Multimodal sentiment analysis has attracted increasing attention and lots of models have been proposed. However, the performance of the state-of-the-art models decreases sharply when they are deployed in the real world. We find that the main reason is that real-world applications can only access the text outputs by the automatic speech recognition (ASR) models, which may be with errors because of the limitation of model capacity. Through further analysis of the ASR outputs, we find that in some cases the sentiment words, the key sentiment elements in the textual modality, are recognized as other words, which makes the sentiment of the text change and hurts the performance of multimodal sentiment models directly. To address this problem, we propose the sentiment word aware multimodal refinement model (SWRM), which can dynamically refine the erroneous sentiment words by leveraging multimodal sentiment clues. Specifically, we first use the sentiment word position detection module to obtain the most possible position of the sentiment word in the text and then utilize the multimodal sentiment word refinement module to dynamically refine the sentiment word embeddings. The refined embeddings are taken as the textual inputs of the multimodal feature fusion module to predict the sentiment labels. We conduct extensive experiments on the real-world datasets including MOSI-Speechbrain, MOSI-IBM, and MOSI-iFlytek and the results demonstrate the effectiveness of our model, which surpasses the current state-of-the-art models on three datasets. Furthermore, our approach can be adapted for other multimodal feature fusion models easily. Data and code are available at https://github.com/albertwy/SWRM.
Aspect-based sentiment analysis (ABSA) aims at analyzing the sentiment of a given aspect in a sentence. Recently, neural network-based methods have achieved promising results in existing ABSA datasets. However, these datasets tend to degenerate to sentence-level sentiment analysis because most sentences contain only one aspect or multiple aspects with the same sentiment polarity. To facilitate the research of ABSA, NLPCC 2020 Shared Task 2 releases a new large-scale Multi-Aspect Multi-Sentiment (MAMS) dataset. In the MAMS dataset, each sentence contains at least two different aspects with different sentiment polarities, which makes ABSA more complex and challenging. To address the challenging dataset, we re-formalize ABSA as a problem of multi-aspect sentiment analysis, and propose a novel Transformer-based Multi-aspect Modeling scheme (TMM), which can capture potential relations between multiple aspects and simultaneously detect the sentiment of all aspects in a sentence. Experiment results on the MAMS dataset show that our method achieves noticeable improvements compared with strong baselines such as BERT and RoBERTa, and finally ranks the 2nd in NLPCC 2020 Shared Task 2 Evaluation.
Text mining can be applied to many fields. One of the application is using text mining in digital newspaper to do politic sentiment analysis. In this paper sentiment analysis is applied to get information from digital news articles about its positive or negative sentiment regarding particular politician. This paper suggests a simple model to analyze digital newspaper sentiment polarity using naive Bayes classifier method. The model uses a set of initial data to begin with which will be updated when new information appears. The model showed promising result when tested and can be implemented to some other sentiment analysis problems.
With faster connection speed, Internet users are now making social network a huge reservoir of texts, images and video clips (GIF). Sentiment analysis for such online platform can be used to predict political elections, evaluates economic indicators and so on. However, GIF sentiment analysis is quite challenging, not only because it hinges on spatio-temporal visual contentabstraction, but also for the relationship between such abstraction and final sentiment remains unknown.In this paper, we dedicated to find out such relationship.We proposed a SentiPairSequence basedspatiotemporal visual sentiment ontology, which forms the midlevel representations for GIFsentiment. The establishment process of SentiPair contains two steps. First, we construct the Synset Forest to define the semantic tree structure of visual sentiment label elements. Then, through theSynset Forest, we organically select and combine sentiment label elements to form a mid-level visual sentiment representation. Our experiments indicate that SentiPair outperforms other competing mid-level attributes. Using SentiPair, our analysis frameworkcan achieve satisfying prediction accuracy (72.6%). We also opened ourdataset (GSO-2015) to the research community. GSO-2015 contains more than 6,000 manually annotated GIFs out of more than 40,000 candidates. Each is labeled with both sentiment and SentiPair Sequence.
Sentiment analysis is a domain of study that focuses on identifying and classifying the ideas expressed in the form of text into positive, negative and neutral polarities. Feature selection is a crucial process in machine learning. In this paper, we aim to study the performance of different feature selection techniques for sentiment analysis. Term Frequency Inverse Document Frequency (TF-IDF) is used as the feature extraction technique for creating feature vocabulary. Various Feature Selection (FS) techniques are experimented to select the best set of features from feature vocabulary. The selected features are trained using different machine learning classifiers Logistic Regression (LR), Support Vector Machines (SVM), Decision Tree (DT) and Naive Bayes (NB). Ensemble techniques Bagging and Random Subspace are applied on classifiers to enhance the performance on sentiment analysis. We show that, when the best FS techniques are trained using ensemble methods achieve remarkable results on sentiment analysis. We also compare the performance of FS methods trained using Bagging, Random Subspace with varied neural network architectures. We show that FS techniques trained using ensemble classifiers outperform neural networks requiring significantly less training time and parameters thereby eliminating the need for extensive hyper-parameter tuning.
Due to the lack of large-scale datasets, the prevailing approach in visual sentiment analysis is to leverage models trained for object classification in large datasets like ImageNet. However, objects are sentiment neutral which hinders the expected gain of transfer learning for such tasks. In this work, we propose to overcome this problem by learning a novel sentiment-aligned image embedding that is better suited for subsequent visual sentiment analysis. Our embedding leverages the intricate relation between emojis and images in large-scale and readily available data from social media. Emojis are language-agnostic, consistent, and carry a clear sentiment signal which make them an excellent proxy to learn a sentiment aligned embedding. Hence, we construct a novel dataset of $4$ million images collected from Twitter with their associated emojis. We train a deep neural model for image embedding using emoji prediction task as a proxy. Our evaluation demonstrates that the proposed embedding outperforms the popular object-based counterpart consistently across several sentiment analysis benchmarks. Furthermore, without bell and whistles, our compact, effective and simple embedding outperforms the more elaborate and customized state-of-the-art deep models on these public benchmarks. Additionally, we introduce a novel emoji representation based on their visual emotional response which support a deeper understanding of the emoji modality and their usage on social media.
Aspect-based sentiment analysis plays an essential role in natural language processing and artificial intelligence. Recently, researchers only focused on aspect detection and sentiment classification but ignoring the sub-task of detecting user opinion span, which has enormous potential in practical applications. In this paper, we present a new Vietnamese dataset (UIT-ViSD4SA) consisting of 35,396 human-annotated spans on 11,122 feedback comments for evaluating the span detection in aspect-based sentiment analysis. Besides, we also propose a novel system using Bidirectional Long Short-Term Memory (BiLSTM) with a Conditional Random Field (CRF) layer (BiLSTM-CRF) for the span detection task in Vietnamese aspect-based sentiment analysis. The best result is a 62.76% F1 score (macro) for span detection using BiLSTM-CRF with embedding fusion of syllable embedding, character embedding, and contextual embedding from XLM-RoBERTa. In future work, span detection will be extended in many NLP tasks such as constructive detection, emotion recognition, complaint analysis, and opinion mining. Our dataset is freely available at https://github.com/kimkim00/UIT-ViSD4SA for research purposes.
Sentiment analysis research has been rapidly developing in the last decade and has attracted widespread attention from academia and industry, most of which is based on text. However, the information in the real world usually comes as different modalities. In this paper, we consider the task of Multimodal Sentiment Analysis, using Audio and Text Modalities, proposed a novel fusion strategy including Multi-Feature Fusion and Multi-Modality Fusion to improve the accuracy of Audio-Text Sentiment Analysis. We call this the Deep Feature Fusion-Audio and Text Modal Fusion (DFF-ATMF) model, and the features learned from it are complementary to each other and robust. Experiments with the CMU-MOSI corpus and the recently released CMU-MOSEI corpus for Youtube video sentiment analysis show the very competitive results of our proposed model. Surprisingly, our method also achieved the state-of-the-art results in the IEMOCAP dataset, indicating that our proposed fusion strategy is also extremely generalization ability to Multimodal Emotion Recognition.
The volume of discussions concerning brands within social media provides digital marketers with great opportunities for tracking and analyzing the feelings and views of consumers toward brands, products, influencers, services, and ad campaigns in CGC. The present study aims to assess and compare the performance of firms and celebrities (i.e., influencers that with the experience of being in an ad campaign of those companies) with the automated sentiment analysis that was employed for CGC at social media while exploring the feeling of the consumers toward them to observe which influencer (of two for each company) had a closer effect with the corresponding corporation on consumer minds. For this purpose, several consumer tweets from the pages of brands and influencers were utilized to make a comparison of machine learning and lexicon-based approaches to the sentiment analysis through the Naive algorithm (lexicon-based) and Naive Bayes algorithm (machine learning method) and obtain the desired results to assess the campaigns. The findings suggested that the approaches were dissimilar in terms of accuracy; the machine learning method yielded higher accuracy. Finally, the results showed which influencer was more appropriate according to their existence in previous campaigns and helped choose the right influencer in the future for our company and have a better, more appropriate, and more efficient ad campaign subsequently. It is required to conduct further studies on the accuracy improvement of the sentiment classification. This approach should be employed for other social media CGC types. The results revealed decision-making for which sentiment analysis methods are the best approaches for the analysis of social media. It was also found that companies should be aware of their consumers' sentiments and choose the right person every time they think of a campaign.
In this paper, we investigate the impact of the social media data in predicting the Tehran Stock Exchange (TSE) variables for the first time. We consider the closing price and daily return of three different stocks for this investigation. We collected our social media data from Sahamyab.com/stocktwits for about three months. To extract information from online comments, we propose a hybrid sentiment analysis approach that combines lexicon-based and learning-based methods. Since lexicons that are available for the Persian language are not practical for sentiment analysis in the stock market domain, we built a particular sentiment lexicon for this domain. After designing and calculating daily sentiment indices using the sentiment of the comments, we examine their impact on the baseline models that only use historical market data and propose new predictor models using multi regression analysis. In addition to the sentiments, we also examine the comments volume and the users' reliabilities. We conclude that the predictability of various stocks in TSE is different depending on their attributes. Moreover, we indicate that for predicting the closing price only comments volume and for predicting the daily return both the volume and the sentiment of the comments could be useful. We demonstrate that Users' Trust coefficients have different behaviors toward the three stocks.
Add to Chrome
Add to Firefox