical item are vectorized and fused to represent them; secondly, a convolutional neural network (CNN) introducing Attention Mechanisms is used to obtain the local features of the fused word vectors; a bi-directional gated recurrent unit (BiGRU) is used to serialize the local features to represent them to obtain the global features; finally, they are connected to the Softmax classifier for graceful sentence recognition. Experiments show that the graceful sentence recognition model based on CNN-Att-BiGRU has the highest recognition precision of graceful sentences, reaching 89.46%, with an F1 value of 82.38%. After adding the fused word vector, the F1 value improved by 0.67.

Aspect sentiment triplet extraction (ASTE) aims to extract aspect term, sentiment and opinion term triplets from sentences. Since the initial datasets used to evaluate models on ASTE had flaws, several studies later corrected the initial datasets and released new versions of the datasets independently. As a result, different studies select different versions of datasets to evaluate their methods, which makes ASTE-related works hard to follow. In this paper, we analyze the relation between different versions of datasets and suggest that the entire-space version should be used for ASTE. Besides the sentences containing triplets and the triplets in the sentences, the entire-space version additionally includes the sentences without triplets and the aspect terms which do not belong to any triplets. Hence, the entire-space version is consistent with real-world scenarios and evaluating models on the entire-space version can better reflect the models' performance in real-world scenarios. In addition, experimental results show that evaluating models on non-entire-space datasets inflates the performance of existing models and models trained on the entire-space version can obtain better performance.

Target-oriented opinion words extraction (TOWE) is a subtask of aspect-based sentiment analysis (ABSA). Given a sentence and an aspect term occurring in the sentence, TOWE extracts the corresponding opinion words for the aspect term. TOWE has two types of instance. In the first type, aspect terms are associated with at least one opinion word, while in the second type, aspect terms do not have corresponding opinion words. However, previous researches trained and evaluated their models with only the first type of instance, resulting in a sample selection bias problem. Specifically, TOWE models were trained with only the first type of instance, while these models would be utilized to make inference on the entire space with both the first type of instance and the second type of instance. Thus, the generalization performance will be hurt. Moreover, the performance of these models on the first type of instance cannot reflect their performance on entire space. To validate the sample selection bias problem, four popular TOWE datasets containing only aspect terms associated with at least one opinion word are extended and additionally include aspect terms without corresponding opinion words. Experimental results on these datasets show that training TOWE models on entire space will significantly improve model performance and evaluating TOWE models only on the first type of instance will overestimate model performance.

Aspect Sentiment Triplet Extraction (ASTE) aims to extract aspect term (aspect), sentiment and opinion term (opinion) triplets from sentences and can tell a complete story, i.e., the discussed aspect, the sentiment toward the aspect, and the cause of the sentiment. ASTE is a charming task, however, one triplet extracted by ASTE only includes one opinion of the aspect, but an aspect in a sentence may have multiple corresponding opinions and one opinion only provides part of the reason why the aspect has this sentiment, as a consequence, some triplets extracted by ASTE are hard to understand, and provide erroneous information for downstream tasks. In this paper, we introduce a new task, named Aspect Sentiment Multiple Opinions Triplet Extraction (ASMOTE). ASMOTE aims to extract aspect, sentiment and multiple opinions triplets. Specifically, one triplet extracted by ASMOTE contains all opinions about the aspect and can tell the exact reason that the aspect has the sentiment. We propose an Aspect-Guided Framework (AGF) to address this task. AGF first extracts aspects, then predicts their opinions and sentiments. Moreover, with the help of the proposed Sequence Labeling Attention(SLA), AGF improves the performance of the sentiment classification using the extracted opinions. Experimental results on multiple datasets demonstrate the effectiveness of our approach .

Emotion recognition is the task of recognizing a person’s emotional state. EEG, as a physiological signal, can provide more detailed and complex information for emotion recognition task. Meanwhile, EEG can’t be changed and hidden intentionally makes EEG-based emotion recognition achieve more effective and reliable result. Unfortunately, due to the cost of data collection, most EEG datasets have small number of EEG data. The lack of data makes it difficult to predict the emotion states with the deep models, which requires enough number of training data. In this paper, we propose to use a simple data augmentation method to address the issue of data shortage in EEG-based emotion recognition. In experiments, we explore the performance of emotion recognition with the shallow and deep computational models before and after data augmentation on two standard EEG-based emotion datasets. Our experimental results show that the simple data augmentation method can improve the performance of emotion recognition based on deep models effectively.