Cross-domain Named Entity Recognition (CD-NER) is crucial for Knowledge Graph (KG) construction and natural language processing (NLP), enabling learning from source to target domains with limited data.Previous studies often rely on manually collected entity-relevant sentences from the web or attempt to bridge the gap between tokens and entity labels across domains.These approaches are time-consuming and inefficient, as these data are often weakly correlated with the target task and require extensive pre-training.To address these issues, we propose automatically generating task-oriented knowledge (GTOK) using large language models (LLMs), focusing on the reasoning process of entity extraction.Then, we employ taskoriented pre-training (TOPT) to facilitate domain adaptation.Additionally, current crossdomain NER methods often lack explicit explanations for their effectiveness.Therefore, we introduce the concept of information density to better evaluate the model's effectiveness before performing entity recognition.We conduct systematic experiments and analyses to demonstrate the effectiveness of our proposed approach and the validity of using information density for model evaluation † .

The proliferation of COVID-19 fake news on social media poses a severe threat to the health information ecosystem. We show that affective computing can make significant contributions to combat this infodemic. Given that fake news is often presented with emotional appeals, we propose a new perspective on the role of emotion in the attitudes, perceptions, and behaviors of the dissemination of information. We study emotions in conjunction with fake news, and explore different aspects of their interaction. To process both emotion and ‘falsehood’ based on the same set of data, we auto-tag emotions on existing COVID-19 fake news datasets following an established emotion taxonomy. More specifically, based on the distribution of seven basic emotions (e.g.Happiness, Like, Fear, Sadness, Surprise, Disgust, Anger), we find across domains and styles that COVID-19 fake news is dominated by emotions ofFear(e.g., of coronavirus), andDisgust(e.g., of social conflicts). In addition, the framing of fake news in terms of gain-versus-loss reveals a close correlation between emotions, perceptions, and collective human reactions. Our analysis confirms the significant role of emotionFearin the spreading of the fake news, especially when contextualized in the loss frame. Our study points to a future direction of incorporating emotion footprints in models of automatic fake news detection, and establishes an affective computing approach to information quality in general and fake news detection in particular.

Knowledge distillation is an effective method to transfer knowledge from a large pre-trained teacher model to a compacted student model. However, in previous studies, the distilled student models are still large and remain impractical in highly speed-sensitive systems (e.g., an IR system). In this study, we aim to distill a deep pre-trained model into an extremely compacted shallow model like CNN. Specifically, we propose a novel one-teacher and multiple-student knowledge distillation approach to distill a deep pre-trained teacher model into multiple shallow student models with ensemble learning. Moreover, we leverage large-scale unlabeled data to improve the performance of students. Empirical studies on three sentiment classification tasks demonstrate that our approach achieves better results with much fewer parameters (0.9%-18%) and extremely high speedup ratios (100X-1000X).

The classification of implicit emotions in text has always been a great challenge to emotion processing. Even though the majority of emotion expressed implicitly, most previous attempts at emotions have focused on the examination of explicit emotions. The poor performance of existing emotion identification and classification models can partly be attributed to the disregard of implicit emotions. In view of this, this paper presents the development of a Chinese event-comment social media emotion corpus. The corpus deals with both explicit and implicit emotions with more emphasis being placed on the implicit ones. This paper specifically describes the data collection and annotation of the corpus. An annotation scheme has been proposed for the annotation of emotion-related information including the emotion type, the emotion cause, the emotion reaction, the use of rhetorical question, the opinion target (i.e. the semantic role in an event that triggers an emotion), etc. Corpus data shows that the annotated items are of great value to the identification of implicit emotions. We believe that the corpus will be a useful resource for both explicit and implicit emotion classification and detection as well as event classification.

Previous research on emotions analysis has placed much emphasis in monolingual instead of bilingual text. However, emotions on social media platforms are often found in bilingual or code-switching posts. Different from monolingual text, emotions in code-switching text can be expressed in both monolingual and bilingual forms. Moreover, more than one emotion can be expressed within a single post; yet they tend to be related in some ways which offers some implications. It is thus necessary to consider the correlation between different emotions. In this paper, a joint factor graph model is proposed to address this issue. In particular, attribute functions of the factor graph model are utilized to learn both monolingual and bilingual information from each post, factor functions are used to explore the relationship among different emotions, and a belief propagation algorithm is employed to learn and predict the model. Empirical studies demonstrate the importance of emotion analysis in code-switching text and the effectiveness of our proposed joint learning model.

Emotions in code-switching text can be expressed in either monolingual or bilingual forms. However, relatively little research has emphasized on code-switching text. In this paper, we propose a Bilingual Attention Network (BAN) model to aggregate the monolingual and bilingual informative words to form vectors from the document representation, and integrate the attention vectors to predict the emotion. The experiments show that the effectiveness of the proposed model. Visualization of the attention layers illustrates that the model selects qualitatively informative words.

Sentiment and emotion classification have been popularly but separately studied in natural language processing. In this paper, we address joint learning on sentiment and emotion classification where both the labeled data for sentiment and emotion classification are available. The objective of this joint-learning is to benefit the two tasks from each other for improving their performances. Specifically, an extra data set that is annotated with both sentiment and emotion labels are employed to estimate the transformation probability between the two kinds of labels. Furthermore, the transformation probability is leveraged to transfer the classification labels to benefit the two tasks from each other. Empirical studies demonstrate the effectiveness of our approach for the novel joint learning task.

Most theories of emotion treat recognition of a triggering cause event as an integral part of emotion processing. This paper proposes emotion cause detection as a new research area in emotion processing. As a first step toward fully automatic inference of emotion‐cause correlation, we propose a text‐driven, rule‐based approach to emotion cause detection in Chinese. First, we constructed a Chinese emotion cause annotated corpus based on our proposed annotation scheme. Next, we analyzed the corpus data, which yielded the identification of seven groups of linguistic cues and two sets of generalized linguistic rules for the detection of emotion causes. We then developed a rule‐based system for emotion cause detection based on the linguistic rules. In addition, we proposed an evaluation scheme with two phases for performance assessment. The results of our experiments show that our system achieved a promising performance for cause occurrence detection, as well as for cause event detection. The current study should lay the groundwork for future research on the inferences of implicit information and the discovery of new information based on cause‐event relation.

Sentiment classification has undergone significant development in recent years. However, most existing studies assume the balance between negative and positive samples, which may not be true in reality. In this paper, we investigate imbalanced sentiment classification instead. In particular, a novel clustering-based stratified under-sampling framework and a centroid-directed smoothing strategy are proposed to address the imbalanced class and feature distribution problems respectively. Evaluation across different datasets shows the effectiveness of both the under-sampling framework and the smoothing strategy in handling the imbalanced problems in real sentiment classification applications.