Sophia Yat Mei Lee

Semi-supervised learning for imbalanced sentiment classification

Various semi-supervised learning methods have been proposed recently to solve the long-standing shortage problem of manually labeled data in sentiment classification. However, most existing studies assume the balance between negative and positive samples in both the labeled and unlabeled data, which may not be true in reality. In this paper, we investigate a more common case of semi-supervised learning for imbalanced sentiment classification. In particular, various random subspaces are dynamically generated to deal with the imbalanced class distribution problem. Evaluation across four domains shows the effectiveness of our approach.

Imbalanced sentiment classification

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.

A Cognitive-based Annotation System for Emotion Computing

Emotion computing is very important for expressive information extraction. In this paper, we provide a robust and versatile emotion annotation scheme based on cognitive emotion theories, which not only can annotate both explicit and implicit emotion expressions, but also can encode different levels of emotion information for the given emotion content. In addition, motivated by a cognitive framework, an automatic emotion annotation system is developed, and large and comparatively high-quality emotion corpora are created for emotion computing, one in Chinese and the other in English. Such an annotation system can be easily adapted for different kinds of emotion applications and be extended to other languages.

Detecting emotion causes with a linguistic rule-based approach

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.

A robust web personal name information extraction system

Highlights? Features are extracted with various lightweight methods and from broad resources. ? The unsupervised features improve the robustness of a disambiguation system. ? Our AE system integrates various extraction approaches with high precision. ? Each integrated AE approach exactly extracts some of the right target information. Personal information extraction, which extracts the persons in question and their related information (such as biographical information and occupation) from web, is an important component to construct social network (a kind of semantic web). For this practical task, two important issues are to be discussed: personal named entity ambiguity and the extraction of personal information for a specific person. For personal named entity ambiguity, which is a common phenomenon in the fast growing web resource, we propose a robust system which extracts lightweight features with a totally unsupervised approach from broad resources. The experiments show that these lightweight features not only improve the performances, but also increase the robustness of a disambiguation system. To extract the information of the focus person, an integrated system is introduced, which is able to effectively re-use and combine current well-developed tools for web data, and at the same time, to identify the expression properties of web data. We show that our flexible extraction system achieves state-of-the-art performances, especially the high precision, which is very important for real applications.

Joint learning on sentiment and emotion classification

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.

Sentiment Classification with Polarity Shifting Detection

Sentiment classification is now a hot research issue in the community of natural language processing and the bag-of-words based machine learning approach is the state-of-the-art for this task. However, one important phenomenon, called polarity shifting, remains unsolved in the bag-of-words model, which sometimes makes the machine learning approach fails. In this study, we aim to perform sentiment classification with full consideration of the polarity shifting phenomenon. First, we extract some detection rules for detecting polarity shifting of sentimental words from a corpus which consists of polarity-shifted sentences. Then, we use the detection rules to detect the polarity-shifted words in the testing data. Third, a novel term counting-based classifier is designed by fully considering those polarity-shifted words. Evaluation shows that the novel term counting-based classifier significantly improves the performance of sentiment analysis across five domains. Furthermore, when this classifier is combined with a machine-learning based classifier, the combined classifier yields better performance than either of them.

Emotion in Code-switching Texts: Corpus Construction and Analysis

Previous researches have focused on analyzing emotion through monolingual text, when in fact bilingual or code-switching posts are also common in social media. Despite the important implications of code-switching for emotion analysis, existing automatic emotion extraction methods fail to accommodate for the code-switching content. In this paper, we propose a general framework to construct and analyze the code-switching emotional posts in social media. We first propose an annotation scheme to identify the emotions associated with the languages expressing them in a Chinese-English code-switching corpus. We then make some observations and generate statistics from the corpus to analyze the linguistic phenomena of code-switching texts in social media. Finally, we propose a multiple-classifier-based automatic detection approach to detect emotion in the codeswitching corpus for evaluating the effectiveness of both Chinese and English texts.

Gender versus Politics: When Conceptual Models Collide in the US Senate

One of the main purposes of political speeches is to persuade others of one’s opinion. Nowhere is this more apparent than on the floor of a democratically elected legislative body, where legislators gain floor time to convince others of the validity of their points of view. One method political leaders employ to this end, either consciously or unconsciously, involves incorporating appropriate conceptual metaphors into their speeches. Recent work has focused on the analysis of metaphors used by presidents and prime ministers (Charteris-Black 2004, 2005, 2007, Chilton and Ilyin 1993, Lu and Ahrens 2008, Semino and Masci 1996). However, less attention has been spent on political leaders at the next level of statesmanship: the senators, cabinet ministers and members of parliament, an area which several chapters in this volume now address (Chs 5, 7–9 and 12). In this chapter, we examine the use of lexemes associated with two conceptual metaphor models in US senatorial speech from 2000 to early 2007 in order to determine if gender, political party affiliation, or a combination of both gender and party in the US Senate influences the conceptual models invoked by the senators. We find that as a group, senators do not invoke a particular conceptual model on the basis of gender. Instead, the conceptual model most often invoked across all groups is the model that Lakoff (1996/2002) postulates to be associated with the Democratic political party.

An Event-Based Emotion Corpus

As part of a larger project, this paper presents some of the work done regarding our proposal of an event-based analysis of emotion. We propose that an emotion is treated as a pivot event linking the events inducing (i.e. pre-events), and induced by (i.e. post-events), said emotion. Our study begins with the development of an emotion corpus annotated with pre-and post-events. We then provide a collocational pattern analysis as well as a linguistic analysis of the links between event structures and emotions in the text. The project goal is to develop a theory predicting the dependencies between emotions and events, based on linguistic cues in context.