Sentiment analysis seeks to identify the viewpoint(s) underlying a text span; an example application is classifying a movie review as "thumbs up" or "thumbs down". To determine this sentiment polarity, we propose a novel machine-learning method that applies text-categorization techniques to just the subjective portions of the document. Extracting these portions can be implemented using efficient techniques for finding minimum cuts in graphs; this greatly facilitates incorporation of cross-sentence contextual constraints.

An important part of our information-gathering behavior has always been to find out what other people think. With the growing availability and popularity of opinion-rich resources such as online review sites and personal blogs, new opportunities and challenges arise as people now can, and do, actively use information technologies to seek out and understand the opinions of others. The sudden eruption of activity in the area of opinion mining and sentiment analysis, which deals with the computational treatment of opinion, sentiment, and subjectivity in text, has thus occurred at least in part as a direct response to the surge of interest in new systems that deal directly with opinions as a first-class object. This survey covers techniques and approaches that promise to directly enable opinion-oriented information-seeking systems. Our focus is on methods that seek to address the new challenges raised by sentiment-aware applications, as compared to those that are already present in more traditional fact-based analysis. We include materialon summarization of evaluative text and on broader issues regarding privacy, manipulation, and economic impact that the development of opinion-oriented information-access services gives rise to. To facilitate future work, a discussion of available resources, benchmark datasets, and evaluation campaigns is also provided. 1

We address the rating-inference problem, wherein rather than simply decide whether a review is “thumbs up ” or “thumbs down”, as in previous sentiment analysis work, one must determine an author’s evaluation with respect to a multi-point scale (e.g., one to five “stars”). This task represents an interesting twist on standard multi-class text categorization because there are several different degrees of similarity between class labels; for example, “three stars ” is intuitively closer to “four stars ” than to “one star”. We first evaluate human performance at the task. Then, we apply a metaalgorithm, based on a metric labeling formulation of the problem, that alters a given-ary classifier’s output in an explicit attempt to ensure that similar items receive similar labels. We show that the meta-algorithm can provide significant improvements over both multi-class and regression versions of SVMs when we employ a novel similarity measure appropriate to the problem. 1

What kinds of social relationships can people have with computers? Are there activities that computers can engage in that actively draw people into relationships with them? What are the potential benefits to the people who participate in these human-computer relationships? To address these questions this work introduces a theory of Relational Agents, which are computational artifacts designed to build and maintain long-term, social-emotional relationships with their users. These can be purely software humanoid animated agents--as developed in this work--but they can also be non-humanoid or embodied in various physical forms, from robots, to pets, to jewelry, clothing, hand-helds, and other interactive devices. Central to the notion of relationship is that it is a persistent construct, spanning multiple interactions; thus, Relational Agents are explicitly designed to remember past history and manage future expectations in their interactions with users. Finally, relationships are fundamentally social and emotional, and detailed knowledge of human social psychology--with a particular emphasis on the role of affect--must be incorporated into these agents if they are to effectively leverage the mechanisms of human social cognition in order to build relationships in the most natural manner possible. People build

In this paper we present a linguistic resource for the lexical representation of affective knowledge. This resource (named WORDNET-AFFECT) was developed starting from WORDNET, through a selection and tagging of a subset of synsets representing the affective meanings. 1.

■ A long-standing dream of artificial intelligence has been to put commonsense knowledge into computers—enabling machines to reason about everyday life. Some projects, such as Cyc, have begun to amass large collections of such knowledge. However, it is widely assumed that the use of common sense in interactive applications will remain impractical for years, until these collections can be considered sufficiently complete and commonsense reasoning sufficiently robust. Recently, at the Massachusetts Institute of Technology’s Media Laboratory, we have had some success in applying commonsense knowledge in a number of intelligent interface agents, despite the admittedly spotty coverage and unreliable inference of today’s

In addition to information, text contains attitudinal, and more specifically, emotional content. This paper explores the text-based emotion prediction problem empirically, using supervised machine learning with the SNoW learning architecture. The goal is to classify the emotional affinity of sentences in the narrative domain of children’s fairy tales, for subsequent usage in appropriate expressive rendering of text-to-speech synthesis. Initial experiments on a preliminary data set of 22 fairy tales show encouraging results over a naïve baseline and BOW approach for classification of emotional versus non-emotional contents, with some dependency on parameter tuning. We also discuss results for a tripartite model which covers emotional valence, as well as feature set alternations. In addition, we present plans for a more cognitively sound sequential model, taking into consideration a larger set of basic emotions. 1

Abstract. ConceptNet is a very large semantic network of commonsense knowledge suitable for making various kinds of practical inferences over text. ConceptNet captures a wide range of commonsense concepts and relations like those in Cyc, while its simple semantic network structure lends it an ease-of-use comparable to WordNet. To meet the dual challenge of having to encode complex higher-order concepts, and maintaining ease-of-use, we introduce a novel use of semi-structured natural language fragments as the knowledge representation of commonsense concepts. In this paper, we present a methodology for reasoning flexibly about these semi-structured natural language fragments. We also examine the tradeoffs associated with representing commonsense knowledge in formal logic versus in natural language. We conclude that the flexibility of natural language makes it a highly suitable representation for achieving practical inferences over text, such as context finding, inference chaining, and conceptual analogy. 1 What is ConceptNet? ConceptNet (www.conceptnet.org) is the largest freely available, machine-useable

What are the sources of happiness and sadness in everyday life? In this paper, we employ ‘linguistic ethnography ’ to seek out where happiness lies in our everyday lives by considering a corpus of blogposts from the LiveJournal community annotated with happy and sad moods. By analyzing this corpus, we derive lists of happy and sad words and phrases annotated by their ‘happiness factor.’ Various semantic analyses performed with this wordlist reveal the happiness trajectory of a 24-day (3am and 9-10p are most happy), and a 7-day week (Wednesdays are saddest), and compare the socialness and humancenteredness of happy descriptions versus sad descriptions. We evaluate our corpus-based approach in a classification task and contrast our wordlist with emotionally-annotated

One of the long-term goals of Artificial Intelligence is construction of a machine that is capable of reasoning about the everyday world the way humans are. In this paper, I first argue that construction of a large collection of statements about everyday world (a repository of commonsense knowledge) is a valuable step towards this long-term goal. Then, I point out that volunteer contributors over the Internet — a frequently overlooked source of knowledge — can be tapped to construct such a knowledge repository. To operationalize construction of a large commonsense knowledge repository by volunteer contributors, I then introduce cumulative analogy, a class of analogy-based reasoning algorithms that leverage existing knowledge to pose knowledge acquisition questions to the volunteer contributors. The algorithms have been implemented and deployed as the Learner system. To date, about 3,400 volunteer contributors have interacted with the system over the course of 11 months, increasing a starting collection of 47,147 statements by 362 % to a total of 217,971. The deployed system and the growing collection of knowledge it acquired are publicly available from