Teaching Machines to Read and Comprehend
"Teaching machines to read natural language documents remains an elusive challenge. Machine reading systems can be tested on their ability to answer questions posed on the contents of documents that they have seen, but until now large scale training and test datasets have been missing for this type of evaluation. In this work we define a new methodology that resolves this bottleneck and provides large scale supervised reading comprehension data. This allows us to develop a class of attention based deep neural networks that learn to read real documents and answer complex questions with minimal prior knowledge of language structure."
Google DeepMind Teaches Artificial Intelligence Machines to Read
"The best way for AI machines to learn is by feeding them huge data sets of annotated examples, and the Daily Mail has unwittingly created one."
Supporting the Adaptation of Texts for Poor Literacy Readers: a Text Simplification Editor for Brazilian Portuguese
"In this paper we investigate the task of text simplification for Brazilian Portuguese. Our purpose is three-fold: to introduce a simplification tool for such language and its underlying development methodology, to present an on-line authoring system of simplified text based on the previous tool, and finally to discuss the potentialities of such technology for education. The resources and tools we present are new for Portuguese and innovative in many aspects with respect to previous initiatives for other languages."
Making It Simplext: Implementation and Evaluation of a Text Simplification System for Spanish
"In this article, we present results from the Simplext project, which is dedicated to automatic text simplification for Spanish. We present a modular system with dedicated procedures for syntactic and lexical simplification that are grounded on the analysis of a corpus manually simplified for people with special needs. We carried out an automatic evaluation of the system’s output, taking into account the interaction between three different modules dedicated to different simplification aspects. One evaluation is based on readability metrics for Spanish and shows that the system is able to reduce the lexical and syntactic complexity of the texts. We also show, by means of a human evaluation, that sentence meaning is preserved in most cases. Our results, even if our work represents the first automatic text simplification system for Spanish that addresses different linguistic aspects, are comparable to the state of the art in English Automatic Text Simplification."
Measuring Text Simplification with the Crowd
"This paper focuses on the evaluation of English text simplification using the crowd. We show that leveraging crowds can result in a collective decision that is accurate and converges to a consensus rating. Our results from 2,500 crowd annotations show that the crowd can effectively rate levels of simplicity. This may allow simplification systems and system builders to get better feedback about how well content is being simplified, as compared to standard measures which classify content into ‘simplified’ or ‘not simplified’ categories. Our study provides evidence that the crowd could be used to evaluate English text simplification, as well as to create simplified text in future work."
A Plug-in to Aid Online Reading in Spanish
"Reading text on the Web is a challenging task for many people, such as those with cognitive impairments, reading diffi culties or people who are learning a new language. In this paper we present a web browser plug-in to help with reading Spanish text on the Web. The plug-in is freely available for Chrome and presents definitions and simpler synonyms on demand for the selected web text. The tool was modi fied following the suggestions of 5 people (2 with diagnosed dyslexia) who tested the tool using the think aloud protocol and undertook a subsequent interview."
Dyslexia and Web Accessibility: Synergies and Challenges
"This paper reviews the main challenges of studying dyslexia for web accessibility. These are: (1) measuring the impact of dyslexia in the population; (2) the limitations of the up-todate studies; and (3) including dyslexia in the Web accessibility guidelines. While some aspects are already addressed by the guidelines, we propose the inclusion of simple recommendations for typeface and font size that would benefit both people with and without dyslexia. We also suggest a change in the current methodologies to overcome up-todate research limitations using larger and more representative datasets."
Detecting Readers with Dyslexia Using Machine Learning with Eye Tracking Measures
"Worldwide, around 10% of the population has dyslexia, a specific learning disorder. Most of previous eye tracking experiments with people with and without dyslexia have found differences between populations suggesting that eye movements reflect the difficulties of individuals with dyslexia. In this paper, we present the first statistical model to predict readers with and without dyslexia using eye tracking measures. The model is trained and evaluated in a 10-fold cross experiment with a dataset composed of 1,135 readings of people with and without dyslexia that were recorded with an eye tracker. Our model, based on a Support Vector Machine binary classifier, reaches 80.18% accuracy using the most informative features. To the best of our knowledge, this is the first time that eye tracking measures are used to predict automatically readers with dyslexia using machine learning."
CASSA: A Context-Aware Synonym Simplification Algorithm
"We present a new context-aware method for lexical simplification that uses two free language resources and real web frequencies. We compare it with the state-of-the-art method for lexical simplification in Spanish and the established simplification baseline, that is, the most frequent synonym. Our method improves upon the other methods in the detection of complex words, in meaning preservation, and in simplicity. Although we use Spanish, the method can be extended to other languages since it does not require alignment of parallel corpora."
Exercises for German-Speaking Children with Dyslexia
"In this work-in-progress we present a computer-based method to design German reinforcement exercises for children with dyslexia. From different schools, we collected more than 1,000 errors written by children with dyslexia. Then, we created a classification of dyslexic errors in German and annotated the errors with different language specific features, such as phonetic and visual features. For the creation of the exercises we took into account the linguistic knowledge extracted from the analyses and designed more than 2,500 word exercises in German that have been integrated in a game available for iOS. The game and the resource of dyslexic errors are available online (https://itunes. apple.com/us/app/dyseggxia/id534986729?mt=8)"
How to Present more Readable Text for People with Dyslexia
"This paper presents a set of recommendations to customize texts on a computer screen in a more accessible way for this target group. This set is based in an eye tracking study with 92 people, 46 with dyslexia and 46 as control group, where the reading performance of the participants was measured. The following parameters were studied: color combinations for the font and the screen background, font size, column width as well as character, line and paragraph spacing. It was found that larger text and larger character spacing lead the participants with and without dyslexia to read significantly faster...So far, these recommendations based on eye tracking data are the most complete for people with dyslexia."
A Game to Target the Spelling of German Children with Dyslexia
"Playing error-based exercises presented in a computer game was found to significantly improve the spelling skills of children with dyslexia in Spanish. Since there are no similar error-based exercises for German, we adapted the method to German and created 2,500 new word exercises. Since dyslexia manifestations are language dependent, the replication of the method required (i) collecting new texts written by German children with dyslexia; (ii) the annotation and the linguistic analysis of the errors; and (iii) the creation of exercises as well as their integration in the tool."
Dytective: Toward a Game to Detect Dyslexia
"Detecting dyslexia is crucial so that people who have dyslexia can receive training to avoid associated high rates of academic failure. In this paper we present Dytective, a game designed to detect dyslexia. The results of a within-subjects experiment with 40 children (20 with dyslexia) show significant differences between groups who played Dytective. These differences suggest that Dytective could be used to help identify those likely to have dyslexia."
A Spellchecker for Dyslexia
"Poor spelling is a challenge faced by people with dyslexia throughout their lives. Spellcheckers are therefore a crucial tool for people with dyslexia, but current spellcheckers do not detect real-word errors, which are a common type of errors made by people with dyslexia. Real-word errors are spelling mistakes that result in an unintended but real word, for instance, form instead of from. In this paper, we introduce a system called Real Check that uses a probabilistic language model, a statistical dependency parser and Google n-grams to detect real-world errors. We evaluated Real Check on text written by people with dyslexia, and showed that it detects more of these errors than widely used spellcheckers. In an experiment with 34 people (17 with dyslexia), people with dyslexia corrected sentences more accurately and in less time with Real Check"
Effect of web page menu orientation on retrieving information by people with learning disabilities
"The Internet facilitates the provision of accessible information to people with learning disabilities. However, problems with navigation and retrieval represent a barrier for this cohort. This article addresses one aspect of page design, testing whether a horizontal or vertical contents arrangement facilitates faster access to content for people with learning disabilities. Participants were timed as they looked for one-word “dummy” menu entries appearing in various locations along a horizontal or vertical grid. The words corresponded to images shown at random in a word-search type activity. Results were analyzed using mixed effects models. Results showed that mean search times increased as the position shifted from left to right and from top to bottom. Thus, participants undertook the test as if it were a reading exercise, despite the images appearing in the center of the page and the words appearing at random positions. The research also suggests that a horizontal menu may be more effective than a vertical one, with the most important links placed on the left. The propensity to imbibe information “serially” (word-for-word) rather than to skim or look “globally” has important website design implications."
Cognitive Web Accessibility: Readability 2015
Published in 2015, these resources are original studies, literature reviews and/or related articles that cite references.
Note: If no resources are displayed above, please refresh this page.
Visit The Clear Helper Blog: Developing best practices of Web accessibility for people with intellectual / cognitive disabilities.