Search

Sentiment Analysis (SA) can be applied to dialogues to determine the emotional tone throughout the conversation. This is beneficial for dialogue systems because it may improve human-computer interaction. For instance, in case of negative sentiment, the system may switch to a human operator who can handle the situation more effectively. However, given that dialogues are a series of utterances, the context, including the previous text, plays a crucial role in analyzing the current sentiment. Our aim is to investigate the importance of context when monitoring the sentiment of every utterance during a conversation. To accomplish this goal, we assess sentiment analysis in dialogues with varying levels of context, specifically differing in the number and author of preceding utterances. We conduct experiments on Portuguese customer-support conversations, with each utterance manually labeled as having negative or non-negative sentiment. We test a wide range of text classification approaches, from traditional, as simplicity should not be overlooked, to more recent methods, as they are more likely to achieve better performances. Results indicate that the relevance of context varies. However, context assumes particular value in human-computer dialogues, when considering both speakers, and in shorter human-human conversations, when focusing on the client. Moreover, the best classifier for both scenarios, based on BERT, achieves the highest scores when considering the context.

The article reveals the concept of language strategies and tactics, touches on the concept of linguistic influence and its classification, the role of communicators in the act of speaking. A communicative strategy is perceived as a type of behavior of one of the partners in a situation of dialogic communication. For the analysis of language strategies, the concepts of goal, motive, and action, thanks to which the linguistic goal is achieved, are the most important. This article examines the concepts of language strategies and tactics, their classification and influence on the communication process. In addition, attention is paid to the role of communicators in the speech process and their influence on language strategy. The basic idea is that communicators can use different strategies and tactics depending on the goal, motive and action leading to the achievement of the linguistic goal. For the analysis of language strategies, the most important concepts are purpose and motive. A goal is a desired result that communication participants want to achieve. Motive is what motivates communicators to use a specific language strategy. In addition, the article highlights the role of such factors as language experience, socio-cultural context and psychological characteristics of communicators in choosing a language strategy. The article also considers in detail the classification of language strategies and tactics. In particular, examples of such strategies as "dominant", "regulatory", "cooperative" and "conflict" are given. For each strategy, its purpose and ways of achieving the language goal are described. In general, the article is devoted to the understanding of different language strategies and their impact on the communication process. Knowing about these strategies and tactics can help communication participants achieve more effective and productive communication. [ABSTRACT FROM AUTHOR]

This paper proposes a spoken dialogue analysis platform (SDAP) that could assist counsellors in person-to-person counselling by analysing counselling conversations and providing key information that could enhance the counsellors' understanding of the counselees' conditions and situations. The proposed platform has two main modules: a speech recognition module and a text analysis module that are specifically built for the Korean language. The speech recognition module uses NAVER CLOVA Speech service to convert voice recordings of counselling dialogues into text. The Korean text analysis environment of the text analysis module was built using NLTK, KoNLPy and scikit-learn library, and, for now, the module provides two types of text analysis: keyword analysis and sentiment analysis. The results of the text analyses that provide keywords and analysis of customers' emotional state can help counsellors to provide appropriate feedback to the counselees easily and more quickly, making the counselling fast and effective and reducing the counselees' waiting time. In the experiments, the text analysis module building process is elaborated in detail, and the usefulness of the proposed SDAP is exemplified by case studies on actual counselling conversations at a dental clinic and a fitness centre.

The characteristics of the so-called discourse markers play an important role in the stylistic analysis of spontaneous spoken communication. The term discourse markers refers to the group of synsemantic words (or their collocations) that convey primarily pragmatic information. When describing discourse markers, their polyfunctionality is always emphasized. Discourse markers can perform a variety of (textual and interaction-related) functions, depending on the context, their structural position in the utterance, and especially their sound modifications. The article is based on the analysis of discourse markers in terms of their expressive and emotional functions. The use of discourse markers to express the speaker’s emotional attitudes is connected with three processes: sound modification, multiplication (repetition), and cumulation (formation of multiwords markers). The main source of language material for the research was the corpus of contemporary spoken Czech ORAL v1.

Philosophical dialogue is an effective way to deepen one's thoughts, but it is not easy to practice it because humans have emotions. We proposed the use of a robot in practicing philosophical dialogue and experimentally investigated how philosophical dialogue with a robot differs from philosophical dialogue with a human. The results of the experiment showed that (1) participants talking to a human spend more time answering than those talking to a robot, (2) the increase in the time participants spend answering comes from the increase in the time participants spend speaking and is not influenced by reaction latency and pause time so much, (3) the reason for the increase in the time spent speaking is that some participants who talked to a human were likely to choose their words so as not to make the interlocutor uncomfortable and rephrased their thoughts so that they were easier for the interlocutor to understand, and some participants who talked to a robot might have thought that the robot would not be concerned even if they give a brief answer, and finally (4) it seems there is no significant difference in the depth of thought between participants talking to a human and participants talking to a robot. These results suggest the effectiveness of using robots for philosophical dialogue, in particular for people who feel nervous about talking to others. [ABSTRACT FROM AUTHOR]

Copyright of Signata is the property of Presses Universitaires de Liege and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Complex services, such as healthcare, struggle to realize the benefits of value co-creation due to the substantial challenges of managing such services over the long-term. Key to overcoming these challenges to value co-creation is a profound understanding of dialogue (i.e., 'quality of discourse' facilitating shared meaning) during service interactions. Contributing to an emerging literature, we undertake a longitudinal, ethnographic study to assess dialogue between professionals and patients through the lens of dialogic engagement (i.e., iterative mutual learning processes that bring about action through dialogue). We develop and empirically support six dialogic co-creation and co-destruction mechanisms that impact on the resolution of tensions and integration of knowledge resources between service providers and consumers. We reveal the multidimensional and dynamic nature of value created or destroyed through these mechanisms in dialogue over time. Taking healthcare as an exemplar, we offer a research agenda for developing our understanding of DE in complex services. [ABSTRACT FROM AUTHOR]

Philosophical dialogue is an effective way to deepen one’s thoughts, but it is not easy to practice it because humans have emotions. We proposed the use of a robot in practicing philosophical dialogue and experimentally investigated how philosophical dialogue with a robot differs from philosophical dialogue with a human. The results of the experiment showed that (1) participants talking to a human spend more time answering than those talking to a robot, (2) the increase in the time participants spend answering comes from the increase in the time participants spend speaking and is not influenced by reaction latency and pause time so much, (3) the reason for the increase in the time spent speaking is that some participants who talked to a human were likely to choose their words so as not to make the interlocutor uncomfortable and rephrased their thoughts so that they were easier for the interlocutor to understand, and some participants who talked to a robot might have thought that the robot would not be concerned even if they give a brief answer, and finally (4) it seems there is no significant difference in the depth of thought between participants talking to a human and participants talking to a robot. These results suggest the effectiveness of using robots for philosophical dialogue, in particular for people who feel nervous about talking to others.

Currently, task-oriented dialogue systems that perform specific tasks based on dialogue are widely used. Moreover, research and development of non-task-oriented dialogue systems are also actively conducted. One of the problems with these systems is that it is difficult to switch topics naturally. In this study, we focus on interview dialogue systems. In an interview dialogue, the dialogue system can take the initiative as an interviewer. The main task of an interview dialogue system is to obtain information about the interviewee via dialogue and to assist this individual in understanding his or her personality and strengths. In order to accomplish this task, the system needs to be flexible and appropriate for detecting topic switching and topic breaks. Given that topic switching tends to be more ambiguous in interview dialogues than in task-oriented dialogues, existing topic modeling methods that determine topic breaks based only on relationships and similarities between words are likely to fail. In this study, we propose a method for detecting topic breaks in dialogue to achieve flexible topic switching in interview dialogue systems. The proposed method is based on multi-task learning neural network that uses embedded representations of sentences to understand the context of the text and utilizes the intention of an utterance as a feature. In multi-task learning, not only topic breaks but also the intention associated with the utterance and the speaker are targets of prediction. The results of our evaluation experiments show that using utterance intentions as features improves the accuracy of topic separation estimation compared to the baseline model.

• Proposed an encoder–decoder for joint dialog act segmentation and recognition. • Used an attentional mechanism to integrate contextual information in dialogs. • Showed that exploiting context improves recognition accuracy. • The proposed model outperforms conventional models significantly. A dialog act represents the communicative function of an utterance in a conversation, and thus provides informative cues for understanding, managing, and generating dialog. While most spoken dialog systems process user input and system output at the turn level, a single turn can consist of multiple dialog acts in human conversations. Therefore, segmenting turn-level tokens into a meaningful dialog act unit is just as important as recognizing the dialog act. Towards joint segmentation and recognition of dialog acts, we propose an encoder–decoder model featuring joint coding and incorporate contextual information by means of an attentional mechanism. The proposed encoder–decoder outperforms other models in segmentation, and the application of attentions significantly reduces recognition error rates. By combining the encoder–decoder model with contextual attention, we achieve state-of-the-art performance in the joint evaluation of dialog act segmentation and recognition. [ABSTRACT FROM AUTHOR]

This study investigates participation problems in teams with mixed language proficiencies. Utilising an in-depth single case study approach and drawing on interactional data and interviews, it explores participation in team meetings. It takes a positioning theory approach, and analyses how the least proficient speaker is subtly positioned in various ways: as silent, different, difficult and incompetent. It argues that these positionings contribute to the marginalisation of his contributions in team meetings and in effectively silencing him and that this occurred through interactional patterns in which his contributions were a) ignored, b) dismissed outright and c) treated with only token interest. The paper ends by considering the range of factors, both interactional and attitudinal, that seem to have contributed to this silencing, including cultural stereotypes that seem to influence the dynamics of the interactions. • Less-proficient team members are marginalised in intercultural teams. • They are positioned as difficult, different & incompetent independent of expertise. • Contributions are silenced interactionally, excluding them from influencing tasks. • Repetition perpetuates positionings and prevents repositioning. [ABSTRACT FROM AUTHOR]

The use of WhatsApp as a means of communication is widespread amongst today's youth, many of whom spent hours in virtual space, in particular during the evenings and nighttime in the privacy of their own homes. This article seeks to contribute to the discussion of the dialogical language and "conversations" conducted in virtual-space encounters and the way in which young people perceive this space, its affect on them, and their interrelations within it. It presents the findings of a study based on a community of philosophical inquiry in which young adults students discussed the "I" and "Thou" (the other) and the interaction between them in a WhatsApp community. The results evince that the youth related to the virtual space in very similar fashion to Buber's "I-Thou" concept, the language they employed to describe what happened in it enabling an expansion of the conceptualization and research language to an "I-Space-Thou" model. [ABSTRACT FROM AUTHOR]

Viewing dialogue management as a reinforcement learning task enables a system to learn to act optimally by maximising a reward function. This reward function is designed to induce the system behaviour required for the target application and for goal-oriented applications, this usually means fulfilling the user’s goal as efficiently as possible. However, in real-world spoken dialogue system applications, the reward is hard to measure because the user’s goal is frequently known only to the user. Of course, the system can ask the user if the goal has been satisfied but this can be intrusive. Furthermore, in practice, the accuracy of the user’s response has been found to be highly variable. This paper presents two approaches to tackling this problem. Firstly, a recurrent neural network is utilised as a task success predictor which is pre-trained from off-line data to estimate task success during subsequent on-line dialogue policy learning. Secondly, an on-line learning framework is described whereby a dialogue policy is jointly trained alongside a reward function modelled as a Gaussian process with active learning. This Gaussian process operates on a fixed dimension embedding which encodes each varying length dialogue. This dialogue embedding is generated in both a supervised and unsupervised fashion using different variants of a recurrent neural network. The experimental results demonstrate the effectiveness of both off-line and on-line methods. These methods enable practical on-line training of dialogue policies in real-world applications. [ABSTRACT FROM AUTHOR]

Though never comfortable with the term competency, I reluctantly accepted it as a way to teach noticing and attending to relational process (NARP) for couple and family social work practice. Didactic training combined with experiential activities of video-recorded role-play and audio-recorded feedback comprised my strategy. Although this teaching approach seemed successful, I struggled theoretically with competency, and my ideas morphed through ongoing practice, teaching, and research experiences. I eventually concluded that NARP was not a single competency but was best characterized as "ways of being" as a social work practitioner; I consequently worked to become more articulate and transparent in the classroom. Informed by constructivist pedagogy and relational ontology, I use the following reflection to immerse myself in an ongoing dialogue to understand how my own and my students' unique discourses shape NARP. [ABSTRACT FROM AUTHOR]

This article introduces a new methodology to enhance an existing traditional Spoken Dialogue System (SDS) with optimal turn-taking capabilities in order to increase dialogue efficiency. A new approach for transforming the traditional dialogue architecture into an incremental one at a low cost is presented: a new turn-taking decision module called the Scheduler is inserted between the Client and the Service. It is responsible for handling turn-taking decisions. Then, a User Simulator which is able to interact with the system using this new architecture has been implemented and used to train a new Reinforcement Learning turn-taking strategy. Compared to a non-incremental and a handcrafted incremental baselines, it is shown to perform better in simulation and in a real live experiment. [ABSTRACT FROM AUTHOR]

Dialogue act recognition is an important component of a large number of natural language processing pipelines. Many research works have been carried out in this area, but relatively few investigate deep neural networks and word embeddings. This is surprising, given that both of these techniques have proven exceptionally good in most other language-related domains. We propose in this work a new deep neural network that explores recurrent models to capture word sequences within sentences, and further study the impact of pretrained word embeddings. We validate this model on three languages: English, French and Czech. The performance of the proposed approach is consistent across these languages and it is comparable to the state-of-the-art results in English. More importantly, we confirm that deep neural networks indeed outperform a Maximum Entropy classifier, which was expected. However, and this is more surprising, we also found that standard word2vec embeddings do not seem to bring valuable information for this task and the proposed model, whatever the size of the training corpus is. We thus further analyse the resulting embeddings and conclude that a possible explanation may be related to the mismatch between the type of lexical-semantic information captured by the word2vec embeddings, and the kind of relations between words that is the most useful for the dialogue act recognition task. [ABSTRACT FROM AUTHOR]

Purpose: The purpose of this paper is to explore the aspects of leadership within the military that set it apart from traditional leadership in the business environment. Design/method/approach: The approach employed in this paper is a form of critical dialogue analysis. The dialogue is derived from reports from the US and UK military and juxtaposed against the dialogue from the movie "A few Good Men". Findings: The analysis leads to a general model of trust pertaining to issues of leadership in a military setting. Originality/value: This paper addresses a gap in the literature, in which differences of leadership and trust in a military setting have been assumed to be identical with the business environment. [ABSTRACT FROM AUTHOR]

Museum researchers have long acknowledged the importance of dialogue in informal learning, particularly for open-ended exploratory exhibits. Novel interaction techniques like full-body interaction are appealing for these exploratory exhibits, but designers have not had a metric for determining how their designs are supporting productive learning talk. Moreover, with the incorporation of digital technologies into museums, researchers and designers now have the opportunity for in situ A/B testing of multiple exhibit designs not previously possible with traditionally constructed exhibits, which once installed were difficult and expensive to iterate. Here we present a method called Scoring Qualitative Informal Learning Dialogue (SQuILD) for quantifying idiosyncratic social learning talk, in order to conduct in situ testing of group learning at interactive exhibits. We demonstrate how the method was applied to a 2 × 2 experiment varying the means of control (full-body vs. handheld tablet controller) and the distribution of control (single-user-input vs. multi-user-input) of an interactive data map exhibit. Though pilot testing in the lab predicted that full-body and multi-input designs would best support learning talk, analysis of dialogue from 119 groups' interactions revealed surprising nuances in the affordances of each. Implications for embodied interaction design are discussed. [ABSTRACT FROM AUTHOR]

ObjectivesAnaphora recognition is a process to identify exactly which noun has been used previously and relates to a pronoun that is included in a specific sentence later. Therefore, anaphora recognition is an essential element of a dialogue agent system. In the current study, all the merits of rule-based, machine learning-based, semantic-based anaphora recognition systems were combined to design and realize a new hybrid-type anaphora recognition system with an optimum capacity.MethodsAnaphora recognition rules were encoded on the basis of the internal traits of referred expressions and adjacent contexts to realize a rule-based system and to serve as a baseline. A semantic database, related to predicate instances of sentences including referred expressions, was constructed to identify semantic co-relationships between the referent candidates (to which semantic tags were attached) and the semantic information of predicates. This approach would upgrade the anaphora recognition system by reducing the number of referent candidates. Additionally, to realize a machine learning-based system, an anaphora recognition model was developed on the basis of training data, which indicated referred expressions and referents. The three methods were further combined to develop a new single hybrid-based anaphora recognition system.ResultsThe precision rate of the rule-based systems was 54.9%. However, the precision rate of the hybrid-based system was 63.7%, proving it to be the most efficient method.ConclusionsThe hybrid-based method, developed by the combination of rule-based and machine learning-based methods, represents a new system with enhanced functional capabilities as compared to other pre-existing individual methods.

In this article, and after sketching the state of the art on semantic representations, we have deduced that an important technological lock that hinders the H/M Spoken Dialogue System (SDS) is the adoption of the attribute/value schemes as a semantic representation. Besides, current systems, using such a representation (form of a formula), constrain too much their users to behave according to the model set by the designer. Thus, in order to overcome all the limitations of the modeling by a pre-established form, we decided to choose the use of semantic frames as a representation of the meanings of Arabic user statements. In this perspective, the choice of the FrameNet paradigm as a high level semantic representation seems appropriate. In addition to the robustness of the paradigm, FrameNet has the advantage of producing standard annotations that are easy to share and compare within the scientific community. To represent more complex user rounds, we propose to use, in the understanding module, a semantic model that is more flexible than that of a predefined form. [ABSTRACT FROM AUTHOR]

Alzheimer's disease (AD) is an increasingly prevalent cognitive disorder in which memory, language, and executive function deteriorate, usually in that order. There is a growing need to support individuals with AD and other forms of dementia in their daily lives, and our goal is to do so through speech-based interaction. Given that 33% of conversations with people with middle-stage AD involve a breakdown in communication, it is vital that automated dialogue systems be able to identify those breakdowns and, if possible, avoid them. In this article, we discuss several linguistic features that are verbal indicators of confusion in AD (including vocabulary richness, parse tree structures, and acoustic cues) and apply several machine learning algorithms to identify dialogue-relevant confusion from speech with up to 82% accuracy. We also learn dialogue strategies to avoid confusion in the first place, which is accomplished using a partially observable Markov decision process and which obtains accuracies (up to 96.1%) that are significantly higher than several baselines. This work represents a major step towards automated dialogue systems for individuals with dementia. [ABSTRACT FROM AUTHOR]

Referring to visually perceivable objects is a very common occurrence in everyday language use. In order to produce expressions that refer, the speaker needs to be able to pick out visual properties that the referred object has and determine the words that name those properties, such that the expression can direct a listener's attention to the intended object. The speaker can aid the listener by looking in the direction of the object and by providing a pointing gesture to indicate it. In order to resolve the reference, the listener has a difficult job to do: simultaneously use all of the linguistic and non-linguistic information; the words of the referring expression that denote properties of the object, such as its colour or shape, need to already be known, and the non-linguistic gaze direction and pointing gesture of the speaker need to be incorporated. Crucially, the listener does not wait until the end of the referring expression before she begins to resolve it; rather, she is interpreting it as it unfolds. A model that resolves referring expressions as the listener must be able to do all of these things. In this paper, we present such a generative model of reference resolution. We explain our model and show empirically through a series of experiments that the model can work incrementally (i.e., word for word) as referring expressions unfold, can incorporate multimodal information such as gaze and pointing gestures in two ways, can learn a grounded meaning of words in the referring expression, can incorporate contextual (i.e., saliency) information, and is robust to noisy input such as automatic speech recognition transcriptions, as well as uncertainty in the representation of the candidate objects. [ABSTRACT FROM AUTHOR]

As knowledge-grounded dialogue systems are attracting intense research interest, technology that facilitates reference to various types of external knowledge as dialogue-system data is also developing. A knowledge-grounded dialogue system must be capable of (1) accurately interpret the conversation content, (2) determining whether external knowledge should be referenced for the current turn, (3) identifying the external knowledge to be referenced, and (4) generating a response. If the referenced external knowledge is in the form of a question–answer pair, this pair is closely related to the domain to which the external knowledge belongs and the entity to which it pertains. This study proposes a knowledge-grounded dialogue system that predicts the domain and entity associated with a question–answer pair referenced by the dialogue system; the system then leverages this information to effectively implement the above three external-knowledge-based capabilities. As the dialogue data associated with external knowledge is diverse, adaptation to the slots and entities of different dialogue datasets is challenging. In this study, the Triple Copy (TripPy) model, which is one of the leading benchmark models for dialogue-state tracking with the Multi-domain Wizard-of-Oz (MultiWOZ) dataset, is further developed to adapt to DSTC10 data for the external knowledge-based dialogue system; hence, dialogue content is effectively interpreted. The developed knowledge-grounded dialogue system incorporates knowledge-seeking turn detection, knowledge selection, and knowledge-grounded response generation models. The model, in DSTC10 dataset, achieves 15.49%p, 12.54%p, and 36.45%p improvements over the baseline, the state-of-the-art model for the previous version of the data (DSTC9 dataset), in terms of the F1 score, Recall@1 score, and BLEU-1 score, respectively. Moreover, in a dialogue-state tracking task for DSTC10 dataset, a 15.41%p improvement in joint goal accuracy score is achieved compared to the TripPy model. • A goal-oriented dialogue system using external knowledge to respond to a user is proposed. • Adapting a dialogue-state tracking model to targeting datasets to well understand a conversation. • The advanced knowledge-grounded dialogue system using the extracted entities and tracked domain. [ABSTRACT FROM AUTHOR]

Task-oriented dialogue systems need to take appropriate actions not only for clear user requests but also for ambiguous and vague ones. In this study, "ambiguous" denotes that although users have potential requests, they failed to clearly define and verbalize their content and conditions which can be associated with system actions. For such ambiguous requests, taking reflective actions is one plausible choice for such systems. In our study, "reflective" denotes taking actions that satisfy user requests before the users themselves clarify their demands. We constructed such a reflective dialogue agent by collecting a corpus that includes pairs of ambiguous user requests and corresponding reflective system actions on sightseeing navigation with a smartphone. Since annotating every possible combination of user requests and system actions is impossible, this study built a corpus where one reflective action is annotated to one ambiguous user request. To train an action selection model on such incomplete training data in which only one action is associated with a request, we applied the positive/unlabeled (PU) learning method, which assumes that only part of the data is labeled with positive examples. In addition, we enhanced the action selection by extracting and distilling knowledge that corresponds to causality from the training data using a causality detection model. The experimental results show that both the PU learning method and the causality detection model improved the performances of the reflective action selection compared to the conventional positive/negative (PN) learning method. • We built a reflective action selection system given ambiguous user requests. • We collected a dialogue corpus that bridges ambiguous requests to reflective actions. • We explored whether several actions can be reflective for one request on test data. • We applied the PU learning method to deal with partially annotated training data. • We distilled causality knowledge to utilize in the reflective action selection. [ABSTRACT FROM AUTHOR]

We define epistemic order as the way in which the exchange and development of knowledge takes place in the classroom, breaking this down into a system of three components: epistemic initiative relating to who sets the agenda in classroom dialogue, and how, epistemic appraisal relating to who judges contributions to classroom dialogue, and how, and epistemic framing relating to the terms in which development and exchange of knowledge are represented, particularly in reflexive talk. These components are operationalised in terms of various types of structural and semantic analysis of dialogue. It is shown that a lesson segment displays a multi-layered epistemic order differing from that of conventional classroom recitation. [ABSTRACT FROM AUTHOR]

Observing others in conversation is a common format for comprehending language, yet little work has been done to understand dialog comprehension. We tested whether overhearers use addressee backchannels as predictive cues for how to integrate information across speaker turns during comprehension of spontaneously produced collaborative narration. In Experiment 1, words that followed specific backchannels (e.g., really, oh) were recognized more slowly than words that followed either generic backchannels (e.g., uh huh, mhm) or pauses. In Experiment 2, we found that when the turn after the backchannel was a continuation of the narrative, specific backchannels prompted the fastest verification of prior information. When the turn after was an elaboration, they prompted the slowest, indicating that overhearers took specific backchannels as cues to integrate preceding talk with subsequent talk. These findings demonstrate that overhearers capitalize on the predictive relationship between backchannels and the development of speakers' talk, coordinating information across conversational roles. [ABSTRACT FROM AUTHOR]

This study develops a communicative model of collaboration failure to address one of the key challenges of collaboration theory and practice: the discrepancy between the promise of collaboration and the reality of persistent failure. A theoretical framework is developed based on notions of dialogue, discourse, and coorientation, which informs three key aspects of collaboration: knowledge production, shared identity, and collective agency. This theoretical framework is then combined with analytic themes from an empirical case study of a failed civil society collaboration. Themes of communication practice that constitute collaborative failure are detailed, while also contrasting these with alternative practices that can enable more successful collaboration. Further implications are discussed, specifically in terms of rethinking common collaboration dualisms of structure/process and talk/action. [ABSTRACT FROM AUTHOR]

This article is concerned with optimizing human-machine turn-taking. In particular, the article covers an in-depth analysis of the timings when users respond to system query in spoken dialog systems. The goal of this work is to obtain a broad understanding of such timing patterns independent of dialog system type and dialog state context. Therefore, the analysis was based on a large volume of data both from a number of deployed spoken dialog system and an experimental study. The data from the experimental study showed that too short timeout settings can cause the system to interrupt a user and thus cause turn-taking problems. Next, the response timing patterns both during a system prompt as well as after prompt-end were analyzed for a number of different question types. It is shown that user responses while the system is playing a prompt (aka 'barge-in') tend to occur in the range of 10-25 % of all user responses, where the exact percentage of barge-in is context-dependent. It was also found that the timing of user responses after a system finishes speaking always follows the same uni-modal pattern independent of system domain and question type. This pattern can be modeled with a rational distribution. Based on these findings, a probabilistic response time model is presented, that allows calculating the likelihood of a user response at any time in a system. This response timing model can be used for multiple purposes, among them timeout setting optimization. [ABSTRACT FROM AUTHOR]

In spoken dialog systems, dialog state tracking refers to the task of correctly inferring the user's goal at a given turn, given all of the dialog history up to that turn. The Dialog State Tracking Challenge is a research community challenge task that has run for three rounds. The challenge has given rise to a host of new methods for dialog state tracking and also to deeper understanding about the problem itself, including methods for evaluation. [ABSTRACT FROM AUTHOR]

The idea of a just and lasting peace seems to be unachievable when people are suffering a civil or international war, a violent revolutionary process or a dictatorship, or when they are immersed in a long period of structural injustice. However, during the last decades there have been a number of attempted transitions from violent situations to a just peace through peace-making processes and talks, followed by reconciliaation, some of which have been remarkably successful, whilst others are still open (e.g. El Salvador, South Africa, Northern Ireland, the Israel-Palestine conflict, among others). These processes, and their extremely technical complexity, are the object of analysis in the Political Science discipline; yet should they not also be the object of study in other disciplines such as Anthropology and Ethics? Opposing issues such as the possibility of self-aggression yet of radical transformation, the restoration of dialogue, the recognition of others as being equal, and the achievement of a just peace for all are at the centre of being human. In this regard, the work of Ignacio Ellacuría in the context of the civil war in El Salvador (1981-1991) is extremely interesting. This paper presents an analysis of his theoretical contribution structured as follows: the analysis of the concepts of dialogue, social dialogue and political dialogue; the context of violence in which he constructs his ethical and anthropological thought; his defence that social dialogue involves the affirmation of the possibility of moral transformation as an anthropological feature of human beings; the possibility to contest any of the multiple objections to dialogue with reason; his view that the reconciliation process goes from ideological monism to dualism, and from there to a wider and more well-defined pluralism; his conviction that political dialogue makes sense as the fulfilment of social dialogue so it can never replace the latter without losing legitimacy; his definition of the mediator as a key figure, essential to overcoming ideological dichotomy; his objective of achieving a just and lasting peace. The Jesuit Ignacio Ellacuría was a theologian and a philosopher, a sharp political analyst, and a mediator in the Salvadorean conflict, where 75,000 human lives were lost in 10 years. He applied the "Philosophy of Reality" of his maestro Xavier Zubiri to the postcolonial and revolutionary reality of Central America from the 1960s to the 1980s. He was killed by the army in 1989. The 25th anniversary of his death (1989-2014) is an appropriate moment to undertake an anthropological and ethical analysis of his contribution to political and social peace processes. [ABSTRACT FROM AUTHOR]