In this paper, we systematically explore feature definition and selection strategies for sentiment polarity classification. We begin by exploring basic questions, such as whether to use stemming, term frequency versus binary weighting, negation-enriched features, n-grams or phrases. We then move onto more complex aspects including feature selection using frequency-based vocabulary trimming, part-of-speech and lexicon selection (three types of lexicons), as well as using expected Mutual Information (MI). Using three product and movie review datasets of various sizes, we show, for example, that some techniques are more beneficial for larger datasets than the smaller. A classifier trained on only few features ranked high by MI outperformed one trained on all features in large datasets, yet in small dataset this did not prove to be true. Finally, we perform a space and computation cost analysis to further understand the merits of various feature types.

Online news sites often include a comments section where readers are allowed to leave their thoughts. These comments often contain interesting and insightful conversations between readers about the news article. However the richness of these conversations is often lost among other meaningless comments, and moreover all comments are found at the bottom of the web page. In this article, we discuss how our system inserts reader conversations into the news article to create a multimedia presentation called Shout Out. Shout Out features two virtual news anchors: one anchor reads the news and when appropriate the anchor pauses to have a conversation about the news with another anchor. This current iteration of Shout Out combines natural language techniques and reader conversations to create an engaging system.

This paper addresses the task of user classification in social media, with an application to Twitter. We automatically infer the values of user attributes such as political orientation or ethnicity by leveraging observable information such as the user behavior, network structure and the linguistic content of the user’s Twitter feed. We employ a machine learning approach which relies on a comprehensive set of features derived from such user information. We report encouraging experimental results on 3 tasks with different characteristics: political affiliation detection, ethnicity identification and detecting affinity for a particular business. Finally, our analysis shows that rich linguistic features prove consistently valuable across the 3 tasks and show great promise for additional user classification needs.

Spoken dialogue systems promise efficient and natural access to a large variety of information sources and services from any phone. However, current spoken dialogue systems are deficient in their strategies for preventing, identifying and repairing problems that arise in the conversation. This paper reports results on automatically training a Problematic Dialogue Predictor to predict problematic human-computer dialogues using a corpus of 4692 dialogues collected with the 'How May I Help You' (SM) spoken dialogue system. The Problematic Dialogue Predictor can be immediately applied to the system's decision of whether to transfer the call to a human customer care agent, or be used as a cue to the system's dialogue manager to modify its behavior to repair problems, and even perhaps, to prevent them. We show that a Problematic Dialogue Predictor using automatically-obtainable features from the first two exchanges in the dialogue can predict problematic dialogues 13.2% more accurately than the baseline.

Designing the dialogue policy of a spoken dialogue system involves many nontrivial choices. This paper presents a reinforcement learning approach for automatically optimizing a dialogue policy, which addresses the technical challenges in applying reinforcement learning to a working dialogue system with human users. We report on the design, construction and empirical evaluation of NJFun, an experimental spoken dialogue system that provides users with access to information about fun things to do in New Jersey. Our results show that by optimizing its performance via reinforcement learning, NJFun measurably improves system performance.

This paper describes a novel method by which a spoken dialogue system can learn to choose an optimal dialogue strategy from its experience interacting with human users. The method is based on a combination of reinforcement learning and performance modeling of spoken dialogue systems. The reinforcement learning component applies Q-learning (Watkins, 1989), while the performance modeling component applies the PARADISE evaluation framework (Walker et al., 1997) to learn the performance function (reward) used in reinforcement learning. We illustrate the method with a spoken dialogue system named ELVIS (EmaiL Voice Interactive System), that supports access to email over the phone. We conduct a set of experiments for training an optimal dialogue strategy on a corpus of 219 dialogues in which human users interact with ELVIS over the phone. We then test that strategy on a corpus of 18 dialogues. We show that ELVIS can learn to optimize its strategy selection for agent initiative, for reading messages, and for summarizing email folders.

But each They are linked by an arrow which is labeled by discourse method for representing a context is quite different. Our relations R. We represent SDRS in the form of boxes like study is based on two representational methods of temporal DRS. To induce a temporal and hierarchical structure, relations: the Segmented Discourse Representation Theory SDRT distinguish discourse relations'coordinating' from (SDRT) and the model of Cognitive and Applicative'subordinating', therefore coordination and subordination Grammar (CAG). This paper presents a comparison of affect the temporal order of text: the former indicate a continuation these two approaches about aspect and tense by an analysis of some discourses pattern, like relations of'Narration' of relations between events. We are not going to show all or'Result' in discourse segmentation, and the later steps of SDRT's representations, but we take a simple discourse indicate with types of information like relations of'Elaboration' (Asher and Lascarides 2003) and we analyze the or'Explanation'. These relations are appeared same discourse with the framework of the CAG.

Augmenting word tokens with a phonetic representation, derived from a dictionary, improves the performance of a Natural Language Understanding component that interprets speech recognizer output: we observed a 5% to 7% reduction in errors across a wide range of response return rates. The best performance comes from mixture models incorporating both word and phone features. Since the phonetic representation is derived from a dictionary, the method can be applied easily without the need for integration with a specific speech recognizer. The method has similarities with autonomous (or bottom-up) psychological models of lexical access, where contextual information is not integrated at the stage of auditory perception but rather later.

The aim of this paper is threefold: (1) it explores methods for the detection of affective states in text, (2) it presents the usage of such affective cues in a conversational system and (3) it evaluates its effectiveness in a virtual reality setting. Valence and arousal values, used for generating facial expressions of users' avatars, are also incorporated into the dialog, helping to bridge the gap between textual and visual modalities. The system is evaluated in terms of its ability to: (i) generate a realistic dialog, (ii) create an enjoyable chatting experience, and (iii) establish an emotional connection with participants. Results show that user ratings for the conversational agent match those obtained in a Wizard of Oz setting.

When people interact with language-producing agents (other people or computers), they assume that the shared experience leads to shared representations — of the world, the interaction, and the language used in the interaction. This phenomenon occurs even during interaction with systems that give no evidence of building shared representations. The absence of shared representations leads to errors and delays; alternatively, even simple shared representations can lead to reduced error rates and more efficient interaction. In this talk, we present three case studies: a mobile local business search application that builds no interaction representations; a telephone-based recommendation and review system that builds limited representations of the shared language in the interaction; and computer models of coreference that use shared representations to permit both coreference resolution and referring expression generation. We lay out a range of possibilities for shared representations, show that they can be built incrementally as an interaction progresses, and point to possibilities for future work in probabilistic shared representations for interactive systems.