Dialogue Management Research Articles

Traditional and new forms of patient expertise: Together with doctors or versus them?

The article outlines and discusses two opposite perspectives of medical practitioners in relation to patients’ attempts to form lay medical expertise. The author conceptualizes these two approaches as thought styles and the corresponding communities of doctors as “thought collectives” (L.Fleck). The two thought styles are distinguished based on four characteristics: attitudes towards interaction with patients, particular ways of making sense of alternative forms of expertise, strategies for resolving conflicts with patients, and outlooks on professional knowledge and its place outside the medical community. One of the thought collectives — monopolization collective — is committed to a paternalistic model of medical care, and the corresponding style of thought does not allow noticing and acknowledging patient expertise. Patients’ attempts at joining the professional medical rationale are resisted by these medical professionals, and in situations of disagreement or conflicts doctors use a closed strategy for their resolution. Physicians from monopolization collective are skeptical about translating medical knowledge outside of practitioners. In contrast, doctors who adopt a co-construction thought style follow the logic of partnership when interacting with patients. Such a style allows medical practitioners to recognize the value in patient expertise and include it in a dialogue and disease management. When conflicts and differences of opinions arise, doctors follow open and intermediate strategies for their resolution, which implies additional efforts to build communication with patients. At the same time, there is a clear demand from this collective for the creation of high-quality popular science and educational materials for patients. High quality translation of medical knowledge is seen as a necessary means of including patients in productive co-production of expertise.

Hierarchical Reinforcement Learning With Guidance for Multi-Domain Dialogue Policy

Achieving high performance in a multi-domain dialogue system with low computation is undoubtedly challenging. Previous works applying an end-to-end approach have been very successful. However, the computational cost remains a major issue since the large-sized language model using GPT-2 is required. Meanwhile, the optimization for individual components in the dialogue system has not shown promising result, especially for the component of dialogue management due to the complexity of multi-domain state and action representation. To cope with these issues, this article presents an efficient guidance learning where the imitation learning and the hierarchical reinforcement learning (HRL) with human-in-the-loop are performed to achieve high performance via an inexpensive dialogue agent. The behavior cloning with auxiliary tasks is exploited to identify the important features in latent representation. In particular, the proposed HRL is designed to treat each goal of a dialogue with the corresponding sub-policy so as to provide efficient dialogue policy learning by utilizing the guidance from human through action pruning and action evaluation, as well as the reward obtained from the interaction with the simulated user in the environment. Experimental results on ConvLab-2 framework show that the proposed method achieves state-of-the-art performance in dialogue policy optimization and outperforms the GPT-2 based solutions in end-to-end system evaluation.

複数の目標を持つ誘導対話のための話題管理部の階層型制御アーキテクチャ

In a dialogue system, when there is a dialogue goal for the dialogue system that is independent of the user's dialogue goal, a dialogue that tries to reach a topic for that goal while coming to terms with the user (i.e., target-guided dialogue) has the potential to be useful for applications such as advertising and recommendation. However, in the case of multiple target-guided strategies, the complexity and multiplicity of designs for switching between target-guided strategies is a challenge. In this paper, we propose the hierarchical structure for dialogue management method, which alleviates the complexity and multiplicity of designing such target-guided strategy switching. It is found that the proposed method automatically complements the topic transitions of skipping, modifying, and sliding, and enhances naturalness of transitions and interest in guided topics.

Characterizing the Response Space of Questions: data and theory

The main aim of this paper is to provide a characterization of the response space for questions using a taxonomy grounded in a dialogical formal semantics. As a starting point we take the typology for responses in the form of questions provided in \cite{lupginz-jlm}. This work develops a wide coverage taxonomy for question/question sequences observable in corpora including the BNC, CHILDES, and BEE, as well as formal modeling of all the postulated classes. Our aim is to extend this work to cover \emph{all} responses to questions. We present the extended typology of responses to questions based on a corpus studies of BNC, BEE, Maptask and CornellMovie with include 506, 262, 467, and 678 question/response pairs respectively. We compare the data for English with data from Polish using the Spokes corpus (694 question/response pairs). We discuss annotation reliability and disagreement analysis. We sketch how each class can be formalized using a dialogical semantics appropriate for dialogue management.

Artificial intelligence in virtual standardized patients: Combining natural language understanding and rule based dialogue management to improve conversational fidelity

Introduction Advances in natural language understanding have facilitated the development of Virtual Standardized Patients (VSPs) that may soon rival human patients in conversational ability. We describe herein the development of an artificial intelligence (AI) system for VSPs enabling students to practice their history taking skills. Methods Our system consists of (1) Automated Speech Recognition (ASR), (2) hybrid AI for question identification, (3) classifier to choose between the two systems, and (4) automated speech generation. We analyzed the accuracy of the ASR, the two AI systems, the classifier, and student feedback with 620 first year medical students from 2018 to 2021. Results System accuracy improved from ∼75% in 2018 to ∼90% in 2021 as refinements in algorithms and additional training data were utilized. Student feedback was positive, and most students felt that practicing with the VSPs was a worthwhile experience. Conclusion We have developed a novel hybrid dialogue system that enables artificially intelligent VSPs to correctly answer student questions at levels comparable with human SPs. This system allows trainees to practice and refine their history-taking skills before interacting with human patients.

Adapting conversational strategies in information-giving human-agent interaction.

In this work, we focus on human-agent interaction where the role of the socially interactive agent is to optimize the amount of information to give to a user. In particular, we developed a dialog manager able to adapt the agent's conversational strategies to the preferences of the user it is interacting with to maximize the user's engagement during the interaction. For this purpose, we train an agent in interaction with a user using the reinforcement learning approach. The engagement of the user is measured using their non-verbal behaviors and turn-taking status. This measured engagement is used in the reward function, which balances the task of the agent (giving information) and its social goal (maintaining the user highly engaged). Agent's dialog acts may have different impact on the user's engagement depending on several factors, such as their personality, interest in the discussion topic, and attitude toward the agent. A subjective study was conducted with 120 participants to measure how third-party observers can perceive the adaptation of our dialog model. The results show that adapting the agent's conversational strategies has an influence on the participants' perception.

A study of interactive robot architecture through the practical implementation of conversational android.

This study shows an autonomous android robot that can have a natural daily dialogue with humans. The dialogue system for daily dialogue is different from a task-oriented dialogue system in that it is not given a clear purpose or the necessary information. That is, it needs to generate an utterance in a situation where there is no clear request from humans. Therefore, to continue a dialogue with a consistent content, it is necessary to essentially change the design policy of dialogue management compared with the existing dialogue system. The purpose of our study is to constructively find out the dialogue system architecture for realizing daily dialogue through implementing an autonomous dialogue robot capable of daily natural dialogue. We defined the android’s desire necessary for daily dialogue and the dialogue management system in which the android changes its internal (mental) states in accordance to the desire and partner’s behavior and chooses a dialogue topic suitable for the current situation. The developed android could continue daily dialogue for about 10 min in the scene where the robot and partner met for the first time in the experiment. Moreover, a multimodal Turing test has shown that half of the participants had felt that the android was remotely controlled to some degree, that is, the android’s behavior was humanlike. This result suggests that the system construction method assumed in this study is an effective approach to realize daily dialogue, and the study discusses the system architecture for daily dialogue.

An Open Source Virtual Reality Training Framework for the Criminal Justice System

We developed an open-source training framework to practice conversation skills in a controlled and immersive virtual reality (VR) environment. Virtual characters with different biographies were developed with which a conversation using natural language is possible. The virtual characters integrate a dialog management system (ChatScript) to provide different biographical memories. Natural language processing for the German language is integrated by using Kaldi, an open-source speech recognition toolkit. As the framework allows for interchangeable content there are many different possible application cases to apply within the criminal justice system. The VR frameworks code is available under an open-source license. In this article, an overview of the framework’s functionality is given as well as an outlook on possible areas of application. Statements about user acceptance and usability cannot yet be made, as relevant data have first to be gathered through a concrete application case.

Dialogue Management in Conversational Systems: A Review of Approaches, Challenges, and Opportunities

Attracted by their easy-to-use interfaces and captivating benefits, conversational systems have been widely embraced by many individuals and organizations as side-by-side digital co-workers. They enable the understanding of user needs, expressed in natural language, and on fulfilling such needs by invoking the appropriate backend services (e.g., APIs). Controlling the conversation flow, known as <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Dialogue Management</i> , is one of the essential tasks in conversational systems and the key to its success and adoption as well. Nevertheless, designing scalable and robust dialogue management techniques to effectively support intelligent conversations remains a deeply challenging problem. This article studies dialogue management from an in-depth design perspective. We discuss the state-of-the-art approaches, identify their recent advances and challenges, and provide an outlook on future research directions. Thus, we contribute to guiding researchers and practitioners in selecting the appropriate dialogue management approach aligned with their objectives, among the variety of approaches proposed so far.

FINANCIAL PROVISION MANAGEMENT FOR THE ENVIRONMENTAL PROTECTION OF THE TRANSBORDER REGION

The aim of the article is to conduct a comparative analysis of the Volyn region (Ukraine) and Lublin Voivodeship (Republic of Poland) environmental protection financing. As well as to determine the level of partnership participation in the financing of the environmental measures within the cross-border region. The system of financial support for environmental protection in the cross-border region is explored in the article. The content of the transboundary natural resources management mechanism is considered. The main sources of funding for environmental measures in the Volyn region and Lublin Voivodeship are identified. A comparative analysis of the dynamics and structure of current costs spent and capital investments for the environmental protection of the Volyn region and Lublin Voivodeship is carried out. It is determined that the structure of environmental protection expenditures within the cross-border region is dominated by costs spent on the protection of water resources, air, waste management, and biodiversity conservation. Linear trend models were built for the amount of funding in Volyn Oblast and Lublin Voivodeship using the trend forecasting methodology. The forecast of the dynamics of the environmental protection financing within the cross-border region for 2022-2023 is determined by the method of extrapolation.Determination of the degree of cooperation in the field of transboundary natural resources management and the division of responsibilities between its key players are important aspects of the financial model’s effectiveness. The degree of cooperation in the field of transboundary natural resources management, including independent control, dialogue, cooperation, coordination mechanisms and joint management is revealed in the article. It has been proved that the cross-border relations within the Volyn region and the Lublin Voivodeship are at the level of cooperation. But it is noted that efficiency will be achieved only at the level of joint cross-border management in this area. The main tools of integrated management of financial support for environmental protection are identified. The distribution of partnership participation in the financing of environmental measures is proposed depending on their location in the border areas of Volyn region and Lublin Voivodeship.

Graph-based representations of clarification strategies supporting automatic dialogue management

This paper aims at presenting a dialogue-oriented approach to the construction of a graph knowledge base (KB) supporting task-oriented human-machine interactions. In particular, we focus on different pragmatic scenarios, facing the Common Ground issue and arguing that knowledge bases (in the form of graphs) are needed to make a clarification and recover pieces of information when inconsistencies occur during the communicative exchange.The main contributions of this work are: 1) a flexible dialog system architecture designed to be plugged into existing service infrastructures, 2) a graph-based knowledge representation protocol to manage both dialog domain and dialog management, 3) a detailed investigation of clarification requests forms with respect to their functions. After a brief introduction (see Section 1), we present: the theoretical underpinnings of the paper and the background work (see Section 2) our system architecture (see Sections 3 and 4) and the clarification requests (CRs) issue (see Section 5); our CRs classification, and some examples in context (see Section 6).

DAISY: An Implementation of Five Core Principles for Transparent and Accountable Conversational AI

We present a detailed implementation of five core principles for transparent and acccountable conversational AI, namely interpretability, inherent capability to explain, independent data, interactive learning, and inquisitiveness. This implementation is a dialogue manager called DAISY that serves as the core part of a conversational agent. We show how DAISY-based agents are trained with human-machine interaction, a process that also involves suggestions for generalization from the agent itself. Moreover, these agents are capable to provide a concise and clear explanation of the actions required to reach a conclusion. Deep neural networks (DNNs) are currently the de facto standard in conversational AI. We therefore formulate a comparison between DAISY-based agents and two methods that use DNNs, on two popular data sets involving multi-domain task-oriented dialogue. Specifically, we provide quantitative results related to entity retrieval and qualitative results in terms of the type of errors that may occur. The results show that DAISY-based agents achieve superior precision at the price of lower recall, an outcome that might be preferable in task-oriented settings. Ultimately, and especially in view of their high degree of interpretability, DAISY-based agents are a fundamentally different alternative to the currently popular DNN-based methods.

Examining AI Methods for Micro-Coaching Dialogs.

Conversational interaction, for example through chatbots, is well-suited to enable automated health coaching tools to support self-management and prevention of chronic diseases. However, chatbots in health are predominantly scripted or rule-based, which can result in a stagnant and repetitive user experience in contrast with more dynamic, data-driven chatbots in other domains. Consequently, little is known about the tradeoffs of pursuing data-driven approaches for health chatbots. We examined multiple artificial intelligence (AI) approaches to enable micro-coaching dialogs in nutrition - brief coaching conversations related to specific meals, to support achievement of nutrition goals - and compared, reinforcement learning (RL), rule-based, and scripted approaches for dialog management. While the data-driven RL chatbot succeeded in shorter, more efficient dialogs, surprisingly the simplest, scripted chatbot was rated as higher quality, despite not fulfilling its task as consistently. These results highlight tensions between scripted and more complex, data-driven approaches for chatbots in health.

Obituary: Martin Kay

Obituary: Martin Kay

An Effective Artificial Intelligence-Enabled Error Detection and Accuracy Estimation Technique for English Speech Recognition System

Error detection and accuracy estimation in automated speech recognition (ASR) systems act a vital part in the design of human-computer spoken dialogue systems, as recognition error can hamper accurate systems in understanding the end user intentions. The major aim is to identify the errors in an utterance, and therefore, the dialogue manager can provide proper clarifications to the user. Therefore, the design of accurate error detection and accuracy determination techniques becomes essential in the ASR system. With this motivation, this paper presents a novel artificial intelligence-enabled accuracy estimation and error detection technique for the English speech recognition system (AIEDAE-ESRS). The goal of the AIEDAE-ESRS technique is to perform three actions such as confidence estimation, out-of-vocabulary (OOV) word identification, and error type categorization. In addition, the AIEDAE-ESRS technique performs different levels of preprocessing including sampling of input speech signal, bandpass filtering, and noise removal. Besides, a new deep neural network with hidden Markov model- (DNN-HMM-) based speech recognition technology is designed, which also aims to estimate the accuracy and error. Finally, the hyperparameters of the DNN-HMM model can be optimally chosen by the use of flower pollination algorithm (FPA) and thereby accomplished improved recognition performance. In order to demonstrate the better performance of the AIEDAE-ESRS technique, a series of simulations were conducted and the results are examined under varying aspects. English voice recognition system’s accuracy estimation and error detection were made possible using artificial intelligence (AIEDAE-ESRS). There are three steps in the AIEDAE-ESRS method: confidence estimation; identifying out-of-vocabulary words (OOV); and categorizing mistake types. The simulation results reported the enhanced performance of the AIEDAE-ESRS methodology over current advanced approaches. Our AIEDAE-ESRS methodology outperforms existing methodologies by a factor of ten. The simulation results demonstrated that the AIEDAE-ESRS methodology outperformed previous approaches in terms of efficiency. The improved experimental results indicated that the AIEDAE-ESRS technique produced superior results across a variety of measures.

Dialog management system based on user persona

Natural language processing (NLP) components are responsible for analysing and contextualising human-like discussions between chatbots or any voice browser or with any live users are known as dialogue management systems (DMS). Dialog management systems, also known as plug-ins, allow the chatbot to complete this functionality with ease. The dialogue management system features a module called the agent for dialogue management that allows the DMS to contextualise information and deliver replies. Chatter-bots frequently employ dialogue management systems, such as ChatScript, to regulate the conversation structure based on themes. In the developed application which emulates the behaviour of a DMS, the functionalities like voice assisted navigation, functional keys implementation, language neutral search are implemented. The system is developed by taking into consideration of user experience as the primary factor. The system facilitates physically disabled users to perform all the above mentioned functionalities using voice commands with approx.90% of accuracy.

Fine-Tuning BERT Models for Intent Recognition Using a Frequency Cut-Off Strategy for Domain-Specific Vocabulary Extension

Intent recognition is a key component of any task-oriented conversational system. The intent recognizer can be used first to classify the user’s utterance into one of several predefined classes (intents) that help to understand the user’s current goal. Then, the most adequate response can be provided accordingly. Intent recognizers also often appear as a form of joint models for performing the natural language understanding and dialog management tasks together as a single process, thus simplifying the set of problems that a conversational system must solve. This happens to be especially true for frequently asked question (FAQ) conversational systems. In this work, we first present an exploratory analysis in which different deep learning (DL) models for intent detection and classification were evaluated. In particular, we experimentally compare and analyze conventional recurrent neural networks (RNN) and state-of-the-art transformer models. Our experiments confirmed that best performance is achieved by using transformers. Specifically, best performance was achieved by fine-tuning the so-called BETO model (a Spanish pretrained bidirectional encoder representations from transformers (BERT) model from the Universidad de Chile) in our intent detection task. Then, as the main contribution of the paper, we analyze the effect of inserting unseen domain words to extend the vocabulary of the model as part of the fine-tuning or domain-adaptation process. Particularly, a very simple word frequency cut-off strategy is experimentally shown to be a suitable method for driving the vocabulary learning decisions over unseen words. The results of our analysis show that the proposed method helps to effectively extend the original vocabulary of the pretrained models. We validated our approach with a selection of the corpus acquired with the Hispabot-Covid19 system obtaining satisfactory results.

A Review of Plan-Based Approaches for Dialogue Management

Dialogue systems deliver a more natural mean of communication between humans and machines when compared to traditional systems. Beyond input/output components that understand and generate natural language utterances, the core of a dialogue system is the dialogue manager. The aim of the dialogue manager is to mimic all cognitive aspects related to a natural conversation and it is responsible for identifying the current state of the dialogue and for deciding the next action to be taken by a dialogue system. Artificial intelligence (AI) planning is one of the techniques available in the literature for dialogue management. In a dialogue system, AI planning deals with the action selection problem by treating each utterance as an action and by choosing the actions that get closer to the dialogue goal. This work aims to provide a systematic literature review (SLR) that investigates recent contributions to plan-based dialogue management. This SLR aims at answering research questions concerning: (i) the types of AI planning exploited for dialogue management; (ii) the planning characteristics that justify its adoption in dialogue system; (iii) and, the challenges posed on the development of plan-based dialogue managers. The present SLR was performed by querying four scientific repositories, followed by a manual search on works from the most eminent authors in the field. Further works that were cited by the retrieved papers were also considered for inclusion. Our final corpus is composed of forty works, including only works published since 2014. The results indicate that AI planning is still an emerging strategy for dialogue management. Although AI planning can offer a strong contribution to dialogue systems, especially to those that require predictability, some relevant challenges might still limit its adoption. Our results contributed to discussions in the field and they highlight some research gaps to be addressed in future studies.

Chatbot integrated with machine learning deployed in the cloud and performance evaluation

Recently human-machine digital assistants gained popularity and commonly used in question-and-answer applications and similar consumer-supporting domains. A class of more sophisticated digital assistants employing longer dialogs follow the trend, and there are several commercial platforms supporting their prototyping such as Google DialogFlow, Manychat, Chatfuel, Amazon Lex, etc. This paper explores cloud deployment of chatbots systems and their performance assessment methodologies. The performance measures includes system response delays and natural language processing capabilities. A case study platform supporting so-called deep-logic chatbots with long cycling capabilities is implemented and used for the assessment. To enable human-like conversations with a chatbot, huge training data, complex natural language understanding models are required and need to be adjusted and trained continuously. We explore implementation formats supporting auto scaling, and uninterrupted availability. In particular, we employ an architecture consisting of separate dialog management, authentication, and Natural Language Understanding (NLU) services. Finally, we present a performance evaluation of such loosely coupled chatbot system. Keywords: Cloud Deployment, Natural language understanding, Chatbot, Performance assessment

Is a Wizard-of-Oz Required for Robot-Led Conversation Practice in a Second Language?

The large majority of previous work on human-robot conversations in a second language has been performed with a human wizard-of-Oz. The reasons are that automatic speech recognition of non-native conversational speech is considered to be unreliable and that the dialogue management task of selecting robot utterances that are adequate at a given turn is complex in social conversations. This study therefore investigates if robot-led conversation practice in a second language with pairs of adult learners could potentially be managed by an autonomous robot. We first investigate how correct and understandable transcriptions of second language learner utterances are when made by a state-of-the-art speech recogniser. We find both a relatively high word error rate (41%) and that a substantial share (42%) of the utterances are judged to be incomprehensible or only partially understandable by a human reader. We then evaluate how adequate the robot utterance selection is, when performed manually based on the speech recognition transcriptions or autonomously using (a) predefined sequences of robot utterances, (b) a general state-of-the-art language model that selects utterances based on learner input or the preceding robot utterance, or (c) a custom-made statistical method that is trained on observations of the wizard’s choices in previous conversations. It is shown that adequate or at least acceptable robot utterances are selected by the human wizard in most cases (96%), even though the ASR transcriptions have a high word error rate. Further, the custom-made statistical method performs as well as manual selection of robot utterances based on ASR transcriptions. It was also found that the interaction strategy that the robot employed, which differed regarding how much the robot maintained the initiative in the conversation and if the focus of the conversation was on the robot or the learners, had marginal effects on the word error rate and understandability of the transcriptions but larger effects on the adequateness of the utterance selection. Autonomous robot-led conversations may hence work better with some robot interaction strategies.