Although computers could offer emotional support as well as task support when aiding a user for a complex task, there is little current understanding of how they might do this. Moreover existing demonstrations of emotional support, though promising, only cover a small number of types of support and investigate a limited number of algorithms designed by hand. In this paper, we present an empirical investigation that starts from first principles, determining different categories of stressors for which emotional support might be useful, different categories of emotional support utterances and promising algorithms for deciding the content and form of textual emotional support messages according to the stressors present. At each stage, the results are validated through empirical experiments with human participants who, for instance, are required to place statements into categories, evaluate possible support messages in different imagined situations and compose their own emotional support from options offered. This development methodology allows us to avoid potentially challenging ethical issues in presenting people with stressful situations. Although our algorithms are attempting to choose emotional support based on the general, “naive” competence of human speakers, we use as a running example situations that can arise when attending a medical emergency and awaiting expert help.

We present an incremental Bayesian model that resolves key issues of crowd size and data quality for consensus labeling. We evaluate our method using data collected from a real-world citizen science program, B ee W atch , which invites members of the public in the United Kingdom to classify (label) photographs of bumblebees as one of 22 possible species. The biological recording domain poses two key and hitherto unaddressed challenges for consensus models of crowdsourcing: (1) the large number of potential species makes classification difficult, and (2) this is compounded by limited crowd availability, stemming from both the inherent difficulty of the task and the lack of relevant skills among the general public. We demonstrate that consensus labels can be reliably found in such circumstances with very small crowd sizes of around three to five users (i.e., through group sourcing). Our incremental Bayesian model, which minimizes crowd size by re-evaluating the quality of the consensus label following each species identification solicited from the crowd, is competitive with a Bayesian approach that uses a larger but fixed crowd size and outperforms majority voting. These results have important ecological applicability: biological recording programs such as B ee W atch can sustain themselves when resources such as taxonomic experts to confirm identifications by photo submitters are scarce (as is typically the case), and feedback can be provided to submitters in a timely fashion. More generally, our model provides benefits to any crowdsourced consensus labeling task where there is a cost (financial or otherwise) associated with soliciting a label.

The CURIOS project explores how digital archives for rural community heritage groups can be made more sustainable so that volunteer members can maintain a lasting digital presence. It is developing software tools to help remote rural communities to collaboratively maintain and present information about their cultural heritage. The objective is to investigate the use of semantic web/linked data technology to build a general, flexible and "future proof" software platform that could help such projects to develop digital archives and to be sustainable over time. As an interdisciplinary project we aim to synthesise a narrative that draws from both social science and computer science perspectives by critically reflecting upon the novel approach taken and the on-going results that are being produced.

Presenting entailments of axioms in a formal ontology is a non-trivial task. This position paper argues that the problem of selecting entailments for presentation to users is not adequately acknowledged or addressed in the literature or in implemented systems. We analyse the problem and consider some di↵erent approaches that can help to address the problem.

Gemma Webster, Chris Mellish, Somayajulu G. Sripada, Rene Van Der Wal, Koen Arts, Yolanda Melero, Xavier Lambin. Proceedings of the 8th International Natural Language Generation Conference (INLG). 2014.

The ourSpaces Virtual Research Environment makes use of Semantic Web technologies to create a platform to support multi-disciplinary research groups. This paper introduces the main semantic components of the system: a framework to capture the provenance of the research process, a collection of services to create and visualise metadata and a policy reasoning service. We also describe different approaches to authoring and accessing metadata within the VRE. Using evidence gathered from data provided by the users of the system we discuss the lessons learnt from deployment with three case study groups.

Community First Responders are often the first on the scene in rural medical emergencies, and experience situations where stress is inherent. One way of reducing this stress is through the use of an Empathic Conversational Agent which can provide Emotional Support, and be applied to systems designed for the pre-hospital care domain. This paper outlines the groundwork for this goal by describing the development of stories which describe a particular stressor, and validated Emotional Support categories and statements. These will be used when developing an algorithm for adapting Emotional Support to different stressful situations. We identified 5 categories of Emotional Support, and a set of 52 statements which were validated as belonging to these categories. We then present a preliminary analysis of the patterns of usage of Emotional Support categories for each of the stressors.

This paper describes the development of an algorithm for use by an Empathic Conversational Agent for choosing appropriate emotional support messages to a learner receiving feedback on their performance. We present a study where we employed a User as Wizard approach to explore how such statements were used by people giving feedback to a fictional learner with high or low conscientiousness and varying grades. We found that the type of emotional support employed depended primarily on the grade that had been achieved, but conscientiousness also influenced the amount of advice given. Interesting differences were found in how people combine Emotional Support messages depending on grade and level of conscientiousness, which inspired an algorithm for automatically choosing appropriate messages depending on context.

In this paper we design and implement an artificial task-based scenario that seeks to induce loneliness and acute stress. We explore how the presence of a virtual teammate called Mary and the differing types of support that she provides affects users stress during a task. We investigate how empathic support, task support and the combination of task and empathic support affect stress. Stress is measured using both physiological sensors (skin conductance and heart rate) and self-reporting questionnaires. The results obtained offer insight into the best type of support to give individuals taking part in critical situations, such as in our chosen domain of pre-hospital care. This work lays down the foundations for future work in the development of an intelligent algorithm for a virtual teammate which can provide benefits to both casualty and carer welfare.

This paper describes how a set of stories, each conveying a personality trait from the Five Factor Model at a high or low level, were developed using Amazon's Mechanical Turk. These stories will be used to develop interfaces that adapt to personality using a User as Wizard method. The paper shows how difficult it is to construct stories that convey a single personality trait. It also shows how such stories can be constructed for most cases, and how Mechanical Turk can aid to achieve this.

Abstract We investigate the use of instance-based ranking methods for surface realization in natural language generation. Our approach to instance-based natural language generation (IBNLG) employs two components: a rule system that ‘overgenerates’ a number of realization candidates from a meaning representation and an instance-based ranker that scores the candidates according to their similarity to examples taken from a training corpus. We develop an efficient search technique for identifying the optimal candidate based on a novel extension of theA* algorithm. The rule system is produced automatically from a semantically annotated fragment of the Penn Treebank II containing management succession texts. We detail the annotation scheme and grammar induction algorithm and evaluate the efficiency and output of the generator. We also discuss issues such as input coverage (completeness) and fluency that are relevant to surface generation in general.

In this article we discuss several metrics of coherence defined using centering theory and investigate the usefulness of such metrics for information ordering in automatic text generation. We estimate empirically which is the most promising metric and how useful this metric is using a general methodology applied on several corpora. Our main result is that the simplest metric (which relies exclusively on NOCB transitions) sets a robust baseline that cannot be outperformed by other metrics which make use of additional centering-based features. This baseline can be used for the development of both text-to-text and concept-to-text generation systems.

Based upon qualitative work done with former Neonatal Intensive Care Unit parents, we propose a potential user model to estimate the level of stress/anxiety that a parent is experiencing and how information given to such parents should be adjusted to meet their informational and emotional needs.

Natural Language Generation (NLG) can be used to generate textual summaries of numeric data sets. In this paper we develop an architecture for generating short (a few sentences) summaries of large (100KB or more) time-series data sets. The architecture integrates pattern recognition, pattern abstraction, selection of the most significant patterns, microplanning (especially aggregation), and realisation. We also describe and evaluate SumTime-Turbine, a prototype system which uses this architecture to generate textualsummaries of sensor data from gas turbines.

Part One: Machine translation: the early years Editor's Introduction 1 The linguistic basis of a mechanical thesaurus, and its application to English preposition classification 2 Linguistics and machine translation Part Two: Probabilistic grammar and the corpus Editor's Introduction 3 Towards probabilistic interpretations 4 Corpus studies and probabilistic grammar 5 Language as system and language as instance: the corpus as a theoretical construct 6 [with Z L James] A quantitative study of polarity and primary tense in the English Finite clause 7 Quantitative studies and probabilities in grammar 8 The spoken language corpus. Part Three: Towards 'intelligent computing' (computing with Meaning) Editor's Introduction 9 On language in relation to fuzzy logic and intelligent computing 10 Fuzzy grammatics: a systemic functional approach to fuzziness in neutral language 11 Computing meaning: some reflections on past experience and present prospects Appendix: Systems of the English clause: a trial grammar for the PENMAN text generation project. [Information Sciences Institute, University of Southern California].

We present the RAGS (Reference Architecture for Generation Systems) framework: a specification of an abstract Natural Language Generation (NLG) system architecture to support sharing, re-use, comparison and evaluation of NLG technologies. We argue that the evidence from a survey of actual NLG systems calls for a different emphasis in a reference proposal from that seen in similar initiatives in information extraction and multimedia interfaces. We introduce the framework itself, in particular the two-level data model that allows us to support the complex data requirements of NLG systems in a flexible and coherent fashion, and describe our efforts to validate the framework through a range of implementations.

We use a reliably annotated corpus to compare metrics of coherence based on Centering Theory with respect to their potential usefulness for text structuring in natural language generation. Previous corpus-based evaluations of the coherence of text according to Centering did not compare the coherence of the chosen text structure with that of the possible alternatives. A corpus-based methodology is presented which distinguishes between Centering-based metrics taking these alternatives into account, and represents therefore a more appropriate way to evaluate Centering from a text structuring perspective.

Generic software architectures aim to support re-use of components, focusing of research and development effort, and evaluation and comparison of approaches. In the field of natural language processing, generic frameworks for understanding have been successfully deployed to meet all of these aims, but nothing comparable yet exists for generation. The nature of the task itself, and the current methodologies available to research it, seem to make it more difficult to reach the necessary level of consensus to support generic proposals. Recent work has made progress towards establishing a generic framework for generation at the functional level, but left open the issue of actual implementation. In this paper, we discuss the requirements for such an implementation layer for generation systems, drawing on two initial attempts to implement it. We argue that it is possible and useful to distinguish “functional architecture” from “implementation architecture” for generation systems.

We present the RAGS (Reference Architecture for Generation Systems) framework, a specification of an abstract Natural Language Generation (NLG) system architecture to support sharing, re-use, comparison and evaluation of NLG technologies. We argue that the evidence from a survey of actual NLG systems calls for a different emphasis in a reference proposal from that seen in similar initiatives in information extraction and multimedia interfaces. We introduce the framework itself, in particular the two-level data model that allows us to support the complex data requirements of NLG systems in a flexible and coherent fashion, and describe our efforts to validate the framework through a range of implementations.

We present a novel evolutionary model for knowledge discovery from texts (KDTs), which deals with issues concerning shallow text representation and processing for mining purposes in an integrated way. Its aims is to look for novel and interesting explanatory knowledge across text documents. The approach uses natural language technology and genetic algorithms to produce explanatory novel hypotheses. The proposed approach is interdisciplinary, involving concepts not only from evolutionary algorithms but also from many kinds of text mining methods. Accordingly, new kinds of genetic operations suitable for text mining are proposed. The principles behind the representation and a new proposal for using multiobjective evaluation at the semantic level are described. Some promising results and their assessment by human experts are also discussed which indicate the plausibility of the model for effective KDT.

MOTIVATION: The field of 'DNA linguistics' has emerged from pioneering work in computational linguistics and molecular biology. Most formal grammars in this field are expressed using Definite Clause Grammars but these have computational limitations which must be overcome. The present study provides a new DNA parsing system, comprising a logic grammar formalism called Basic Gene Grammars and a bidirectional chart parser DNA-ChartParser. RESULTS: The use of Basic Gene Grammars is demonstrated in representing many formulations of the knowledge of Escherichia coli promoters, including knowledge acquired from human experts, consensus sequences, statistics (weight matrices), symbolic learning, and neural network learning. The DNA-ChartParser provides bidirectional parsing facilities for BGGs in handling overlapping categories, gap categories, approximate pattern matching, and constraints. Basic Gene Grammars and the DNA-ChartParser allowed different sources of knowledge for recognizing E.coli promoters to be combined to achieve better accuracy as assessed by parsing these DNA sequences in real-world data sets.

Generating text in a hypermedia environment places different demands on a text generation system than occurs in non-interactive environments. This paper describes some of these demands, then shows how the architecture of one text generation system, ILEX, has been shaped by them. The architecture is described in terms of the levels of linguistic representation used, and the processes which map between them. Particular attention is paid to the processes of content selection and text structuring.

The RAGS project aims to define a reference architecture for Natural Language Generation (NLG) systems. Currently the major part of this architecture consists of a set of datatype definitions for specifying the input and output formats for modules within NLG systems. In this paper we describe our efforts to reinterpret an existing NLG system in terms of these definitions. The system chosen was the Caption Generation System.

It is not a rare phenomenon for human written text to use non-restrictive NP modifiers to express essential pieces of information or support the situation presented in the main proposition containing the NP, for example, "Private Eye, which couldn't afford the libel payment, had been threatened with closure."(from Wall Street Journal) Yet no previous research in NLG investigates this in detail.This paper describes corpus analysis and a psycholinguistic experiment regarding the acceptability of using non-restrictive NP modifiers to express semantic relations that might normally be signalled by 'because' and 'then'.The experiment tests several relevant factors and enables us to accept or reject a number of hypotheses.The results are incorporated into an NLG system based on a Genetic Algorithm.

In natural language generation, different generation tasks often interact with each other in a complex way. We think that how to resolve the complex interactions inside and between tasks is more important to the generation of a coherent text than how to model each individual factor. This paper focuses on the interaction between aggregation and text planning, and tries to explore what preferences exist among the features considered by the two tasks. The preferences are implemented in two generation systems, namely ILEX-TS and a text planner using a Genetic Algorithm. The evaluation emphasises the second implementation and shows that capturing these preferences properly can lead to coherent text.

This paper outlines a text generation system suited to a large class of information sources, relational databases. We focus on one aspect of the problem: the additional information which needs to be specified to produce reasonable text quality when generating from relational databases. We outline how databases need to be prepared, and then describe various types of domain semantics which can be used to improve text quality.

We will demonstrate the ILEX system, a system which dynamically generates descriptions of database objects for the web, adapting the description to the discourse context and user type. Among other improvements in version 3, the system now generates from relational databases, and this demonstration will focus on this ability. We will also show how incremental extensions to the domain semantics improve the quality of the text produced.

This paper describes the latest version of the SOLE concept-tospeech system, which uses linguistic information provided by a natural language generation system to improve the prosody of synthetic speech. We discuss the types of linguistic information that prove most useful and the implications for text-to-speech systems. 1. INTRODUCTION The purpose of the SOLE project is to make use of automaticallygenerated, high-level linguistic information to improve the quality of the intonation of synthetic speech. After choosing an initial set of linguistic constructs thought to have some influence on prosody, we developed an SGML-based mark-up language to serve as a general interface between NLG and speech synthesis systems, and trained our synthesis system to recognise correlations between the mark-up and intonational contours so that it can make use of this mark-up when synthesising. As a result, many of the errors that the synthesiser makes with regard to knowing when to accent or deaccent ...

Dynamic hypertext' is hypertext which is automatically generated at the point of need.A number of NLG systems have now been developed to operate within a hypertext environment; and now that these systems are becoming widely available on the World Wide Web, it is useful to take stock of how well-equipped NLG technology is to work in this new domain.A generation system in a hypertext environment faces a specific set of requirements; here, we discuss those requirements, and the resources that can be provided to help meet them.Examples are drawn from a number of systems, including our own prototype, ILEX-0.We conclude by indicating that the major benefit of such systems could be in the way that they combine flexibility with the illusion of user control.

This paper presents a technique for sentence generation.We argue that the input to generators should have a non-hierarchical nature.This allows us to investigate a more general version of the sentence generation problem where one is not pre-committed to a choice of the syntactically prominent elements in the initial semantics.We also consider that a generator can happen to convey more (or less) information than is originally specified in its semantic input.In order to constrain this approximate matching of the input we impose additional restrictions on the semantics of the generated sentence.Our technique provides flexibility to address cases where the entire input cannot be precisely expressed in a single sentence.Thus the generator does not rely on the strategic component having linguistic knowledge.We show clearly how the semantic structure is declaratively related to linguistically motivated syntactic representation.

The IDAS natural-language generation system uses a KL-ONE type classifier to perform content determination, surface realisation, and part of text planning. Generation-by-classification allows IDAS to use a single representation and reasoning component for both domain and linguistic knowledge, which is difficult for systems based on unification or systemic generation techniques.

We argue for the usefulness of an active chart as the basis of a system that searches for the globally most plausible explanation of failure to syntactically parse a given input. We suggest semantics-free, grammar-independent techniques for parsing inputs displaying simple kinds of ill-formedness and discuss the search issues involved.