Search

Este trabajo presenta el programa CoLE: una aplicación software para la Corrección de los errores en la Lectura y Escritura en niños con dislexia. Nuestra hipótesis es que los niños disléxicos pueden mejorar la lectoescritura a partir del uso de un programa que ejercite capacidades alteradas en este trastorno, como son la discriminación fonológica, la descodificación visual, la memoria a corto plazo, la percepción y la secuenciación (Herrera-Pino et al., 2007). Desarrollar el programa como una aplicación software tiene una doble finalidad: por un lado, aumentar la motivación del niño y por otro, que las actividades puedan ser realizadas sin supervisión (autoevaluables). CoLE cuenta con los siguientes tipos de actividades: conciencia fonológica, memoria visual, ritmo, velocidad lectora y discriminación visual. La aplicación incluye dos modalidades de uso: (i) juego y (ii) test. Por un lado, la modalidad juego permite al sujeto practicar los diferentes tipos de actividades recibiendo feedback inmediato del resultado. Además, se puede seleccionar el nivel de dificultad o que éste vaya aumentando progresivamente. Por otro lado, la modalidad de test (sin feedback) presenta una secuencia fija de actividades de cada tipo para que los resultados sean comparables en diferentes niños. Los test quedan guardados en una base de datos para su posterior análisis. Actualmente se está realizando una evaluación del programa con 4 niños con dificultades en la lectoescritura de edades comprendidas entre los 5 y los 12 años. Hasta el momento se advierte una alta predisposición por su parte a utilizar la aplicación. Esperamos, además, alcanzar aumentos en la velocidad y exactitud lectora, así como una disminución en los errores ortográficos. La aplicación está disponible en la web para la experimentación por parte de otros profesionales (http://www.centroafan.com/afan-app/www/). This paper presents the CoLE program: a software application for error Correction in Reading and Writing in children with dyslexia. Our hypothesis is that dyslexic children can improve reading and writing through the use of a program that exercises the capabilities affected in this disorder, such as phonological discrimination, visual decoding, short term memory, perception and sequencing (Herrera-Pino et al., 2007). Developing the program as a software app has a twofold aim: first, increasing the motivation of the child and secondly, making it possible for the activities to be carried out without supervision (self-assessable). CoLE includes the following types of activities: phonological awareness, visual memory, rhythm, reading speed and visual discrimination. The application can be used in two modes: (i) game and (ii) test. On one hand, the game mode allows the subject to practice the different types of activities receiving immediate feedback on the outcome. Furthermore, the level of difficulty can be selected or otherwise increase gradually. On the other hand, the test mode (without feedback) shows a fixed sequence of activities of each type so that the results are comparable between different children. The tests are stored in a database for further analysis. Currently, the program is being evaluated on 4 children with reading and writing difficulties, aged between 5 and 12 years. So far we observed a high predisposition in children to use the application. Moreover, we expect to achieve increases in reading speed and accuracy, as well as a decrease in the misspellings. The application is available on the web for experimentation by other professionals (http://www.centroafan.com/afan-app/www/).

This paper introduces a new class of temporal expression – named temporal expressions – and methods for recognising and interpreting its members. The commonest temporal expressions typically contain date and time words, like April or hours. Research into recognising and interpreting these typical expressions is mature in many languages. However, there is a class of expressions that are less typical, very varied, and difficult to automatically interpret. These indicate dates and times, but are harder to detect because they often do not contain time words and are not used frequently enough to appear in conventional temporally-annotated corpora for example Michaelmas or Vasant Panchami.Using Wikipedia and linked data, we automatically construct a resource of English named temporal expressions, and use itto extract training examples from a large corpus. These examples are then used to train and evaluate a named temporal ex-pression recogniser. We also introduce and evaluate rules for automatically interpreting these expressions, and we observe thatuse of the rules improves temporal annotation performance over existing corpora This paper introduces a new class of temporal expression – named temporal expressions – and methods for recognising and interpreting its members. The commonest temporal expressions typically contain date and time words, like April or hours. Research into recognising and interpreting these typical expressions is mature in many languages. However, there is a class of expressions that are less typical, very varied, and difficult to automatically interpret. These indicate dates and times, but are harder to detect because they often do not contain time words and are not used frequently enough to appear in conventional temporally-annotated corpora – for example Michaelmas or Vasant Panchami. UsingWikipedia and linked data, we automatically construct a resource of English named temporal expressions, and use it to extract training examples from a largecorpus. These examples are then used to train and evaluate a named temporal expression recogniser. We also introduce and evaluate rules for automatically interpreting these expressions, and we observe that use of the rules improves temporal annotation performance over existing corpora.

Automatic annotation of temporal expressions is a research challenge of great interest in the field of information extraction. Gold standard temporally-annotated resources are limited in size, which makes research using them difficult. Standards have also evolved over the past decade, so not all temporally annotated data is in the same format. We vastly increase available human-annotated temporal expression resources by converting older format resources to TimeML/TIMEX3. This task is difficult due to differing annotation methods. We present a robust conversion tool and a new, large temporal expression resource. Using this, we evaluate our conversion process by using it as training data for an existing TimeML annotation tool, achieving a 0.87 F1 measure -- better than any system in the TempEval-2 timex recognition exercise., Comment: Proc. LREC (2012)