A machine learning-based investigation utilizing the in-text features for the identification of dominant emotion in an email

Multi-label emotion classification of Urdu tweets

Urdu is a widely used language in South Asia and worldwide. While there are similar datasets available in English, we created the first multi-label emotion dataset consisting of 6,043 tweets and six basic emotions in the Urdu Nastalíq script. A multi-label (ML) classification approach was adopted to detect emotions from Urdu. The morphological and syntactic structure of Urdu makes it a challenging problem for multi-label emotion detection. In this paper, we build a set of baseline classifiers such as machine learning algorithms (Random forest (RF), Decision tree (J48), Sequential minimal optimization (SMO), AdaBoostM1, and Bagging), deep-learning algorithms (Convolutional Neural Networks (1D-CNN), Long short-term memory (LSTM), and LSTM with CNN features) and transformer-based baseline (BERT). We used a combination of text representations: stylometric-based features, pre-trained word embedding, word-based n-grams, and character-based n-grams. The paper highlights the annotation guidelines, dataset characteristics and insights into different methodologies used for Urdu based emotion classification. We present our best results using micro-averaged F1, macro-averaged F1, accuracy, Hamming loss (HL) and exact match (EM) for all tested methods.

A novel algorithm for sarcasm detection using supervised machine learning approach

<abstract> <p>Sarcasm means the opposite of what you desire to express, particularly to insult a person. Sarcasm detection in social networks SNs such as Twitter is a significant task as it has assisted in studying tweets using NLP. Many existing study-related methods have always focused only on the content-based on features in sarcastic words, leaving out the lexical-based features and context-based features knowledge in isolation. This shows a loss of the semantics of terms in a sarcastic expression. This study proposes an improved model to detect sarcasm from SNs. We used three feature set engineering: context-based on features set, Sarcastic based on features, and lexical based on features. Two Novel Algorithms for an effective model to detect sarcasm are divided into two stages. The first used two algorithms one with preprocessing, and the second algorithm with feature sets. To deal with data from SNs. We applied various supervised machine learning (ML) such as k-nearest neighbor classifier (KNN), na?ve Bayes (NB), support vector machine (SVM), and Random Forest (RF) classifiers with TF-IDF feature extraction representation data. To model evaluation metrics, evaluate sarcasm detection model performance in precision, accuracy, recall, and F1 score by 100%. We achieved higher results in Lexical features with KNN 89.19 % accuracy campers to other classifiers. Combining two feature sets (Sarcastic and Lexical) has shown slight improvement with the same classifier KNN; we achieved 90.00% accuracy. When combining three feature sets (Sarcastic, Lexical, and context), the accuracy is shown slight improvement. Also, the same classifier we achieved is a 90.51% KNN classifier. We perform the model differently to see the effect of three feature sets through the experiment individual, combining two feature sets and gradually combining three feature sets. When combining all features set together, achieve the best accuracy with the KNN classifier.</p> </abstract>

Opinion mining for national security: techniques, domain applications, challenges and research opportunities

BACKGROUND

Opinion mining, or sentiment analysis, is a field in Natural Language Processing (NLP). It extracts people's thoughts, including assessments, attitudes, and emotions toward individuals, topics, and events. The task is technically challenging but incredibly useful. With the explosive growth of the digital platform in cyberspace, such as blogs and social networks, individuals and organisations are increasingly utilising public opinion for their decision-making. In recent years, significant research concerning mining people's sentiments based on text in cyberspace using opinion mining has been explored. Researchers have applied numerous opinions mining techniques, including machine learning and lexicon-based approach to analyse and classify people's sentiments based on a text and discuss the existing gap. Thus, it creates a research opportunity for other researchers to investigate and propose improved methods and new domain applications to fill the gap.

METHODS

In this paper, a structured literature review has been done by considering 122 articles to examine all relevant research accomplished in the field of opinion mining application and the suggested Kansei approach to solve the challenges that occur in mining sentiments based on text in cyberspace. Five different platforms database were systematically searched between 2015 and 2021: ACM (Association for Computing Machinery), IEEE (Advancing Technology for Humanity), SCIENCE DIRECT, SpringerLink, and SCOPUS.

RESULTS

This study analyses various techniques of opinion mining as well as the Kansei approach that will help to enhance techniques in mining people's sentiment and emotion in cyberspace. Most of the study addressed methods including machine learning, lexicon-based approach, hybrid approach, and Kansei approach in mining the sentiment and emotion based on text. The possible societal impacts of the current opinion mining technique, including machine learning and the Kansei approach, along with major trends and challenges, are highlighted.

CONCLUSION

Various applications of opinion mining techniques in mining people's sentiment and emotion according to the objective of the research, used method, dataset, summarized in this study. This study serves as a theoretical analysis of the opinion mining method complemented by the Kansei approach in classifying people's sentiments based on text in cyberspace. Kansei approach can measure people's impressions using artefacts based on senses including sight, feeling and cognition reported precise results for the assessment of human emotion. Therefore, this research suggests that the Kansei approach should be a complementary factor including in the development of a dictionary focusing on emotion in the national security domain. Also, this theoretical analysis will act as a reference to researchers regarding the Kansei approach as one of the techniques to improve hybrid approaches in opinion mining.

Applying machine learning EEG signal classification to emotion‑related brain anticipatory activity

Machine learning approaches have been fruitfully applied to several neurophysiological signal classification problems. Considering the relevance of emotion in human cognition and behaviour, an important application of machine learning has been found in the field of emotion identification based on neurophysiological activity. Nonetheless, there is high variability in results in the literature depending on the neuronal activity measurement, the signal features and the classifier type. The present work aims to provide new methodological insight into machine learning applied to emotion identification based on electrophysiological brain activity. For this reason, we analysed previously recorded EEG activity measured while emotional stimuli, high and low arousal (auditory and visual) were provided to a group of healthy participants. Our target signal to classify was the pre-stimulus onset brain activity. Classification performance of three different classifiers (LDA, SVM and kNN) was compared using both spectral and temporal features. Furthermore, we also contrasted the performance of static and dynamic (time evolving) approaches. The best static feature-classifier combination was the SVM with spectral features (51.8%), followed by LDA with spectral features (51.4%) and kNN with temporal features (51%). The best dynamic feature classifier combination was the SVM with temporal features (63.8%), followed by kNN with temporal features (63.70%) and LDA with temporal features (63.68%). The results show a clear increase in classification accuracy with temporal dynamic features.

A Study of Multilingual Toxic Text Detection Approaches under Imbalanced Sample Distribution

Multilingual characteristics, lack of annotated data, and imbalanced sample distribution are the three main challenges for toxic comment analysis in a multilingual setting. This paper proposes a multilingual toxic text classifier which adopts a novel fusion strategy that combines different loss functions and multiple pre-training models. Specifically, the proposed learning pipeline starts with a series of pre-processing steps, including translation, word segmentation, purification, text digitization, and vectorization, to convert word tokens to a vectorized form suitable for the downstream tasks. Two models, multilingual bidirectional encoder representation from transformers (MBERT) and XLM-RoBERTa (XLM-R), are employed for pre-training through Masking Language Modeling (MLM) and Translation Language Modeling (TLM), which incorporate semantic and contextual information into the models. We train six base models and fuse them to obtain three fusion models using the F1 scores as the weights. The models are evaluated on the Jigsaw Multilingual Toxic Comment dataset. Experimental results show that the best fusion model outperforms the two state-of-the-art models, MBERT and XLM-R, in F1 score by 5.05% and 0.76%, respectively, verifying the effectiveness and robustness of the proposed fusion strategy.

Imagined character recognition through EEG signals using deep convolutional neural network

Electroencephalography (EEG)-based brain computer interface (BCI) enables people to interact directly with computing devices through their brain signals. A BCI typically interprets EEG signals to reflect the user's intent or other mental activity. Motor imagery (MI) is a commonly used technique in BCIs where a user is asked to imagine moving certain part of the body such as a hand or a foot. By correctly interpreting the signal, one can perform a multitude of tasks such as controlling wheel chair, playing computer games, or even typing text. However, the use of motor-imagery-based BCIs outside the laboratory environment is limited due to the lack of their reliability. This work focuses on another kind of mental imagery, namely, the visual imagery (VI). VI is the manipulation of visual information that comes from memory. This work presents a deep convolutional neural network (DCNN)-based system for the recognition of visual/mental imagination of English alphabets so as to enable typing directly via brain signals. The DCNN learns to extract the spatial features hidden in the EEG signal. As opposed to many deep neural networks that use raw EEG signals for classification, this work transforms the raw signals into band powers using Morlet wavelet transformation. The proposed approach is evaluated on two publicly available benchmark MI-EEG datasets and a visual imagery dataset specifically collected for this work. The obtained results demonstrate that the proposed model performs better than the existing state-of-the-art methods for MI-EEG classification and yields an average accuracy of 99.45% on the two public MI-EEG datasets. The model also achieves an average recognition rate of 95.2% for the 26 English-language alphabets. Overall working of the proposed solution for imagined character recognition through EEG signals.