Lingjia Deng

Sentiment Propagation via Implicature Constraints

Opinions may be expressed implicitly via inference over explicit sentiments and events that positively/negatively affect entities (goodFor/badFor events). We investigate how such inferences may be exploited to improve sentiment analysis, given goodFor/badFor event information. We apply Loopy Belief Propagation to propagate sentiments among entities. The graph-based model improves over explicit sentiment classification by 10 points in precision and, in an evaluation of the model itself, we find it has an 89% chance of propagating sentiments correctly.

MPQA 3.0: An Entity/Event-Level Sentiment Corpus

This paper presents an annotation scheme for adding entity and event target annotations to the MPQA corpus, a rich span-annotated opinion corpus. The new corpus promises to be a valuable new resource for developing systems for entity/event-level sentiment analysis. Such systems, in turn, would be valuable in NLP applications such as Automatic Question Answering. We introduce the idea of entity and event targets (eTargets), describe the annotation scheme, and present the results of an agreement study.

Joint Prediction for Entity/Event-Level Sentiment Analysis using Probabilistic Soft Logic Models

In this work, we build an entity/event-level sentiment analysis system, which is able to recognize and infer both explicit and implicit sentiments toward entities and events in the text. We design Probabilistic Soft Logic models that integrate explicit sentiments, inference rules, and +/-effect event information (events that positively or negatively affect entities). The experiments show that the method is able to greatly improve over baseline accuracies in recognizing entity/event-level sentiments.

Lexical Acquisition for Opinion Inference: A Sense-Level Lexicon of Benefactive and Malefactive Events

Opinion inference arises when opinions are expressed toward states and events which positive or negatively affect entities, i.e., benefactive and malefactive events. This paper addresses creating a lexicon of such events, which would be helpful to infer opinions. Verbs may be ambiguous, in that some meanings may be benefactive and others may be malefactive or neither. Thus, we use WordNet to create a sense-level lexicon. We begin with seed senses culled from FrameNet and expand the lexicon using WordNet relationships. The evaluations show that the accuracy of the approach is well above baseline accuracy.

A Conceptual Framework for Inferring Implicatures

While previous sentiment analysis research has concentrated on the interpretation of explicitly stated opinions and attitudes, this work addresses a type of opinion implicature (i.e., opinion-oriented default inference) in real-world text. This work describes a rule-based conceptual framework for representing and analyzing opinion implicatures. In the course of understanding implicatures, the system recognizes implicit sentiments (and beliefs) toward various events and entities in the sentence, often of mixed polarities; thus, it produces a richer interpretation than is typical in opinion analysis.

An Investigation for Implicatures in Chinese : Implicatures in Chinese and in English are similar !

Implicit opinions are commonly seen in opinion-oriented documents, such as political editorials. Previous work have utilized opinion inference rules to detect implicit opinions evoked by events that positively/negatively affect entities (goodFor/badFor) to improve sentiment analysis for English text. Since people in different languages may express implicit opinions in different ways, in this work we investigate implicit opinions expressed via goodFor/badFor events in Chinese. The positive results have provided evidences that such implicit opinions and inference rules are similar in Chinese and in English. Moreover, we have observed cases where the inferences are blocked.

How can NLP Tasks Mutually Benefit Sentiment Analysis? A Holistic Approach to Sentiment Analysis

Existing opinion analysis techniques rely on the clues within the sentence that focus on the sentiment analysis task itself. However, the sentiment analysis task is not isolated from other NLP tasks (co-reference resolution, entity linking, etc) but they can benefit each other. In this paper, we define dependencies between sentiment analysis and other tasks, and express the dependencies in first order logic rules regardless of the representations of different tasks. The conceptual framework proposed in this paper using such dependency rules as constraints aims at exploiting information outside the sentence and outside the document to improve sentiment analysis. Further, the framework allows exception to the rules.

Entity/Event-Level Sentiment Detection and Inference

Sentiment analysis aims at recognizing and understanding opinions expressed in languages. Previous work in sentiment analysis focused on extracting explicit opinions, which are directly expressed via sentiment words. However, opinions may be expressed implicitly via inferences over explicit sentiments. For example, in the sentence It is great that he was promoted. versus It is great that he was fired, there is an explicitly positive sentiment in both sentences because of the positive sentiment word great. Previous work may stop here. However, the sentiment toward he in the former sentence is positive, while the sentiment toward he in the later sentence is negative. The sentiments toward he in both sentences are implicit since there is no sentiment word directly modifying he. The implicit opinions are indicated in the text, and they are important for a sentiment analysis system to fully understand the documents. While previous work cannot recognize such implicit sentiment, this thesis contributes to developing an entity/event-level sentiment analysis system to recognize both explicit and implicit sentiments expressed from entities toward entities and events. Specifically, we first give the definitions of the entity/event-level sentiment analysis task. Since this is a new task, we develop two corpora serving as resources for this task. The implicit sentiments cannot be recognized merely relying on sentiment lexicons since the implicit sentiments are not directly associated with sentiment words. Inference rules are needed to recognize the implicit sentiments. Instead of developing a rule-based system to automatically infer implicit opinions, we develop computational models which use the inference rules as soft constraints. What’s more important, the models take into account the information not only from sentiment analysis tasks, but also from other Natural Language Processing tasks including information extraction and semantic role labeling. The models jointly solve different NLP tasks in one single model and improve the performances of the tasks. We also contribute to improving recognizing sources of opinions in this thesis. Finally, we conduct an analysis study showing that the idea of sentiment inference defined in this thesis can be applied to Chinese text as well.