Targeted Sentiment Analysis Research Articles

Utilizing RoBERTa and XLM-RoBERTa pre-trained model for structured sentiment analysis

The surge in internet usage has amplified the trend of expressing sentiments across various platforms, particularly in e-commerce. Traditional sentiment analysis methods, such as aspect-based sentiment analysis (ABSA) and targeted sentiment analysis, fall short in identifying the relationships between opinion tuples. Moreover, conventional machine learning approaches often yield inadequate results. To address these limitations, this study introduces an approach that leverages the attention values of pre-trained RoBERTa and XLM-RoBERTa models for structured sentiment analysis. This method aims to predict all opinion tuples and their relationships collectively, providing a more comprehensive sentiment analysis. The proposed model demonstrates significant improvements over existing techniques, with the XLM-RoBERTa model achieving a notable sentiment graph F1 (SF1) score of 64.6% on the OpeNER<sub>EN</sub> dataset. Additionally, the RoBERTa model showed satisfactory performance on the multi-perspective question answer (MPQA) and DS<sub>Unis</sub> datasets, with SF1 scores of 25.3% and 29.9%, respectively, surpassing baseline models. These results underscore the potential of this proposed approach in enhancing sentiment analysis across diverse datasets, making it highly applicable for both academic research and practical applications in various industries.

Large Language Models in Targeted Sentiment Analysis for Russian

Large Language Models in Targeted Sentiment Analysis for Russian

Individual- vs. Multiple-Objective Strategies for Targeted Sentiment Analysis in Finances Using the Spanish MTSA 2023 Corpus

Multitarget sentiment analysis extracts the subjective polarity of text from multiple targets simultaneously in a given context. This approach is useful in finance, where opinions about different entities affect the target differently. Examples of possible targets are other companies and society. However, typical multitarget solutions are resource-intensive due to the need to deploy multiple classification models for each target. An alternative to this is the use of multiobjective training approaches, where a single model is capable of handling multiple targets. In this work, we propose the Spanish MTSACorpus 2023, a novel corpus for multitarget sentiment analysis in finance, and we evaluate its reliability with several large language models for multiobjective training. To this end, we compare three design approaches: (i) a Main Economic Target (MET) detection model based on token classification plus a multiclass classification model for sentiment analysis for each target; (ii) a MET detection model based on token classification but replacing the sentiment analysis models with a multilabel classification model; and (iii) using seq2seq-type models, such as mBART and mT5, to return a response sequence containing the MET and the sentiments of different targets. Based on the computational resources required and the performance obtained, we consider the fine-tuned mBART to be the best approach, with a mean F1 of 80.300%.

Evaluation of transformer models for financial targeted sentiment analysis in Spanish.

Nowadays, financial data from social media plays an important role to predict the stock market. However, the exponential growth of financial information and the different polarities of sentiment that other sectors or stakeholders may have on the same information has led to the need for new technologies that automatically collect and classify large volumes of information quickly and easily for each stakeholder. In this scenario, we conduct a targeted sentiment analysis that can automatically extract the main economic target from financial texts and obtain the polarity of a text towards such main economic target, other companies and society in general. To this end, we have compiled a novel corpus of financial tweets and news headlines in Spanish, constituting a valuable resource for the Spanish-focused research community. In addition, we have carried out a performance comparison of different Spanish-specific large language models, with MarIA and BETO achieving the best results. Our best result has an overall performance of 76.04%, 74.16%, and 68.07% in macro F1-score for the sentiment classification towards the main economic target, society, and other companies, respectively, and an accuracy of 69.74% for target detection. We have also evaluated the performance of multi-label classification models in this context and obtained a performance of 71.13%.

A Spanish dataset for Targeted Sentiment Analysis of political headlines

Los textos subjetivos han sido ampliamente estudiados debido a su potencial para influir en los comportamientos de la población. Si bien la mayoría de las investigaciones se han centrado en los textos generados por los usuarios en las redes sociales, otros tipo de textos ―como los titulares de noticias que expresan opiniones sobre ciertos temas― también pueden influir en los criterios de juicio durante las decisiones políticas. En este artículo abordamos la tarea de Análisis de sentimiento dirigido (Targeted Sentiment Analysis) sobre titulares de noticias relacionados con las elecciones presidenciales de Argentina de 2019, publicados por los principales medios de comunicación. Para facilitar la investigación en esta área, presentamos un conjunto de datos de polaridad de 1976 titulares que mencionan candidatos de esta elección. Nuestros experimentos con algoritmos del estado del arte basados en modelos de lenguaje muestran la utilidad de la información del objetivo para esta tarea. Publicamos nuestros datos y modelos de clasificación para facilitar la investigación futura.

A Lexicon Enhanced Collaborative Network for targeted financial sentiment analysis

The increasing interest around emotions in online texts creates the demand for financial sentiment analysis. Previous studies mainly focus on coarse-grained document-/sentence-level sentiment analysis, which ignores different sentiment polarities of various targets (e.g., company entities) in a sentence. To fill the gap, from a fine-grained target-level perspective, we propose a novel Lexicon Enhanced Collaborative Network (LECN) for targeted sentiment analysis (TSA) in financial texts. In general, the model designs a unified and collaborative framework that can capture the associations of targets and sentiment cues to enhance the overall performance of TSA. Moreover, the model dynamically incorporates sentiment lexicons to guide the sentiment classification, which cultivates the model faculty of understanding financial expressions. In addition, the model introduces a message selective-passing mechanism to adaptively control the information flow between two tasks, thereby improving the collaborative effects. To verify the effectiveness of LECN, we conduct experiments on four financial datasets, including SemEVAL2017 Task5 subset1, SemEVAL2017 Task5 subset2, FiQA 2018 Task1, and Financial PhraseBank. Results show that LECN achieves improvements over the state-of-art baseline by 1.66 p.p., 1.47 p.p., 1.94 p.p., and 1.88 p.p. in terms of F1-score. A series of further analyses also indicate that LECN has a better capacity for comprehending domain-specific expressions and can achieve the mutually beneficial effect between tasks.

AT-ODTSA: a Dataset of Arabic Tweets for Open Domain Targeted Sentiment Analysis

AT-ODTSA: a Dataset of Arabic Tweets for Open Domain Targeted Sentiment Analysis

Understand Students Feedback Using Bi-Integrated CRF Model Based Target Extraction

Educational institutions showing interest to find the opinion of the students about their course and the instructors to enhance the teaching-learning process. For this, most research uses sentiment analysis to track students’ behavior. Traditional sentence-level sentiment analysis focuses on the whole sentence sentiment. Previous studies show that the sentiments alone are not enough to observe the feeling of the students because different words express different sentiments in a sentence. There is a need to extract the targets in a given sentence which helps to find the sentiment towards those targets. Target extraction is the subtask of targeted sentiment analysis. In this paper, we proposed the innovative model to find the targets of the given sentence using Bi-Integrated Conditional Random Fields (CRF). A Parallel fusion neural network model is designed to perform this task. We evaluate the model using the Michigan dataset and we build a dataset for target extraction from student reviews. The experimental results show that our proposed fusion model achieves better results compared to baseline models.

A BERT Fine-tuning Model for Targeted Sentiment Analysis of Chinese Online Course Reviews

Accurate analysis of targeted sentiment in online course reviews helps in understanding emotional changes of learners and improving the course quality. In this paper, we propose a fine-tuned bidirectional encoder representation from transformers (BERT) model for targeted sentiment analysis of course reviews. Specifically, it consists of two parts: binding corporate rules — conditional random field (BCR-CRF) target extraction model and a binding corporate rules — double attention (BCR-DA) target sentiment analysis model. Firstly, based on a large-scale Chinese review corpus, intra-domain unsupervised training of a BERT pre-trained model (BCR) is performed. Then, a Conditional Random Field (CRF) layer is introduced to add grammatical constraints to the output sequence of the semantic representation layer in the BCR model. Finally, a BCR-DA model containing double attention layers is constructed to express the sentiment polarity of the course review targets in a classified manner. Experiments are performed on Chinese online course review datasets of China MOOC. The experimental results show that the F1 score of the BCR-CRF model reaches above 92%, and the accuracy of the BCR-DA model reaches above 72%.

Targeted Sentiment Classification Based on Attentional Encoding and Graph Convolutional Networks

Targeted sentiment classification aims to predict the emotional trend of a specific goal. Currently, most methods (e.g., recurrent neural networks and convolutional neural networks combined with an attention mechanism) are not able to fully capture the semantic information of the context and they also lack a mechanism to explain the relevant syntactical constraints and long-range word dependencies. Therefore, syntactically irrelevant context words may mistakenly be recognized as clues to predict the target sentiment. To tackle these problems, this paper considers that the semantic information, syntactic information, and their interaction information are very crucial to targeted sentiment analysis, and propose an attentional-encoding-based graph convolutional network (AEGCN) model. Our proposed model is mainly composed of multi-head attention and an improved graph convolutional network built over the dependency tree of a sentence. Pre-trained BERT is applied to this task, and new state-of-art performance is achieved. Experiments on five datasets show the effectiveness of the model proposed in this paper compared with a series of the latest models.

Embedding Projection for Targeted Cross-lingual Sentiment: Model Comparisons and a Real-World Study

Sentiment analysis benefits from large, hand-annotated resources in order to train and test machine learning models, which are often data hungry. While some languages, e.g., English, have a vast arrayof these resources, most under-resourced languages do not, especially for fine-grained sentiment tasks, such as aspect-level or targeted sentiment analysis. To improve this situation, we propose a cross-lingual approach to sentiment analysis that is applicable to under-resourced languages and takes into account target-level information. This model incorporates sentiment information into bilingual distributional representations, byjointly optimizing them for semantics and sentiment, showing state-of-the-art performance at sentence-level when combined with machine translation. The adaptation to targeted sentiment analysis on multiple domains shows that our model outperforms other projection-based bilingual embedding methods on binary targetedsentiment tasks. Our analysis on ten languages demonstrates that the amount of unlabeled monolingual data has surprisingly little effect on the sentiment results. As expected, the choice of a annotated source language for projection to a target leads to better results for source-target language pairs which are similar. Therefore, our results suggest that more efforts should be spent on the creation of resources for less similar languages tothose which are resource-rich already. Finally, a domain mismatch leads to a decreased performance. This suggests resources in any language should ideally cover varieties of domains.

Sparse attention based separable dilated convolutional neural network for targeted sentiment analysis

Long short-term memory networks (LSTM) and classical convolutional neural networks (CNN) are two critical methods for the task of targeted sentiment analysis, but LSTM are difficult to parallelize and time-inefficient, and classical CNN can only capture local semantic features. To this end, this paper first proposes a sparse attention based separable dilated convolutional neural network (SA-SDCCN), which consists of multichannel embedding layer, separable dilated convolution module, sparse attention layer, and output layer. Specifically, our work is mainly concentrated on the first three parts. In multichannel embedding layer, semantic and sentiment embeddings are incorporated into an embedding tensor, which builds richer representations over the input sequence. In separable dilated convolution module, long-range contextual semantic information is explored and multi-scale contextual semantic dependencies are aggregated simultaneously through diverse dilation rates. Moreover, the separable structure further reduces the model parameters. In sparse attention layer, sentiment-oriented components are noticed according to the features of specific target entity. Finally, some experiments on three benchmark datasets demonstrate that SA-SDCCN achieves comparable or even better performance than state-of-the-art methods in terms of higher parallelism and lower computational cost.

Granular computing-based multi-level interactive attention networks for targeted sentiment analysis

Targeted sentiment analysis has shown great promise in the field of e-commerce review processing. In recent years, research is ongoing to optimize the current targeted sentiment analysis approaches. This paper, from the innovation perspective, proposes a method referred to as granular computing-based multi-level interactive attention networks (GC-MIA). The method is constructed on the basis of granular computing. Each word is treated as an indivisible granule which constitutes the contexts and the target, respectively. Aiming at exploring the information within each word in the sentence, the multi-level interactive attention networks are established to make full use of every single granule. The representations of the target and the contexts are revised via multi-level interactive learning. Experimental results on the SemEval 2014 dataset are presented to verify the effectiveness of the GC-MIA model. The high accuracy indicates the significance of the proposed method in the targeted sentiment analysis.

Targeted Aspect-Based Sentiment Analysis via Embedding Commonsense Knowledge into an Attentive LSTM

Analyzing people’s opinions and sentiments towards certain aspects is an important task of natural language understanding. In this paper, we propose a novel solution to targeted aspect-based sentiment analysis, which tackles the challenges of both aspect-based sentiment analysis and targeted sentiment analysis by exploiting commonsense knowledge. We augment the long short-term memory (LSTM) network with a hierarchical attention mechanism consisting of a target-level attention and a sentence-level attention. Commonsense knowledge of sentiment-related concepts is incorporated into the end-to-end training of a deep neural network for sentiment classification. In order to tightly integrate the commonsense knowledge into the recurrent encoder, we propose an extension of LSTM, termed Sentic LSTM. We conduct experiments on two publicly released datasets, which show that the combination of the proposed attention architecture and Sentic LSTM can outperform state-of-the-art methods in targeted aspect sentiment tasks.

Learning Latent Sentiment Scopes for Entity-Level Sentiment Analysis

In this paper, we focus on the task of extracting named entities together with their associated sentiment information in a joint manner. Our key observation in such an entity-level sentiment analysis (a.k.a. targeted sentiment analysis) task is that there exists a sentiment scope within which each named entity is embedded, which largely decides the sentiment information associated with the entity. However, such sentiment scopes are typically not explicitly annotated in the data, and their lengths can be unbounded. Motivated by this, unlike traditional approaches that cast this problem as a simple sequence labeling task, we propose a novel approach that can explicitly model the latent sentiment scopes. Our experiments on the standard datasets demonstrate that our approach is able to achieve better results compared to existing approaches based on conventional conditional random fields (CRFs) and a more recent work based on neural networks.

Gated Neural Networks for Targeted Sentiment Analysis

Targeted sentiment analysis classifies the sentiment polarity towards each target entity mention in given text documents. Seminal methods extract manual discrete features from automatic syntactic parse trees in order to capture semantic information of the enclosing sentence with respect to a target entity mention. Recently, it has been shown that competitive accuracies can be achieved without using syntactic parsers, which can be highly inaccurate on noisy text such as tweets. This is achieved by applying distributed word representations and rich neural pooling functions over a simple and intuitive segmentation of tweets according to target entity mentions. In this paper, we extend this idea by proposing a sentence-level neural model to address the limitation of pooling functions, which do not explicitly model tweet-level semantics. First, a bi-directional gated neural network is used to connect the words in a tweet so that pooling functions can be applied over the hidden layer instead of words for better representing the target and its contexts. Second, a three-way gated neural network structure is used to model the interaction between the target mention and its surrounding contexts. Experiments show that our proposed model gives significantly higher accuracies compared to the current best method for targeted sentiment analysis.