19-07-2012, 03:37 PM
Statistical Method for Attitude Classification from Text Corpus
Statistical Method for Attitude Classification from Text Corpus.docx (Size: 290.65 KB / Downloads: 115)
Introduction
In the presented work, we address the task of the classification of attitude in Text documents. In order to achieve truly natural classification of attitude in text documents, we set focus in our research to Sentiment Classification from text.
1.1. AttitudeClassification From Text
Attitudes have been widely studied in psychology and behavior sciences, as they are an important element of human nature. They have also attracted the attention of researchers in computer science, especially in the field of human computer interaction, where studies have been carried out on facial expressions or on the recognition of attitudes through a variety of sensors[1] .
In computational linguistics, all words can potentially convey affective meaning. Every word, even those apparently neutral, can evoke pleasant or painful experiences due to their semantic relation with attitude’s concepts or categories. Some words have attitude’s meaning with respect to an individual story, while for many others the affective power is part of the collective imagination (e.g. words such as .mum., .ghost., .war.). The automatic detection of attitude in texts is becoming increasingly important from an applicative point of view. Consider for example the tasks of opinion mining and market analysis, affective computing, or natural language interfaces such as e-learning environments or educational/edutainment games. Possible beneficial effects of attitudes on memory and attention of the users, and in general on fostering their creativity are also well-known in psychology field.[2]
For instance, the following represent examples of applicative scenarios in which affective analysis would give valuable and interesting contributions:
• Sentiment Analysis. Text categorization according to affective relevance, opinion exploration for market analysis, etc. are just some examples of application of these techniques. While positive/ negative valence checking is an active field of sentiment analysis, we believe that a fine-grained attitude checking would increase the effectiveness of the field.
• Computer Assisted Creativity. The automated generation of evaluative expressions with a bias on some polarity orientation are a key component for automatic personalized advertisement and persuasive communication.
• Verbal Expressivity in Human Computer Interaction. Future human-computer interaction, accordingto a widespread view, will emphasizenaturalness and effectiveness and hence theincorporation of models of possibly many humancognitive capabilities, including affectiveanalysis and generation. For example, attitudesexpression by synthetic characters (e.g.embodied conversational agents) is considerednow a key element for their believability.Affective words selection and understanding iscrucial for realizing appropriate and expressiveconversations.The .Affective Text. task is intended as an explorationof the connection between lexical semanticsand attitudes, and an evaluation of various automaticapproaches to attitude recognition.The task is not easy. Indeed, as (Ortony etal., 1987) indicates, besides words directly referringto attitude’s states (e.g. .threat., .cheerful.) andfor which an appropriate lexicon would help, thereare words that act only as an indirect reference toattitudes depending on the context (e.g. .monster.,.ghost.). We can call the former direct affectivewords and the latter indirect affective words (Strapparavaet al., 2006)[1,2].
Recognition and analysis of human attitudes have attracted a lot of interest in the past two decades and have been researched extensively in neuroscience, psychology, cognitive sciences, and computer sciences. Attitudes have been widely studied in psychology and behavior sciences, as they are an important element of human nature. They have also attracted the attention of researchers in computer science, especially in the field of human computer interaction. Most of the past research in machine analysis of human attitude has focused on recognition of prototypic expressions of six basic attitudes based on data that has been posed on demand and acquired in laboratory settings. In computational linguistics, the automatic detection of attitudes in texts is becoming increasingly important from an applicative point of view.[2]
With the development of Internet, the information which appears by text form are more and more frequent. It becomes us one kind of the most easily to gain and the richest interactive resources. Presently, however, the text attitude analysis aspect's research is little. Nowadays, with the un-causing development of natural language technology, people could extract attitude information from text through analyzing grammar structure, Semantic information and attitude glossary methods etc. From the massive texts withdraws the attitude information which is contained in them has the broad application prospect in many aspects. For example, automated analysis the received mail attitude information, and give relevant attitude computing result before user reads. Through attitude analysis to the information in homepage, it may examine the homepage which contain violent attitude, realize homepage filtration and ensure the net’s information security. Along with natural language processing technology unceasing development, the text attitude computation has a richer method and reliable theory basis. At present, the text attitude computation has the widespread application prospect in many domains. For example, speech synthesis, information security, intelligent robot, pattern recognition, personalized text, and analysis article attitude structure aspects and so on.[4]
1.2. Attitude
Attitudes play the role of a sensitive catalyst, which fosters lively interactions between human beings and assists in the development and regulation of interpersonal relationships. The expression of attitudes shapes social interactions by providing observers a rich channel of information about the conversation partner [6] or his social intentions [7], by evoking positive or negative responses in others [8], and by stimulating other’s social behavior. Keltner et al. [9] highlight that “facial expressions are more than just markers of internal states,” which also serve unique functions in a social environment. By accentuating the functional role of attitudes, Frijda [10,11] argues that they preserve and enhance life, and Lutz [12] emphasizes their communicative, moral, and cultural purposes. The richness of attitude’s communication greatly benefits from the expressiveness of verbal (spoken words, prosody) and nonverbal (gaze, face, gestures, body pose) cues that enable auditory and visual channels of communication [5]. All types of expressive means potentially carry communicative power and promote better understanding [13]. Attitude’s information can be:
(1) encoded lexically within the actual words (affective predicates, intensifiers, modals, hedges, etc.) of the sentence, syntactically by means of subordinate clauses, and morphologically through changes in attitudinal shades of word meaning using suffixes (especially, in languages with rich inflectional system, such as Russian or Italian),
(2) consciously or unconsciously conveyed through vast repertoire of prosodic features like intonation, stress, pitch, loudness, juncture, and rate of speech,
(3) visually reflected through subtle details of contraction of facial muscles, tints of facial skin, eye movements, gestures, and body postures [14]. The attitude’s significance of an utterance is accompanied, complemented, and modified by vocal and visual cues.
To establish a social and friendly atmosphere, people should be able to express attitudes. However, media for online communication lack the physical contact and visualization of attitude’s reactions of partners involved in a remote text-mediated conversation, limiting thus the source of information to text messages and to graphical representations of users (avatars) that are to some degree controlled by a person. Trends show that people often try to enrich their interaction online, introducing affective symbolic conventions or emphases into text (emoticons, capital letters, etc.) [15], coloring attitude’s messages, or manually controlling the expressiveness of avatars in order to supplement the lack of paralinguistic cues.
Attitude theory has been an important field of research for a long time. Generally, attitudes describe subjective feelings in short periods of time that are related to events, persons, or objects. There are different theoretical approaches about the nature and the acquisition of attitudes. Since the attitude’s state of humans is a highly subjective experience it is very hard to find objective definition or universal terms. That is why there are several approaches to model attitudes in the psychological literature. The two most important approaches are the definition of discrete attitude categories, the so called basic attitudes, and the utilization of continuous attitude dimensions. These two approaches can also be utilized for the application of automatic attitude recognition.
Attitudes play the role of a sensitive catalyst, which fosters lively interactions between human beings and assists in the development and regulation of interpersonal relationships. The expression of attitudes shapes social interactions by providing observers a rich channel of information about the conversation partner or his social intentions, by evoking positive or negative responses in others, and by stimulating other’s social behavior. To establish a social and friendly atmosphere, people should be able to express attitudes.