Multilingual emotion classifier using unsupervised pattern extraction from microblog data

Linguistic Patterns for Code Word Resilient Hate Speech Identification

The permanent transition to online activity has brought with it a surge in hate speech discourse. This has prompted increased calls for automatic detection methods, most of which currently rely on a dictionary of hate speech words, and supervised classification. This approach often falls short when dealing with newer words and phrases produced by online extremist communities. These code words are used with the aim of evading automatic detection by systems. Code words are frequently used and have benign meanings in regular discourse, for instance, "skypes, googles, bing, yahoos" are all examples of words that have a hidden hate speech meaning. Such overlap presents a challenge to the traditional keyword approach of collecting data that is specific to hate speech. In this work, we first introduced a word embedding model that learns the hidden hate speech meaning of words. With this insight on code words, we developed a classifier that leverages linguistic patterns to reduce the impact of individual words. The proposed method was evaluated across three different datasets to test its generalizability. The empirical results show that the linguistic patterns approach outperforms the baselines and enables further analysis on hate speech expressions.

Multimodal emotion recognition with hierarchical memory networks

Emotion recognition in conversations is crucial as there is an urgent need to improve the overall experience of human-computer interactions. A promising improvement in this field is to develop a model that can effectively extract adequate contexts of a test utterance. We introduce a novel model, termed hierarchical memory networks (HMN), to address the issues of recognizing utterance level emotions. HMN divides the contexts into different aspects and employs different step lengths to represent the weights of these aspects. To model the self dependencies, HMN takes independent local memory networks to model these aspects. Further, to capture the interpersonal dependencies, HMN employs global memory networks to integrate the local outputs into global storages. Such storages can generate contextual summaries and help to find the emotional dependent utterance that is most relevant to the test utterance. With an attention-based multi-hops scheme, these storages are then merged with the test utterance using an addition operation in the iterations. Experiments on the IEMOCAP dataset show our model outperforms the compared methods with accuracy improvement.

Multilingual Sentiment Analysis: A Systematic Literature Review

With the explosive growth of social media, the online community can freely express their opinions without disclosing their identities. People with hidden agendas can easily post fake opinions to discredit target products, services, politicians, or organizations. With these big data, monitoring opinions and distilling their sentiments remain a formidable task because of the proliferation of diverse sites with a large volume of opinions that are portrayed in multilingual. Therefore, this paper aims to provide a systematic literature review on multilingual sentiment analysis, which summarises the common languages supported in multilingual sentiment analysis, pre-processing techniques, existing sentiment analysis approaches, and evaluation models that have been used for multilingual sentiment analysis. By following the systematic literature review, the findings revealed, most of the models supported two languages, and English is seen as the most used language in sentiment analysis studies. None of the reviewed literature has catered the combination of languages for English, Chinese, Malay, and Hindi language on multilingual sentiment analysis. The common pre-processing techniques for the multilingual domain are tokenization, normalization, capitalization, N-gram, and machine translation. Meanwhile, the sentiment analysis classification techniques for multilingual sentiment are hybrid sentiment analysis, which includes localized language analysis, unsupervised topic clustering, and then followed by multilingual sentiment analysis. In terms of evaluation, most of the studies used precision, recall, and accuracy as the benchmark for the results.