A Joint Model for Sentiment Classification and Opinion Words Extraction

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.

Mining opinions from online reviews is a fundamental step in obtaining the overall sentiment of a product. Detection of opinion relations among the words play an important role in the opinion target (OT) and opinion word (OW) extraction. In this paper, Partially Supervised Word Alignment Model is used to find opinion relations among words. Graph based co-ranking algorithm is used in estimating the confidence of each OT and OW. Candidates having confidence value higher than the threshold are extracted as final OT and OW. We propose a hybrid method that considers semantic relations along with opinion relations that results in fine grained opinion target (OT) and opinion word (OW) extraction. This semantic relations and opinion relations affects the confidence calculation of the OT and OW and improves the precision of extraction.

With proliferation of user-generated reviews, new opportunities and challenges arise. The advance of Web technologies allows people to access a large amount of reviews of products and services online. Knowing what others like and dislike becomes increasingly important for their decision making in online shopping. The retailers also care more than ever about online reviews, because a vast pool of reviews enables them to monitor reputations and collect feedbacks efficiently. However, people often find difficult times in identifying and summarizing fine-grained sentiments buried in the opinion-rich resources. The traditional sentiment analysis, which focuses on the overall sentiments, fails to uncover the sentiments with regard to the aspects of the reviewed entities. This dissertation studied the research problem of Aspect Based Sentiment Analysis (ABSA), which is to reveal the aspect-dependent sentiment information of review text. ABSA consists of several subtasks: 1) aspect extraction, 2) aspect term extraction, 3) aspect category classification, and 4) sentiment polarity classification at aspect level. We focused on the approach of topic models and neural networks for ABSA. First, to extract the aspects from a collection of reviews and to detect the sentiment polarity regarding the aspects in each review, we proposed a few probabilistic graphical models, which can model words distribution in reviews and aspect ratings at the same time. Second, we presented a multi-task learning model based on long-short term memory and convolutional neural network for aspect category classification and aspect term extraction. Third, for aspect-level sentiment polarity classification, we developed a gated convolution neural network, which can be applied to aspect category sentiment analysis as well as aspect target sentiment analysis.

In opinion mining, the precision of extracting opinion words and opinion targets from online review dataset can severely influence the precision of opinion mining systems.In this paper, we propose a graph-based filtering approach to extract effective opinion targets and opinion words from the candidate words, which can be mined by word-based translation model in a monolingual scenario.Experimental results on four real Chinese review datasets show that our approach is promising performance.

Weibo messages sentiment polarity classification towards given topics refers to that the machine automatically classifies whether the weibo message is of positive, negative, or neutral sentiment towards the given topic.The algorithm the sentiment analysis system CUCsas adopts to perform this task includes three steps: (1) whether there is an "exp" (short for "expression having evaluation meaning") in the weibo message; (2) whether there is a semantic orientation relationship between the exp and topic; (3) the sentiment polarity classification of the exp.CUCsas completes step (1) based on the sentiment lexicon and sentiment value assignment rules, completes step (2) based on the topic extraction and sentiment polarity classification rule base, and completes step (3) based on the sentiment computing rules.Taking 20 given topics and a total of 19,469 weibo messages released by SIGHAN-2015 Bake-off as the test data, the overall F value of the rule-based system CUCsas is 0.69 in the unrestricted test.

Aspect-based sentiment classification (ABSC) (also called fine-grained sentiment analysis) is an essential and challenging task among sentiment analysis tasks. Aspect level sentiment analysis overcomes the limitation of the document and sentence level when multiple aspects appear in a review. Recently, neural network approaches like LSTM has achieved better results in ABSC than traditional machine learning algorithms. Aspect extraction and polarity detection for specific aspects in the review becoming an important task in aspect level sentiment analysis. Therefore, the paper aims to predict the sentiment polarity of each aspect in the review into 3 classes by developing the multi-aspect attention (MAA) model and combine it with the BiLSTM neural network, called MAA-BLSTM. Unlike unidirectional LSTM, the BiLSTM network runs the input in two directions: forward and backward, therefore, it can understand the context better than LSTM. In this paper, the hyperparameter tuning approach is also considered to be the high-performing model. Experiments are tested on restaurant and laptop data from SemEval (2014, 2015, and 2016) datasets and compare the result with other LSTM-based methods. Finally, the result proves that the accuracy of the proposed sentiment model reaches 89.9 on the restaurant dataset and 82.1 on the laptop dataset which are higher than other LSTM approaches.

Aspect term extraction and opinion word extraction are two fundamental subtasks of aspect-based sentiment analysis. The internal relationship between aspect terms and opinion words is typically ignored, and information for the decision-making of buyers and sellers is insufficient. In this paper, we explore an aspect–opinion pair extraction (AOPE) task and propose a Question-Driven Span Labeling (QDSL) model to extract all the aspect–opinion pairs from user-generated reviews. Specifically, we divide the AOPE task into aspect term extraction (ATE) and aspect-specified opinion extraction (ASOE) subtasks; we first extract all the candidate aspect terms and then the corresponding opinion words given the aspect term. Unlike existing approaches that use the BIO-based tagging scheme for extraction, the QDSL model adopts a span-based tagging scheme and builds a question–answer-based machine-reading comprehension task for an effective aspect–opinion pair extraction. Extensive experiments conducted on three tasks (ATE, ASOE, and AOPE) on four benchmark datasets demonstrate that the proposed method significantly outperforms state-of-the-art approaches.

Sentiment analysis or opinion mining used to capture the community's attitude who have experienced the specific service/product. Sentiment analysis usually concentrates to classify the opinion of whole document or sentence. However, in most comments, users often express their opinions on different aspects of the mentioned entity rather than express general sentiments on entire document. In this case, using aspect-based sentiment analysis (ABSA) is a solution. ABSA emphases on extracting and synthesizing sentiments on particular aspects of entities in opinion text. The previous studies have difficulty working with aspect extraction and sentiment polarity classification in multiple domains of review. We offer an innovative deep learning approach with the integrated construction of bidirectional Long Short Term Memory (BiLSTM) and Convolutional Neural Network (CNN) for multidomain ABSA in this article. Our system finished the following tasks: domain classification, aspect extraction and opinion determination of aspect in the document. Besides applying GloVe word embedding for input sentences from mixed Laptop_Restaurant domain of the SemEval 2016 dataset, we also use the additional layer of POS to pick out the word morphological attributes before feeding to the CNN_BiLSTM architecture to enhance the flexibility and precision of our suggested model. Through experiment, we found that our proposed model has performed the above mentioned tasks of domain classification, aspect and sentiment extraction concurrently on a mixed domain dataset and achieved the positive results compared to previous models that were performed only on separated domain dataset.

With the continuous development of online shopping, more and more online shopping review data has been accumulated. These reviews contain consumers' attitudes and opinions on the products, which are of great value for mining. This research introduces a set of methods to mine review information by using aspect extraction and sentiment analysis. By analyzing the aspects involved in the reviews and the sentiment polarity of the reviewers, the potential opinions on the products can be summarized and presented to the merchants so as to understand consumers' preferences and make reasonable improvements to the products. This research combines and adopts several techniques, such as aspect extraction, paragraph vector and sentiment classification, to build a complete process which can handle review text and get the results of review aspects and sentiment polarity. By taking an Amazon review dataset as an example and applying this process, the relevant aspects and sentiment polarities contained in the reviews are obtained, which reveals that the idea of this research is feasible.

Aspect-level sentiment analysis gives a detailed view of user opinions expressed towards each feature of a product. Aspect extraction is a challenging task in aspect-level sentiment analysis. Hence, several researchers worked on the problem of aspect extraction during the past decade. The authors begin this chapter with a brief introduction to aspect-level sentimental analysis, which covers the definition of key terms used in this chapter, and the authors also illustrate various subtasks of aspect-level sentiment analysis. The introductory section is followed by an explanation of the various feature learning methods like supervised, unsupervised, semi-supervised, etc. with a discussion regarding their merits and demerits. The authors compare the aspect extraction methods performance with respect to metrics and a detailed discussion on the merits and demerits of the approaches. They conclude the chapter with pointers to the unexplored problems in aspect-level sentiment analysis that may be beneficial to the researchers who wish to pursue work in this challenging and mature domain.

Sentiment analysis, a critical task in natural language processing, aims to automatically identify and classify the sentiment expressed in textual data. Aspect-level sentiment analysis focuses on determining sentiment at a more granular level, targeting specific aspects or features within a piece of text. In this paper, we explore various techniques for sentiment analysis, including traditional machine learning approaches and state-of-the-art deep learning models. Additionally, deep learning techniques has been utilized to identifying and extracting specific aspects from text, addressing aspect-level ambiguity, and capturing nuanced sentiments for each aspect. These datasets are valuable for conducting aspect-level sentiment analysis. In this article, we explore a language model based on pre-trained deep neural networks. This model can analyze sequences of text to classify sentiments as positive, negative, or neutral without explicit human labeling. To evaluate these models, data from Twitter's US airlines sentiment database was utilized. Experiments on this dataset reveal that the BERT, RoBERTA and DistilBERT model outperforms than the ML based model in accuracy and is more efficient in terms of training time. Notably, our findings showcase significant advancements over previous state-of-the-art methods that rely on supervised feature learning, bridging existing gaps in sentiment analysis methodologies. Our findings shed light on the advancements and challenges in sentiment analysis, offering insights for future research directions and practical applications in areas such as customer feedback analysis, social media monitoring, and opinion mining.

Aspect level sentiment analysis is a basic task to determine the sentiment bias based on the contextual information near the aspect words. Some sentences contain many confusing feature words due to incomplete structure or high complexity in sentiment prediction, which can easily lead to the problem that the model pays too much attention to unimportant features. Furthermore, the role of aspect-related sentiment features is not significant in the feature extraction process. The capsule network, due to its complex network structure, leads to capsule detachment when too much information is found in the dataset. Too much computational resources are easily consumed during dynamic routing, which in turn affects the model's judgement of the polarity of aspect-related sentiment. To address these issues, we propose a capsule network (AGCDFF-Caps) with aspect-gated convolution and Dual-Feature Filtering layers for aspect-level sentiment analysis. A pre-trained BERT model is used to generate a serialised representation of the text, and the semantic representation in the contextual text is enhanced by the self-attention mechanism and bi-GRU. An aspect-gating based convolutional neural network is constructed to selectively extract contextual sentiment information about aspects and discard irrelevant information. A capsule network is then used to learn the multispatial semantic features of the aspect and the text. In particular, we incorporate a Dual-Feature Filtering network structure into the capsule network structure to strengthen the interaction between the particular aspect and the context from global and local perspectives, filtering the redundant information in local semantics and global semantics. The opinion feature representations that can more accurately express the emotional tendency of the aspect are obtained. Experimental results on SemEval2014 and Twitter datasets show that the proposed hybrid network structure has superior classification performance compared to 12 advanced baseline methods.

With the recent emergence of Web-based applications and use of social networking sites, number of people are eager in expressing their views and opinions online. The sentimental analysis also referred to as opinion mining aims at processing user reviews (about products, movies, services, books, places, etc.). These reviews are often unstructured and need processing to evolve into the productive knowledge. Majority of the sentiment analysis works on the classification of opinion polarity with the use of simple classifiers. Handling diverse data distribution is one of the major issues that simple classifiers suffer. To cope up with the issue in this paper, we utilized the ensemble learners on the polarity prediction of the movie reviews. The proposed work processes the review data through some elementary steps that are conducted for the feature extraction in sentiment analysis. In addition to the feature extraction, we further perform the feature selection for the sake of dimensionality reduction. However, in contrast to the conventional simple learner, we applied the ensemble learner in the proposed model and evaluated its performance. To compare the ensemble model competence, we conducted the experiment on both individual as well as ensemble learner (random forest, AdaBoost, extra trees) and computed classification measures on both the model. IMDB dataset is used, and the polarity of a review, i.e., whether it is positive or negative, is predicted. With an extensive experimentation, it is found that results of ensemble classifiers are outperforming than individual learner in the classification of sentiment polarity.

LLarge language models (LLMs) demonstrate strong natural language processing capabilities but come with significant computational costs, raising questions about their practical utility compared to small language models (SLMs). This study systematically compares SLMs (DistilBERT, ELECTRA) and LLMs (Flan-T5, Flan-UL2) on two customer review analysis tasks: sentiment polarity classification and product correlation analysis. Our results show that while LLMs outperform in sentiment classification, they do so at a much higher computational cost, whereas fine-tuned SLMs excel in domain-specific correlation analysis with greater efficiency. To balance accuracy and cost, we propose a context-enhanced hybrid (CE-Hybrid) model, which refines traditional hybrid methods by enriching LLM input with SLM-generated insights, reducing redundant computation while maintaining accuracy. Our findings quantify the trade-offs between model performance and resource efficiency, offering actionable insights for businesses to optimize AI deployment. These results have significant implications for real-world applications such as e-commerce, customer service automation, and business analytics.

Classification model performance can be degraded by label noise in the training set. The sentiment classification domain also struggles with this issue, whereby customer reviews can be mislabeled. Some customers give a rating score for a product or service that is inconsistent with the review content. If business owners are only interested in the overall rating picture that includes mislabeling, this can lead to erroneous business decisions. Therefore, this issue became the main challenge of this study. If we assume that customer reviews with noisy labels in the training data are validated and corrected before the learning process, then the training set can generate a predictive model that returns a better result for the sentiment analysis or classification process. Therefore, we proposed a mechanism, called polarity label analyzer, to improve the quality of a training set with noisy labels before the learning process. The proposed polarity label analyzer was used to assign the polarity class of each sentence in a customer review, and then polarity class of that customer review was concluded by voting. In our experiment, datasets were downloaded from TripAdvisor and two linguistic experts helped to assign the correct labels of customer reviews as the ground truth. Sentiment classifiers were developed using the k-NN, Logistic Regression, XGBoost, Linear SVM and CNN algorithms. After comparing the results of the sentiment classifiers without training set improvement and the results with training set improvement, our proposed method improved the average scores of F1 and accuracy by 20.59%.