Framework For Sentiment Classification Research Articles

MFSC: A Multimodal Aspect-Level Sentiment Classification Framework with Multi-Image Gate and Fusion Networks

Currently, there is a great deal of interest in multimodal aspect-level sentiment classification using both textual and visual information, which changes the traditional use of only single-modal to identify sentiment polarity. Considering that existing methods could be strengthened in terms of classification accuracy, we conducted a study on aspect-level multimodal sentiment classification with the aim of exploring the interaction between textual and visual features. Specifically, we construct a multimodal aspect-level sentiment classification framework with multi-image gate and fusion networks called MFSC. MFSC consists of four parts, i.e., text feature extraction, visual feature extraction, text feature enhancement, and multi-feature fusion. Firstly, a bidirectional long short-term memory network is adopted to extract the initial text feature. Based on this, a text feature enhancement strategy is designed, which uses text memory network and adaptive weights to extract the final text features. Meanwhile, a multi-image gate method is proposed for fusing features from multiple images and filtering out irrelevant noise. Finally, a text-visual feature fusion method based on an attention mechanism is proposed to better improve the classification performance by capturing the association between text and images. Experimental results show that MFSC has advantages in classification accuracy and macro-F1.

LexiSNTAGMM: an unsupervised framework for sentiment classification in data from distinct domains, synergistically integrating dictionary-based and machine learning approaches

LexiSNTAGMM: an unsupervised framework for sentiment classification in data from distinct domains, synergistically integrating dictionary-based and machine learning approaches

Airline reviews processing: Abstractive summarization and rating-based sentiment classification using deep transfer learning

Opinion summarization and sentiment classification are key processes for understanding, analyzing, and leveraging information from customer opinions. The rapid and ceaseless increase in big data of reviews on e-commerce platforms, social media, or review portals becomes a stimulus for the automation of these processes. In recent years, deep transfer learning has opted to solve many challenging tasks in Natural Language Processing (NLP) relieving the hassles of exhaustive training and the requirement of extensive labelled datasets. In this work, we propose frameworks for Abstractive Summarization (ABS) and Sentiment Analysis (SA) of airline reviews using Pretrained Language Models (PLM). The abstractive summarization model goes through two finetuning stages, the first one, for domain adaptation and the second one, for final task learning. Several studies in the literature empirically demonstrate that review rating has a positive correlation with sentiment valence. For the sentiment classification framework, we used the rating value as a signal to determine the review sentiment, and the model is built on top of BERT (Bidirectional Encoder Representations from Transformers) architecture. We evaluated our models comprehensively with multiple metrics. Our results indicate competitive performance of the models in terms of most of the evaluation metrics.

A Nano Scale Machine Learning Assisted Natural Language Processing Framework for Sentiment Classification in Smart System Platform

A Nano Scale Machine Learning Assisted Natural Language Processing Framework for Sentiment Classification in Smart System Platform

Financial sentiment classification with fresh and hot public opinions

Financial sentiment analysis aims to extract public opinion about an institution to help financial researchers make better decisions. To predict sentiment more accurately, it is necessary for models to improve their capability to capture long-term temporal information and support multi-user interaction. However, existing methods only analyze sentiment based on one comment from a user, which fails to fully exploit the latent emotions of the public, and they lack effective temporal modeling and interaction capabilities. In this paper, we analyze a company from two perspectives to alleviate the above issues: (1) the fresh opinions can reflect timely public attitudes towards a company, while (2) the hot opinions provide the most influential views. A comprehensive exploration of fresh and hot financial sentiment can help researchers make more accurate determinations. To this end, we propose a novel financial sentiment classification framework (FSCN), that can capture temporal information and interact with the opinions of users to make a more comprehensive decision. Our approach takes into account the inherent temporal dependencies in public opinions and combines both views of information to achieve an accurate classification of financial sentiment. Specifically, the FSCN contains (1) a multi-opinion extractor to filter and extract features from massively fresh and hot opinions, respectively. (2) a fresh-hot bilinear pooling (FHBP) module to effectively fuse fresh and hot features. Additionally, to verify the effectiveness of the proposed method, we crawl data from the Internet and create a real-world public opinion dataset that consists of 79,350 comments from 837 companies. Extensive experiments demonstrate that our framework achieves state-of-the-art results on this real-world dataset and is capable of providing reliable service in the financial system. Codes will be released at https://github.com/zjfgh2015/FSCN.

Semi-Supervised Sentiment Classification and Emotion Distribution Learning Across Domains

In this study, sentiment classification and emotion distribution learning across domains are both formulated as a semi-supervised domain adaptation problem, which utilizes a small amount of labeled documents in the target domain for model training. By introducing a shared matrix that captures the stable association between document clusters and word clusters, non-negative matrix tri-factorization (NMTF) is robust to the labeled target domain data and has shown remarkable performance in cross-domain text classification. However, the existing NMTF-based models ignore the incompatible relationship of sentiment polarities and the relatedness among emotions. Besides, their applications on large-scale datasets are limited by the high computation complexity. To address these issues, we propose a semi-supervised NMTF framework for sentiment classification and emotion distribution learning across domains. Based on a many-to-many mapping between document clusters and sentiment polarities (or emotions), we first incorporate the prior information of label dependency to improve the model performance. Then, we develop a parallel algorithm based on message passing interface (MPI) to further enhance the model scalability. Extensive experiments on real-world datasets validate the effectiveness of our method.

Correction to: An Optimized Deep Neural Aspect Based Framework for Sentiment Classification

Correction to: An Optimized Deep Neural Aspect Based Framework for Sentiment Classification

Transfer Learning-Based Framework for Sentiment Classification of Cosmetics Products Reviews

The exponential growth in online reviews and recommendations availability drives sentiment classification, an interesting topic in industrial research. There is a vital requirement for organizations to explore client behaviour to assess the competitive business environment. This study aspires to examine and predict customer reviews using Transfer learning (TL) approaches. Reviews can span so many domains that it is challenging to gather annotated training data for all of them. Hence, this paper proposed an annotation algorithm to label a large unlabeled dataset. These reviews must be pulled and examined to predict the sentiment polarity, whether the review is positive, neutral, or negative. We propose a deep learning-based approach that learns to extract a meaningful representation for each review in an unsupervised fashion. Sentiment classifiers trained with this high-level feature representation outperform state-of-the-art methods on a benchmark of reviews of cosmetics brands on Amazon or other platforms. Using the BERT for sentiment analysis, we achieved the highest accuracy of 93.21% compared to previous studies.

An Optimized Deep Neural Aspect Based Framework for Sentiment Classification

An Optimized Deep Neural Aspect Based Framework for Sentiment Classification

Building an enhanced sentiment classification framework based on natural language processing

Sentiment classification is one of the major tasks of natural language processing (NLP) and has gained much attention by researchers and businesses in recent years. However, the semantics of the social networking language is becoming increasingly complex and unpredictable, affecting the accuracy of the associated NLP systems. In this paper, we propose a hybrid sentiment analysis (SA) framework that classifies the opinions of Vietnamese reviews into one of two types: positive or negative. The special feature of the proposed framework is that it is built on a combination of three different text representation models that focus on analyzing social media network language characteristics. Our system achieved an accuracy score of 81.54% on the test set, which is better than other strategies. Based on the experimental results, this work proves that the choice of text representation model determines the performance of the system.

Sentiment classification of microblog: A framework based on BERT and CNN with attention mechanism

BERT, a boon to natural language understanding, extracts the context information of words and forms the basis of the newly-designed sentiment classification framework for Chinese microblogs. Coupled with a CNN and an attention mechanism, the BERT model takes Chinese characters as inputs for vectorization and outputs two kinds of vectors: character-level vectors and sentence-level vectors. The character-level vectors are input into the CNN, and both top-K-average pooling and attention pooling are used to mine emotions from the microblogs. The attention mechanism is used to fuse various features to the microblog vectors. The final classification results are derived via a softmax function over the microblog vectors. The results show an outstanding improvement over benchmark methods on the two classification tasks.

Learning for amalgamation: A multi-source transfer learning framework for sentiment classification

Transfer learning plays an essential role in Deep Learning, which can remarkably improve the performance of the target domain, whose training data is not sufficient. Our work explores beyond the common practice of transfer learning with a single pre-trained model. We focus on the task of Vietnamese sentiment classification and propose LIFA, a framework to learn a unified embedding from several pre-trained models. We further propose two more LIFA variants that encourage the pre-trained models to either cooperate or compete with one another. Studying these variants sheds light on the success of LIFA by showing that sharing knowledge among the models is more beneficial for transfer learning. Moreover, we construct the AISIA-VN-Review-F dataset, the first large-scale Vietnamese sentiment classification database. We conduct extensive experiments on the AISIA-VN-Review-F and existing benchmarks to demonstrate the efficacy of LIFA compared to other techniques. To contribute to the Vietnamese NLP research, we publish our source code and datasets to the research community upon acceptance.

Tourism web app with Aspect Based Sentiment Classification Framework for Tourist Review

Abstract: According to studies, current tourism recommendation systems make false recommendations that do not live up to tourist expectations. Among The majority of these systems are inefficient, which is one of the main causes of the problem. A recommendation system that incorporates user feedback element.Tourist reviews are sources of information for travellers interested in learning more about tourist destinations. Regrettably, some reviews are irrelevant, resulting in noisy statistics. Sentiment categorization algorithms based on aspects have showed potential in reducing noise. We proposed a framework for sentiment classification based on aspects that can not only detect aspects quickly but also execute classification tasks with high accuracy. The framework has been deployed to assists travellers in finding the best restaurant or lodging in a city, and its performance has been evaluated with outstanding results using real-world datasets. Keywords: Pre-processing, Classifier algorithm, Feature extraction NLP, Tourism Strategy,Machine Learning, Tourist Reviews, Aspect Based Sentiment Analysis etc.

Chinese sentiment classification based on Word2vec and vector arithmetic in human–robot conversation

With the aim of enhancing human–chatbot interactions, this paper presents a novel sentiment classification framework that recognizes both semantics emotion terms and emojis. Underpinned by Word2vec and vector arithmetic, a deep model learns the vectors of emotion words and emojis from a large-scale corpus of conversations. The vectors are then assembled into a dictionary, evaluated with an improved k-means similarity calculation, and assigned to a category of emotions. To accommodate negative grammar constructions, e.g., “not happy” as opposed to “unhappy”, a separate module combines the vectors of negatives and emotion words using vector addition. The final sentiment classification results are derived by calculating the similarity between an utterance and the emotion categories. Experiments show this strategy results in multi-emotion and polarity classifications between 3 and 4% more precise than the next best performing baseline classifier.

Grammar guided embedding based Chinese long text sentiment classification

AbstractAlthough the state‐of‐the‐art sentiment classification approaches, such as LSTM and TextCNN, have achieved a good performance on Chinese short text sentiment analysis, the Chinese long text sentiment classification is still a challenge because of the sentiment change problem and the long text structure problem. Therefore, we propose a grammar guided embedding model (GGE) and a novel Chinese long text sentiment classification framework. First, the part‐of‐speech (POS) tags are introduced as the Chinese long text grammar guided information which can help classification approaches to model the Chinese long text structure and the important structure of sentiment change. Second, we proposed a simple GGE training method which considers the combination representation of word sequence and POS sequence. Finally, we proposed a unified framework which combines our novel GGE with TextCNN. Experiment results show that after using GGE, the model outperforms the state‐of‐the‐art approaches. At the same time, we also found that the GGE achieves the model converge faster, that is, it can achieve better results than without GGE when there is only a small amount of training data. Thus, we believe that the GGE can help machines better understand human language sentiment expression structure.

Kernel Optimized-Support Vector Machine and MapReduce Framework for Sentiment Classification of Train Reviews

Sentiment analysis is one of the popular techniques gaining attention in recent times. Nowadays, people gain information on reviews of users regarding public transportation, movies, hotel reservation, etc., by utilizing the resources available, as they meet their needs. Hence, sentiment classification is an essential process employed to determine the positive and negative responses. This paper presents an approach for sentiment classification of train reviews using MapReduce model with the proposed Kernel Optimized-Support Vector Machine (KO-SVM) classifier. The MapReduce framework handles big data using a mapper, which performs feature extraction and reducer that classifies the review based on KO-SVM classification. The feature extraction process utilizes features that are classification-specific and SentiWordNet-based. KO-SVM adopts SVM for the classification, where the exponential kernel is replaced by an optimized kernel, finding the weights using a novel optimizer, Self-adaptive Lion Algorithm (SLA). In a comparative analysis, the performance of KO-SVM classifier is compared with SentiWordNet, NB, NN, and LSVM, using the evaluation metrics, specificity, sensitivity, and accuracy, with train review and movie review database. The proposed KO-SVM classifier could attain maximum sensitivity of 93.46% and 91.249% specificity of 74.485% and 70.018%; and accuracy of 84.341% and 79.611% respectively, for train review and movie review databases.

Tourism Mobile App With Aspect-Based Sentiment Classification Framework for Tourist Reviews

Tourist reviews are information sources for travelers to know about tourist places. Unfortunately, some reviews are irrelevant and become noisy data. Aspect-based sentiment classification methods have shown promise in suppressing the noise. However, not much research has been done on automatic aspect identification, and identification of implicit, infrequent and co-referential aspects, resulting in misclassifications. This paper presents a framework of aspect-based sentiment classification that will not only identify the aspects very efficiently but can perform classification task with high accuracy. The framework has been implemented as a mobile app that helps tourists find the best restaurant or hotel in a city, and performance has been evaluated by conducting experiments on real-world datasets with excellent results (85% identification and 90% classification).

Modified framework for sarcasm detection and classification in sentiment analysis

<span>Sentiment analysis is directed at identifying people's opinions, beliefs, views and emotions in the context of the entities and attributes that appear in text. The presence of sarcasm, however, can significantly hamper sentiment analysis. In this paper a sentiment classification framework is presented that incorporates sarcasm detection. The framework was evaluated using a non-linear Support Vector Machine and Malay social media data. The results obtained demonstrated that the proposed sarcasm detection process could successfully detect the presence of sarcasm in that better sentiment classification performance was recorded. A best average F-measure score of 0.905 was recorded using the framework; a significantly better result than when sentiment classification was performed without sarcasm detection.</span>

An Effective Framework for Tweet Level Sentiment Classification using Recursive Text Pre-Processing Approach

With around 330 million people around the globe tweet 6000 times per second to express their feelings about a product, policy, service, or an event. Twitter message majorly consists of thoughts. Thoughts are mostly expressed as a text and it is an open challenge to extract some insight from free text. The scope of this work is to build an effective tweet level sentiment classification framework that may use these thoughts to know collective sentiment of the folk on a particular subject. Furthermore, this work also analyses the impact of proposed tweet level recursive text pre-processing approach on overall classification results. This work achieved up to 4 points accuracy improvement over baseline approach besides mitigating feature vector space.

Kernel Optimized-Support Vector Machine and Mapreduce framework for sentiment classification of train reviews

Sentiment analysis is one of the popular techniques gaining attention in recent times. Nowadays, people gain information on reviews of users regarding public transportation, movies, hotel reservation, etc., by utilizing the resources available, as they meet their needs. Hence, sentiment classification is an essential process employed to determine the positive and negative responses. This paper presents an approach for sentiment classification of train reviews using MapReduce model with the proposed Kernel Optimized-Support Vector Machine (KO-SVM) classifier. The MapReduce framework handles big data using a mapper, which performs feature extraction and reducer that classifies the review based on KO-SVM classification. The feature extraction process utilizes features that are classification-specific and SentiWordNet-based. KO-SVM adopts SVM for the classification, where the exponential kernel is replaced by an optimized kernel, finding the weights using a novel optimizer, self-adaptive lion algorithm. In a comparative analysis, the performance of KO-SVM classifier is compared with SentiWordNet, Naive Bayes, neural network, and LSVM, using the evaluation metrics, specificity, sensitivity, and accuracy, with train review and movie review database. The proposed KO-SVM classifier could attain maximum sensitivity of 93.46% and 91.249%, specificity of 74.485% and 70.018%; and accuracy of 84.341% and 79.611%, respectively, for train review and movie review databases.