Vi-AbSQA: Multi-task Prompt Instruction Tuning Model for Vietnamese Aspect-based Sentiment Quadruple Analysis

Aspect-based sentiment analysis (ABSA) is currently among the most vigorous areas in natural language processing (NLP). Individuals, private and government institutions are increasingly using media sources for decision making. In the last decade, aspect extraction has been the most essential phase of sentiment analysis (SA) to conduct an abridged sentiment classification. However, previous studies on sentiment analysis mostly focused on explicit aspects extraction with limited work on implicit aspects. To the best of our knowledge, this is the first systematic review that covers implicit, explicit, and the combination of both implicit and explicit aspect extractions. Therefore, this systematic review has been conducted to, 1) identify techniques used for extracting implicit, explicit, or both implicit and explicit aspects; 2) analyze the various evaluation metrics, data domains, and languages involved in the implicit and explicit aspect extraction in sentiment analysis from years 2008 to 2019; 3) identify the key challenges associated with the techniques based on the result of a comprehensive comparative analysis; and finally, 4) highlight the feasible opportunities for future research directions. This review can be used to assist novice and prominent researchers to understand the concept of both implicit and explicit aspect extractions in aspect-based sentiment analysis domain.

Sentiment analysis is becoming rapidly important for exploring social media Bangla text. The lack of sufficient resources is considered to be an important challenge for Aspect Based Sentiment Analysis (ABSA) of the Bangla language. The ABSA is a technique that divides the text and defines its sentiment based on its aspects. In this paper, we developed a high-quality Bangla ABSA annotated dataset namely BANGLA_ABSA. The datasets are labelled with aspects category and their respective sentiment polarity to do the ABSA task in Bangla. Four open domains namely Restaurant, Movie, Mobile phone, and Car are considered to make the dataset. The datasets are called Restaurant_ABSA, Movie_ABSA, Mobile_phone_ABSA, and Car_ABSA respectively that contain 801, 800, 975, and 1149 comments. All the comments are either complex or compound sentences. We created the dataset manually and annotated the same by exerting opinions. We organized the dataset as three tuples in Excel format namely 〈Id, Comment, {Aspect category, Sentiment Polarity}〉. These data are very important that facilitate the efficient handling of sentiment for any machine learning and deep learning research, especially for Bangla text.

Aspect-Based Sentiment Analysis (ABSA) is one of the highly challenging tasks in natural language processing. It extracts fine-grained sentiment information in user-generated reviews, as it aims at predicting the polarities towards predefined aspect categories or relevant entities in free text. Previous deep learning approaches usually rely on large-scale pre-trained language models and the attention mechanism, which applies the complete computed attention weights and does not place any restriction on the attention assignment. We argue that the original attention mechanism is not the ideal configuration for ABSA, as for most of the time only a small portion of terms are strongly related to the sentiment polarity of an aspect or entity. In this paper, we propose a masked attention mechanism customized for ABSA, with two different approaches to generate the mask. The first method sets an attention weight threshold that is determined by the maximum of all weights, and keeps only attention scores above the threshold. The second selects the top words with the highest weights. Both remove the lower score parts that are assumed to be less relevant to the aspect of focus. By ignoring part of input that is claimed irrelevant, a large proportion of input noise is removed, keeping the downstream model more focused and reducing calculation cost. Experiments on the Multi-Aspect Multi-Sentiment (MAMS) and SemEval-2014 datasets show significant improvements over state-of-the-art pre-trained language models with full attention, which displays the value of the masked attention mechanism. Recent work shows that simple self-attention in Transformer quickly degenerates to a rank-1 matrix, and masked attention may be another cure for that trend.

Aspect-based sentiment analysis refers to the task of determining the sentiment polarity associated with particular aspects mentioned in a sentence or document. Previous studies have used attention-based neural network models to connect aspect terms with context words, but these models often perform poorly due to limited interaction between aspect terms and opinion words. Furthermore, these models typically focus only on explicitly stated aspect objects, which can be overly restrictive in certain scenarios. Current sentiment analysis methods that rely on aspect categories also often fail to consider the implicit placement of aspect-category information within the context. While existing models may produce strong results, they often lack domain knowledge. To address these issues, this study proposes an Aspect-position and Entity-oriented Knowledge Convolutional Graph (APEKCG) consisting of two modules: the Aspect position-aware module (APA) and the Entity oriented Knowledge Dependency Convolutional Graph (EKDCG). The APA module is designed to integrate aspect-specific sentiment features for sentiment classification by incorporating information about aspect categories into different parts of the context. The EKDCG module incorporates entity-oriented knowledge, dependency labels, and syntactic path using a dependence graph. Experimental results on five benchmarks Natural Language Processing (NLP) datasets of the English language demonstrate the effectiveness of the proposed APEKCG framework. Furthermore, the APEKCG outperformed previous state-of-the-art models with its accuracy, achieving 89.13%, 84.32%, 89.02%, 79.64%, and 90.22% on the MAMS, Laptop, Restaurant, AWARE, and SemEval-15&16 datasets, respectively.

The advent of pre-trained language models has directed a new era of Natural Language Processing (NLP), enabling us to create powerful language models. Among these models, Transformer-based models like BERT have grown in popularity due to their cutting-edge effectiveness. However, these models heavily rely on resource-intensive languages, forcing other languages into multilingual models(mBERT). The two fundamental challenges with mBERT become significantly more challenging in a resource-constrained language like Bangla. It was trained on a limited and organized dataset and contained weights for all other languages. Besides, current research on other languages suggests that a language-specific BERT model will exceed multilingual ones. This paper introduces Bangla-BERT^a, a monolingual BERT model for the Bangla language. Despite the limited data available for NLP tasks in Bangla, we perform pre-training on the largest Bangla language model dataset, BanglaLM, which we constructed using 40 GB of text data. Bangla-BERT achieves the highest results in all datasets and vastly improves the state-of-the-art performance in binary linguistic classification, multilabel extraction, and named entity recognition, outperforming multilingual BERT and other previous research. The pre-trained model is assessed against several non-contextual models such as Bangla fasttext and word2vec the downstream tasks. Finally, this model is evaluated by transfer learning based on hybrid deep learning models such as LSTM, CNN, and CRF in NER, and it is observed that Bangla-BERT outperforms state-of-the-art methods. The proposed Bangla-BERT model is assessed by using benchmark datasets, including Banfakenews, Sentiment Analysis on Bengali News Comments, and Cross-lingual Sentiment Analysis in Bengali. Finally, it is concluded that Bangla-BERT surpasses all prior state-of-the-art results by 3.52%, 2.2%, and 5.3%.

Aspect-Based Sentiment Analysis (ABSA) plays a vital role in interpreting user opinions in customer reviews by identifying sentiment polarity linked to specific service aspects. However, traditional ABSA methods largely focus on explicit aspect mentions, often neglecting implicit aspects that are contextually implied. This limitation is especially critical in the hospitality domain, where customer reviews frequently include indirect references to hotel features. This study proposes an implicit aspect analysis framework for sentiment classification using hotel reviews, aiming to improve the detection and interpretation of implicit aspect expressions. A dataset of 2,000 TripAdvisor hotel reviews was collected and manually annotated by linguistic experts for four aspects: staff service, cleanliness, value for money, and location convenience. Each sentence was labeled as either explicitly or implicitly expressing an aspect, with associated sentiment polarity (positive or negative). A two-stage framework was developed: the first stage employed Word2Vec Skip-gram modeling and TF-IDF to construct an implicit aspect corpus; the second stage implemented machine learning and deep learning classifiers—including SVM, LR, MNB, RF, CNN, and LSTM—to predict sentiment. The hybrid feature representation combining TF-IDF and Word2Vec significantly enhanced the model's ability to detect implicit sentiments. Deep learning models, particularly LSTM and CNN, outperformed traditional classifiers in both sentiment polarity classification and implicit aspect association. Clustering results confirmed strong alignment with expert-labeled data, validating the framework’s effectiveness. The proposed framework demonstrates substantial improvements in classifying implicit aspects, offering valuable insights for service quality evaluation in hospitality and other domains relying on user-generated content.

Aspect-Based Sentiment Analysis (ABSA) is a Natural Language Processing (NLP) task that extracts referred aspects from text and assigns polarities to opinions about those aspects. Most research on ABSA focuses on English. Only a few ABSA works deal with the Portuguese language. In this work, we used BERTimbau to create a Question-Answer approach to ABSA in Portuguese. First, we post-trained this model with text from the same domain as our target corpus. Then, we constructed an auxiliary sentence from the aspect and converted ABSA to a sentence-pair classification task, such as question answering (QA) and natural language inference (NLI). Our experiments show that ABSA based on BERT for Portuguese achieved Balanced Accuracy (BACC) of 77% on a corpus of reviews about the accommodation sector using a post-trained model with QA approach.

TripAdvisor is one of the most popular e-commerce platforms in the tourism sector in Indonesia. TripAdvisor give Traveler Choice Award every year in Indonesia through user reviews. However, online text-based reviews are often associated only with evaluation scores that do not pay attention to the context and meaningful content of the review itself, either explicitly or implicitly. Moreover, the sentence structure of the review can have an impact on the goal of the target sentiment which is nothing but an aspect of the review itself. This research discusses aspect based sentiment analysis for explicit and implicit aspects. This research starting with taking the TripAdvisor website restaurant product review dataset to measure customer satisfaction based on four aspect categories of Ambience, Food, Service, and Price. Furthermore, the aspect word extraction and opinion word extraction processes in the case of explicit sentences for simple, compound, complex, and compound-complex sentence structures are carried out using grammatical rule extraction. This research also works on implicit sentence cases for simple sentence structures. Aspect categorization process uses hybrid approach. Aspect and opinion keyword extraction process uses the ELMo-Wikipedia. Then WordNet and TF-ICF are used to expand the meaning of aspect and opinion that has been taken. The last stage is the aspect based sentiment analysis process, both explicit and implicit sentences using SentiCircle. This research can produce two evaluations of sentiment classification, namely positive and negative. The results of the aspect extraction obtained an evaluation of the aspect categorization for each precision 0.82, recall 0.87, and f1measure 0.86. Meanwhile, the results of the sentiment analysis showed that the respective evaluations for precision 0.87, recall 0.92, and f1-measure 0.89.

Inspired by the tremendous success of the transformer-based model in natural language processing (NLP), many efforts introduce the transformer-based model into the image processing tasks. However, naive transformer models have to down-sample the image resolution to satisfy computational restrictions, thus discarding the local information, which is catastrophic for high-performance remote sensing image segmentation. Hence, this paper proposes a novel trainable boundary-aware bias loss function to enhance transformer-based models of extracting local information. On the Challenging ISPRS Potsdam dataset, two representative transformer-based models achieve remarkable performance improvements, proving the effectiveness of the proposed method.

Aspect-Based Sentiment Analysis (ABSA) is essential for extracting detailed sentiment polarity regarding specific aspects in product reviews, providing deeper insights into customer opinions on various product attributes. Unlike document-level sentiment analysis, ABSA allows a more granular understanding, crucial for e-commerce analytics and decision-making systems. This study investigates the effectiveness of transformer-based models, such as BERT and RoBERTa, in performing ABSA for product review mining.Purpose:This research aims to explore the application of transformer-based models for aspect-based sentiment analysis, comparing their performance with traditional deep learning models (LSTM and CNN) in the context of mining product reviews. The study evaluates how transformer-based models can more effectively capture sentiment polarity at the aspect level.Methods/Study design/approach:The study uses publicly available product review datasets from large-scale e-commerce platforms, where each review is annotated with aspect terms and sentiment polarities (positive, negative, neutral). The datasets were split into training, validation, and test sets in an 80:10:10 ratio. The models—BERT, RoBERTa, LSTM, and CNN—were fine-tuned on the ABSA task. Performance was evaluated using standard metrics: accuracy, precision, recall, and F1-score.Result/Findings:The results show that transformer-based models, especially RoBERTa, significantly outperform conventional deep learning baselines like LSTM and CNN. RoBERTa achieved the best performance with an accuracy of 0.93 and an F1-score of 0.92, while BERT achieved an accuracy of 0.91 and an F1-score of 0.90. In contrast, LSTM and CNN achieved F1-scores of 0.82 and 0.84, respectively. The transformer models excel in capturing contextual dependencies and associating sentiment polarity with the correct aspects, particularly in complex and multi-aspect sentences. This highlights the superior ability of transformers to handle long-range dependencies and complex sentence structures compared to LSTM and CNN.The findings confirm that transformer-based models are highly effective for aspect-based sentiment analysis, providing a more reliable approach for product review mining. Future research should address the efficiency and interpretability of these models, particularly for large-scale deployment in real-world e-commerce applications.

As a fine-grained sentiment analysis subtask in the natural language processing (NLP) community, aspect-based sentiment analysis (ABSA) aims to predict the sentiment polarity of the aspect terms according to the given sentence. In previous research, external affective knowledge has been utilized to enhance the representation of the sentiment information contained in the text, which has achieved impressive progress in ABSA task. However, it is worth noting that the external knowledge employed is usually leveraged from the single-dimensional perspective, which lacks diversified and multiple-dimensional exploration and leads to insufficient comprehension of the complex sentiment information. Thus, this paper proposes a novel ABSA network to fully exploit the sentiment features from triple dimensions, namely the psychology knowledge assisted graph attention networks (VADGAT). Preliminarily, to obtain the task-relevant semantic representation, an aspect-oriented template is devised firstly to guide the fine-tuning of the pre-trained language model. Besides, to fully leverage the triple-dimensional sentiment, they are injected into the normal dependency graph to construct three mutually independent adjacent matrices, which are implemented by graph attention networks (GAT) to extract the relations among graph nodes, progressively. Meanwhile, to be treated as the fourth dimension of human beings’ sentiment, SenticNet is also leveraged to enhance the sentiment intensity of the words in the text, which are expected to highlight implicit information and support the final sentiment inference. Moreover, an intentional shadow network is devised and leveraged to reinforce the screening ability of sentiment information, which is integrated with the mainstream representation proportionally. Furthermore, to improve the effectiveness of this innovative strategy, this paper utilizes the Jensen–Shannon divergence to evaluate the similarity between the mainstream and the shadow modules and optimizes the designed VADGAT eventually. To validate the effectiveness of the proposed approach, this paper conducted extensive experiments on five publicly available datasets, and the results show that the approach outperforms the state-of-the-art methods.

In the era of e-commerce, tremendous product reviews can provide quick and valuable insight into the market trends. Aspect-based sentiment analysis (ABSA) on product reviews is becoming increasingly important for both companies and consumers. Implicit aspect extraction, a subtask of ABSA, aims to identify the aspect of a review sentence without explicit aspect terms. For Chinese online reviews, this paper proposes a fine-tuning BERT based method to extract implicit aspects. Fine-tuning BERT performs well on a variety of natural language processing tasks. Using labeled data, the hidden layer parameters of pre-training BERT are fine-tuned in a supervised manner to generate fine-tuning BERT. Through experiments on Chinese online reviews of clothing, it is believed that fine-tuning BERT can improve the effect of implicit aspects identification.

A Pre-Trained Model for Aspect-based Sentiment Analysis Task: using Online Social Networking

The 36.98% growth in the quantity of electronic word of mouth (e-WOM) over the past five years presents opportunities for the tourism industry to understand tourists' needs and desires better when analyzed effectively. Aspect-based sentiment analysis (ABSA) is proposed as a solution, as it can identify the sentiment at a more detailed aspect level. Prior research revealed two issues in ABSA: imbalanced datasets and poor performance in representing implicit aspects and opinions. The authors proposed a combination of the bidirectional and auto-regressive transformer (BART) and bidirectional encoder representations from transformers (BERT) models. Leveraging BART capability in modeling context and BERT expertise in modeling text semantics and nuances, the author proposed an ABSA model that combines BART and BERT using the ensemble method. The experimental results reveal that combining these models significantly enhances the performance of the ABSA model, with an F1-score reaching 70%. Furthermore, text augmentation and preprocessing did not bring improvements in ABSA performance.

The recent advances of Natural Language Processing (NLP) have significantly benefited people across the globe who speak and write some specific languages, commonly termed high-resource languages (HRLs). These are languages with abundant digital resources. One of the most significant areas where NLP has recently made its mark is environment preservation. However, a significant digital divide is observed for low-resource languages (LRLs), which are not rich in terms of digital resources. Even though many LRLs, like Assamese, are very important for the cultural and linguistic identity of many indigenous communities, they remain digitally underrepresented due to a lack of annotated corpora and computational tools. This gap is even more crucial in the environmental domain, where knowledge is very limited even for high-resource languages. Multilingual information is very important for NLP applications related to climate change, disaster management and other issues related to environmental science. This paper reports a study of the challenges and strategies to address the gap in NLP, both in the context of the global and Indian landscape. The paper also highlights the key problems while addressing issues related to environmental science due to the lack of digital resources. Some of the significant factors include the absence of large corpora, labelled datasets and socio-economic factors. The paper proposes to emphasize data collection and digitization, among many other measures, to address this gap. The integration of multilingual pretrained models and transfer learning approaches also provides a pathway for enhancing performance with limited resources. The paper also presents a detailed analysis of resources available for global and Indian languages in addition to proposing a set of strategic actions, including government policies, and others, for inclusive development in NLP and sustainable environment. Bridging the divide and enabling linguistic equity will ensure participation of all sections of the society in the advancement of inclusive technological growth and solutions to the emerging problems.