Raquel Hervás

Story plot generation based on CBR

In this paper we present a system for automatic story generation that reuses existing stories to produce a new story that matches a given user query. The plot structure is obtained by a case-based reasoning (CBR) process over a case base of tales and an ontology of explicitly declared relevant knowledge. The resulting story is generated as a sketch of a plot described in natural language by means of natural language generation (NLG) techniques.

One Half or 50%? An Eye-Tracking Study of Number Representation Readability

Are numbers expressed as digits easier to read and understand than written with letters? What about fractions and percentages? Exact or rounded values? We present an eye-tracking study that attempts to answer these questions for Spanish, using fixation and reading time to measure readability as well as comprehension questions to score understandability. We find that digits are faster to read but do not help comprehension. Fractions help understandability while percentages help readability. No significant results were found concerning the influence of rounding. Our experiments were performed by 72 persons, half of them with dyslexia. To the best of our knowledge, this is the first study that addresses the cognitive load of number representation in any language, even more for people with dyslexia.

EmoTales: creating a corpus of folk tales with emotional annotations

Emotions are inherent to any human activity, including human–computer interactions, and that is the reason why recognizing emotions expressed in natural language is becoming a key feature for the design of more natural user interfaces. In order to obtain useful corpora for this purpose, the manual classification of texts according to their emotional content has been the technique most commonly used by the research community. The use of corpora is widespread in Natural Language Processing, and the existing corpora annotated with emotions support the development, training and evaluation of systems using this type of data. In this paper we present the development of an annotated corpus oriented to the narrative domain, called EmoTales, which uses two different approaches to represent emotional states: emotional categories and emotional dimensions. The corpus consists of a collection of 1,389 English sentences from 18 different folk tales, annotated by 36 different people. Our model of the corpus development process includes a post-processing stage performed after the annotation of the corpus, in which a reference value for each sentence was chosen by taking into account the tags assigned by annotators and some general knowledge about emotions, which is codified in an ontology. The whole process is presented in detail, and revels significant results regarding the corpus such as inter-annotator agreement, while discussing topics such as how human annotators deal with emotional content when performing their work, and presenting some ideas for the application of this corpus that may inspire the research community to develop new ways to annotate corpora using a large set of emotional tags.

Enrichment of automatically generated texts using metaphor

Computer-generated texts are yet far from human-generated ones. Along with the limited use of vocabulary and syntactic structures they present, their lack of creativeness and abstraction is what points them as artificial. The use of metaphors and analogies is one of the creative tools used by humans that is difficult to reproduce in a computer. A human writer would not have difficulties to find conceptual relations between the domain he is writing about and his knowledge about other domains in the world, using this information in the text avoiding possible confusion. However, this task is not trivial for a computer. This paper presents an approach to the use of metaphors for referring to concepts in an automatically generated text. From a given mapping between the concepts of two domains we intend to generate metaphors for some concepts relating them with the target metaphoric domain and insert these metaphorical references in a text. We also study the ambiguity induced by metaphor and how to reduce it.

Assessing the Novelty of Computer-Generated Narratives Using Empirical Metrics

Novelty is a key concept to understand creativity. Evaluating a piece of artwork or other creation in terms of novelty requires comparisons to other works and considerations about the elements that have been reused in the creative process. Human beings perform this analysis intuitively, but in order to simulate it using computers, the objects to be compared and the similarity metrics to be used should be formalized and explicitly implemented. In this paper we present a study on relevant elements for the assessment of novelty in computer-generated narratives. We focus on the domain of folk-tales, working with simple plots and basic narrative elements: events, characters, props and scenarios. Based on the empirical results of this study we propose a set of computational metrics for the automatic assessment of novelty. Although oriented to the implementation of our own story generation system, the measurement methodology we propose can be easily generalized to other creative systems.

How to make numerical information accessible: experimental identification of simplification strategies

Public information services and documents should be accessible to the widest possible readership. Information in newspapers often takes the form of numerical expressions which pose comprehension problems for people with limited education. A first possible approach to solve this important social problem is making numerical information accessible by rewriting difficult numerical expressions in a simpler way. To obtain guidelines for performing this task automatically, we have carried out a survey in which experts in numeracy were asked to simplify a range of proportion expressions, with three readerships in mind: (a) people who did not understand percentages; (b) people who did not understand decimals; (c) more generally, people with poor numeracy. Responses were consistent with our intuitions about how common values are considered simpler and how the value of the original expression influences the chosen simplification.

Case-based reasoning for knowledge-intensive template selection during text generation

The present paper describes a case-based reasoning solution for solving the task of selecting adequate templates for realizing messages describing actions in a given domain. This solution involves the construction of a case base from a corpus of example texts, using information from WordNet to group related verbs together. A case retrieval net is used as a memory model. A taxonomy of the concepts involved in the texts is used to compute similarity between concepts. The set of data to be converted into text acts as a query to the system. The process of solving a given query may involve several retrieval processes – to obtain a set of cases that together constitute a good solution for transcribing the data in the query as text messages – and a process of knowledge-intensive adaptation which resorts to a knowledge base to identify appropriate substitutions and completions for the concepts that appear in the cases, using the query as a source. We describe this case-based solution, and we present examples of how it solves the task of selecting an appropriate set of templates to render a given set of data as text.

NIL-UCM: most-frequent-value-first attribute selection and best-scoring-choice realization

The NIL entry for the challenge has been constructed upon the general architecture for developing Natural Language Generation systems provided by the TAP project (Gervas, 2007). TAP (Text Arranging Pipeline) is a set of interfaces that define generic functionality for a pipeline of tasks oriented toward natural language generation, from an initial conceptual input to surface realization as a string, with intervening stages of content planning and sentence planning.

Evolutionary and Case-Based Approaches to REG: NIL-UCM-EvoTAP, NIL-UCM-ValuesCBR and NIL-UCM-EvoCBR

We propose the use of evolutionary algorithms (EAs) (Holland, 1992) to deal with the attribute selection task of referring expression generation. Evolutionary algorithms operate over a population of individuals (possible solutions for a problem) that evolve according to selection rules and genetic operators. The fitness function is a metric that evaluates each of the possible solutions, ensuring that the average adaptation of the population increases each generation. Repeating this process hundreds or thousands of times leads to very good solutions for the problem.

A model for human readable instruction generation using level-based discourse planning and dynamic inference of attributes disambiguation

This paper shows a model of automatic instruction giving for guiding human users in virtual 3D environments. A multilevel model for choosing what instruction to give in every state is presented, and so are the different modules that compose the whole generation system. How 3D information in the virtual world is used is explained, and the final order generation is detailed. This model has been implemented as a solution for the GIVE Challenge, an instruction generation challenge.