People with Autism Spectrum Disorder experience difficulties with reading comprehension and information processing, which affect their school performance, employability and social inclusion. The main goal of this work is to investigate new ways to evaluate and improve text and web accessibility for adults with autism. The first stage of this research involved using eye-tracking technology and comprehension testing to collect data from a group of participants with autism and a control group of participants without autism. This series of studies resulted in the development of the ASD corpus, which is the first multimodal corpus of text and gaze data obtained from participants with and without autism. We modelled text complexity and sentence complexity using sets of features matched to the reading difficulties people with autism experience. For document-level classification we trained a readability classifier on a generic corpus with known readability levels (easy, medium and difficult) and then used the ASD corpus to evaluate with unseen user-assessed data. For sentence-level classification, we used for the first time gaze data and comprehension testing to define a gold standard of easy and difficult sentences, which we then used as training and evaluation sets for sentence-level classification. The results showed that both classifiers outperformed other measures of complexity and were more accurate predictors of the comprehension of people with autism. We conducted a series of experiments evaluating easy-to-read documents for people with cognitive disabilities. Easy-to-read documents are written in an accessible way, following specific writing guidelines and containing both text and images. We focused mainly on the image component of these documents, a topic which has been significantly under-studied compared to the text component; we were also motivated by the fact that people with autism are very strong visual thinkers and that therefore image insertion could be a way to use their strengths in visual thinking to compensate for their difficulties in reading. We investigated the effects images in text have on attention, comprehension, memorisation and user preferences in people with autism (all of these phenomena were investigated both objectively and subjectively). The results of these experiments were synthesised in a set of guidelines for improving text accessibility for people with autism. Finally, we evaluated the accessibility of web pages with different levels of visual complexity. We provide evidence of existing barriers to finding relevant information on web pages that people with autism face and we explore their subjective experiences with searching the web through survey questions.

O crescente número de pessoas que utilizam a Web e sua vasta quantidade de conteúdo têm motivado especialistas em computação a investigar e acompanhar a experimentação desses usuários. Podese observar que um grande número de usuários que adere ao uso da Internet é o das pessoas com meia-idade, de 40 a 59 anos, e os idosos com idade superior a 60 anos. Trata-se de um público interessado e potencialmente pró-ativo para usufruir das possibilidades que a Web proporciona a cada dia. No entanto, barreiras que dificultam o acesso à informação são naturalmente encontradas com o avançar da idade. Dificuldades como a perda parcial ou total da visão, a redução de precisão dos movimentos, e a diminuição de memorização dos passos a serem realizados são alguns exemplos dessas barreiras. Para superar as dificuldades apresentadas pelos usuários, estudos têm sido realizados visando minimizá-las, de maneira geral. Seus resultados encontram-se sintetizados em normas / orientações a serem seguidas visando proporcionar melhor usabilidade e acessibilidade para toda diversidade de usuários. Nesse contexto, na presente pesquisa, investiga-se quais dessas normas junto com os padrões para a criação de websites que disponibilizam conteúdos proporcionam melhor adequação para atender à demanda das pessoas com idade mais avançada. Assim, foram estudados diferentes tipos de menus de navegação que disponibilizam as subnavegações conforme a necessidade do usuário (no inglês conhecido como menus drop-down) com diferentes propriedades e avaliado qual menu e suas propriedades apresentavam melhores resultados, entre eles, foi verificado o tempo de uso do menu e o número de erros cometidos para uma determinada tarefa. Conclui-se que, menus com melhores contraste e um tempo mediano de resposta ao realizar a interação, apresentaram melhores resultados / With the constant number of people who use facilitated through the use of the Web and its vast amount of content, have motivated computer specialists to investigate and monitor the trial of those users. It may be noted that a large number of users joining the Internet use is that of people aged mature (middle-aged, 40 to 59 years, and the elderly aged 60 years). It is a potentially interested public and pro-active to enjoy the possibilities that the Web provides each day. However, barriers to access to information are found naturally with advancing age, difficulties as the partial or total loss of vision, reducing accuracy of movements, and decreased retention of steps to be performed are some examples of these barriers . To overcome the difficulties presented by the users, studies have been conducted in order to minimize them in general. Their results are summarized in standards / guidelines to be followed in order to provide better usability and accessibility for the whole range of different types of users. In this context, the present study, we investigate which of these standards along with standards for the creation of websites that provide content provide better adaptation to meet the demand of people with older age. Thus, we studied different types of navigation menus that provide the subnavegações as needed by the user with different properties and evaluated their properties which menu and had better results, among them was found the time to use the menu and the number of errors for a particular task. It can be concluded that menus with better contrast and a median time of response to make the interaction, showed better results

Webbens centrala position i det moderna informationssamhället kan svårligen ifrågasättas. Men övergången till ett digitalt informationssamhälle kan också innebära problem för en grupp människor i samhället. I takt med webbens tilltagande inflytande i samhället ökar den demokratiska betydelsen av att innehållet är tillgängligt för alla som vill ta del av det, oavsett fysiska och kognitiva förutsättningar. För att underlätta för personer med funktionsnedsättningar att använda internet växte begreppet webbtillgänglighet fram. Trots ämnets ökade relevans har studier konstaterat att webben av idag är behäftad med stora tillgänglighetsbarriärer. Webbutvecklare är dem yrkeskategori som har störst inflytande över webbplatsers tekniska utformning och således ett starkt inflytande över webbens tillgänglighet. Den här studien undersöker varför professionella webbutvecklare inte efterlever standarder och inte skapar tillgängliga webbplatser – trots att en rad uppsättning riktlinjer och verktyg finns tillhands för att åstadkomma detta. Studien följer en interpretativ forskningsansats och bygger på semi-strukturerade intervjuer med fem professionella webbutvecklare. Ett av studiens grundläggande slutsatser är att webbtillgänglighet idag ofta är nedprioriterad aspekt som endast beaktas explicit om beställaren efterfrågar detta. Ett hinder utvecklare möter i arbetet med att skapa tillgängliga webbplatser är således brist på efterfrågan från beställare vilket i sin tur ger begränsade tids- och budgetresurser och affärsmässiga incitament. I de fall där tillgänglighet efterfrågas görs det nästan uteslutande av företrädare från offentliga organisationer vars regelmässiga kontext kräver det. I dessa sammanhang kan otydliga riktlinjer samt otillräckliga kunskaper vara en faktorer som försvårar arbetet med tillgänglighet. WCAG-riktlinjer upplevs av webbutvecklare som alltför omfattande och omoderna vilket medför att riktlinjerna inte fungerar som den stödjande resurs som är avsikten. Dessutom verkar det råda förvirring kring vad som definierar god tillgänglighet och hur en webbplats tillgänglighet ska utvärderas. Avsaknaden av moderna och lättillgängliga riktlinjer tillsammans med tydligt definierade mätpunkter och tillhörande utvärderingsverktyg är ett hinder för utvecklare i arbetet med webbtillgänglighet.

Tillgänglighet på webben är ett fenomen som blivit allt mer uppmärksammat på senare tid, i synnerhet för myndigheter, i och med nya lagar och internationella standarder. Begreppet innebär att personer med funktionsnedsättningar ska kunna uppfatta, förstå, navigera och interagera med webben. Resultatet av denna studie visar dock på att även språk kan ses som en del av arbetet med tillgänglighet. Den ökade uppmärksamheten till trots har det visat sig att det i dagsläget finns många webbplatser med bristande tillgänglighet. Vissa av dessa brister går att koppla till hur man arbetar med tillgänglighet och de problem och utmaningar som finns i detta arbete, något som denna studie har syftat att undersöka. Studien har genomförts i en kvalitativ och tolkande anda genom en fallstudie på Migrationsverket som är en av 22 så kallade strategiska myndigheter med ett särskilt ansvar i funktionshinderspolitiken. Insamlingen av det empiriska materialet har gjorts med hjälp av sex intervjuer av fem olika respondenter med ett uttalat tillgänglighetsansvar hos fallorganisationen. Det empiriska materialet har sedan jämförts och relaterats till tidigare forskning för att skapa ett trovärdigt resultat och för att möjliggöra generella slutsatser. Resultatet av studien visar att riktlinjer, användarinvolvering och tekniska hjälpmedel är vanliga komponenter i arbetet med tillgänglighet och att språk bör inkluderas som en del av tillgänglighet. Studien redogör även för att myndigheter kan ställas inför problem och utmaningar i tillgänglighetsarbetet i form av bristande attityd, bristande kompetens, bristande stöd från ledningen, otillräckliga verktyg, otillräckliga standardriktlinjer, komplicerad användarinvolvering, offentlig upphandling samt föråldrade system. / Web accessibility is a phenomenon that recently has drawn a lot of attention, particularly for governments, with enforcement of new laws and international standards. The term Web accessibility is about how people with disabilities are able to perceive, understand, navigate and interact with the web. The results of this study, however, show that language also can be a part of web accessibility. Even though the attention has increased, it has been shown that lots of websites suffer from poor accessibility. Some of these deficiencies can relate to how people work with web accessibility and the problems and challenges this work face, which is something this study has aimed to investigate.   The study has been conducted in a qualitative interpretive spirit with a case study at Migrationsverket, which is one of 22 so called strategic governments in Sweden with a special responsibility towards disability politics. The data has been gathered through six interviews with five different people with responsibility towards web accessibility within the studied organisation. The empirical data were then compared with previous research in the area in order to achieve a credible result and enable general conclusions.   The results of the study highlights guidelines, user involvement and technical tools as common components in the process of working with web accessibility and that language should be included as a part of web accessibility. The study also emphasizes that governments can bump into problems and challenges in this process in the form of lack of attitude, lack of competence, lack of managerial support, inadequate tools, inadequate standard guidelines, complicated user involvement, government procurement and out dated systems.

The World Wide Web (Web) has evolved from a collection of static pages that need reloading every time the content changes, into dynamic pages where parts of the page updates independently, without reloading it. As such, users are required to work with dynamic pages with components that react to events either from human interaction or machine automation. Often elderly and visually impaired users are the most disadvantaged when dealing with this form of interaction. Operating widgets require the user to have the conceptual design knowledge of the widget to complete the task. Users must have prior experience with the widget or have to learn to operate it independently, because often no user documentation is available. An automated Widget Prediction Framework (WPF) is proposed to address the issues discussed. It is a pre-emptive approach that predicts different types of widget and their locations in the page. Widgets with similar characteristics and functionalities are categorised based on a definition provided by widget design pattern libraries. Some design patterns are more loosely defined than others, and this causes confusion and ambiguity when identifying them. A formal method to model widgets based on a Widget Ontology was developed. The paradigm of the ontology provides a framework for developers to communicate their ideas, while reducing ambiguity between different types of widget. A Widget Prediction System (WPS) was developed using the concepts of the WPF. To select the types of widget for WPS evaluation, a widget popularity investigation was conducted. Seven of the most popular widgets from the investigation, done across fifty Websites, were selected. To demonstrate how WPF can be applied to predict widgets, fifty websites were used to evaluate the feasibility of the approach using WPS. On average, WPS achieved 61.98% prediction accuracy with two of the widgets > 84% accuracy. These results demonstrated the feasibility of the framework as the backend for tools that support elderly or visually impaired users.

Acessibilidade na Web tem como objetivo possibilitar que qualquer pessoa possa entender e interagir com o conteúdo de uma página Web, independente de deficiências ou outros fatores. Embora exista uma grande quantidade de trabalhos realizados na área, pode-se ainda perceber que pessoas com alguns tipos específicos de deficiências encontram mais dificuldades ao navegar na Web, como no caso dos usuários cegos. Com intuito de minimizar os problemas enfrentados por esses usuários cegos ao navegar na Web, foi proposta nesta tese uma abordagem que consiste em extrair automaticamente o conhecimento de navegação implícito na apresentação visual de uma página Web e alterar o design dessa navegação de uma maneira otimizada a esses usuários. Com esse objetivo, primeiramente foram estudados os trabalhos relacionados aos principais problemas enfrentados pelos usuários cegos ao interagir com a Web, assim como os conjuntos de diretrizes de acessibilidade e usabilidade específicos para minimizar esses problemas, fundamentados em um mapeamento sistemático conduzido sobre o tema. Após a identificação dos problemas e diretrizes, foi desenvolvido um classificador baseado no método Naïve Bayes, considerando-se três características comumente encontradas nos padrões de design de menus de navegação, com intuito de classificar automaticamente os links contidos nesta página em diferentes classes referentes às estruturas de navegação. Após desenvolvido, o método foi testado e avaliado em sites governamentais por meio de uma análise estatística, apresentando um boa taxa de acerto em suas classificações. Com base nas dificuldades encontradas na primeira etapa de pesquisa, foi proposto um modelo de interface próprio aos usuários cegos que utilizam a Web com auxílio de um leitor de tela. Com base no classificador desenvolvido, foi proposta uma abordagem para automaticamente transformar os links de navegação classificados pelo método, em um padrão de navegação otimizado aos leitores de tela. Como prova de conceito, foi realizada uma extração semiautomática de conteúdo das páginas Web, com intuito de transformar esse conteúdo para o novo modelo de interface proposto. Nesse contexto, foi desenvolvido um protótipo de mecanismo de apoio, denominado NavAux. Esse protótipo foi avaliado através de um percurso cognitivo. Para isso, foram definidas algumas tarefas e a exploração da interface foi realizada com apoio do leitor de tela JAWS. Os resultados apresentaram indícios de que a abordagem proposta pode ser considerada uma alternativa viável para minimizar os problemas de acessibilidade enfrentados pelos usuários cegos ao navegar na Web. / Web Accessibility aims to enable that any individual may understand and interact with a Web page content, regardless of disabilities or other factors. Although there is a large number of research studies in this area, it can be noted individuals with some specific types of disabilities find it more difficulty browsing the Web, such as those who are blind. In order to minimize the problems faced by blind users when browsing the Web, the study reported in this thesis proposed an approach which consists in automatically extracting the navigation knowledge implicit in the visual presentation of a Web page and changing this navigation design to an optimized way to those users. For this purpose, the research started with a review of the papers related to the main problems faced by blind users when interacting to the Web, as well as the specific sets of accessibility and usability guidelines to minimize these problems, based on a systematic mapping study conducted on the topic. After identifying the problems and guidelines, the next step involved the development of a classifier based on the Naïve Bayes methods, considering three features commonly found in the design patterns of navigation menus, in order to automatically classify the links contained on each page into different classes relating to the navigation structures. After developed, the method was tested and assessed in governmental websites through an statistical analysis, showing a good success rate in its classifications. Based on the difficulties encountered during the first step of the research, the study proposed an interface model tailored to blind users who use the Web with the help of a screen reader. Based on the developed classifier, the study proposed an approach to automatically transform the browsing links classified by the method into an optimized navigation standard to screen readers. As a proof of concept, the following step involved a semi-automatic extraction of Web pages content, in order to transform this content to a new proposed interface model. In this context, was developed a prototype of the mechanism, named NavAux. This prototype was evaluated through a cognitive walkthrough. To this, were defined some tasks and the interface operation was performed with support of the JAWS screen reader. The results presented signs that the proposed approach can be considered a viable alternative to minimize the accessibility problems faced by blind users when browsing the Web.

A diversificação nas formas de acesso à Web, alavancada pelo crescente uso de dispositivos ubíquos, principalmente smartphones e tablets, tem motivado o desenvolvimento de métodos para adaptação do conteúdo disponível na Web, dado que grande parte do conteúdo não foi projetado para ser acessado nesse novo contexto. O conteúdo Web é composto por diversos elementos que, em geral, fornecem interatividade aos usuários. Alguns destes elementos, tais como menus, são responsáveis por auxiliar a navegação no site, ajudando na localização e acesso das informações que por ventura o usuário esteja procurando. No entanto, muitos dos menus disponíveis na Web não foram desenvolvidos de uma forma acessível e nem suportam a interação em dispositivos, criando dificuldades que impedem os usuários interagirem satisfatoriamente com esses elementos. Além da diversificação de modos de acesso, existe também a diversificação do perfil de usuários que fazem uso dos recursos da Web. Alguns desses usuários encontram barreiras que tendem a dificultar ou limitar seu acesso às aplicações e conteúdos Web em geral. Nesse sentindo, este trabalho teve como objetivo principal abordar o problema de adaptação de conteúdo Web para dispositivos móveis, com foco nos menus, fornecendo uma solução personalizada de acordo com as preferências do usuário, a fim de que as barreiras de interação fossem reduzidas ou eliminadas. Inicialmente estudou-se os diferentes padrões de menus, bem como as diretrizes de acessibilidade e usabilidade para criação de menus acessíveis. A partir desses estudos, foi desenvolvido uma metamodelo que deu origem a linguagem AMenu, contendo todos os detalhes técnicos sobre a acessibilidade e usabilidade. Em seguida, foi desenvolvida a ferramenta AMeG, para validar a linguagem AMenu. Com base na avaliação da linguagem, foi desenvolvido o mecanismo de adaptação, que faz uso da linguagem AMenu para geração dos menus acessíveis e adaptados para dispositivos móveis. Finalmente, um estudo de caso com usuários idosos foi conduzido, a fim de comparar os menus gerados pelo mecanismo em relação aos menus originais. Os resultados obtidos apontam um melhor desempenho na interação com os menus gerados pelo mecanismo, uma vez que os participantes obtiveram uma redução de 54% no tempo de realização das tarefas e cometeram 82% menos erros. Por outro lado, os resultados apontam que todos os participantes responderam melhor à abordagem de adaptação desenvolvida neste estudo, uma vez que eles conseguiram concluir com sucesso todas as tarefas definidas para o estudo. / The diversity of means to use the Internet, ensured by the growing upcoming of ubiquitous devices such as tablets and smartphones, has promoted research on adaptation of the content that used to be mostly targeted only to desktop platform into this new context now known as the mobile devices. Some of the elements such as the menus are designed to facilitate the usage of the website once they help at finding and using the information that the user looks for. However, most of the menus available on websites do not apply the mobile devices, nor provide users with interaction, which nearly disables the interaction with the mentioned devices. There yet must be added the profile diversity of the web content users. Some of these users, such as elderly people, find limitations that tend to hamper their access to web content on the mobile devices, once aging commonly represents both physical (visual, hearing and motor abilities) and cognitive (attention, memory and reasoning issues) impairment that directly interferes on the usage of the computing devices that do not portray adjustments to such audience profile. This paper aims at investigating the issues related to the adaptation of web contents into the mobile devices. We focused on the menus in order to reduce or eliminate the limitations elderly people could find while using the mobile devices. We initially analyzed different patterns and structure of menus, as well as the guidelines upon accessibility and usability involved in the menus designing process. Based on this analysis, we developed a metamodel that originated the AMenu language, in which all technical data upon accessibility and usability involved in the menus designing process was included. Then, we developed the AMeG tool in order to assess the language created under the development perspective. Based on this assessment, we developed of an adaption mechanism that uses the AMenu language to generate menus adapted to the mobile devices. We then run a study case with elderly people in order to assess the menus generated by the mechanism, based on the comparison with the original ones. The results demonstrated that the participants had a better performance in the interaction while using the menus generated by the mechanism rather than the original ones, since they spent 54% less time performing tasks and made 82% less mistakes compared to the original menus. We also verified a quitting average of 33% among the participants while performing the assigned tasks using the original menus. On the other hand, the results point out that all participants responded better to the approach developed in this study as they all managed to successfully finish all the assigned tasks.

Users with different types of handicaps access web pages with wide scale of device. Thus, it is often difficult to develop web sites in order to be accessible for everyone. This thesis takes into consideration specific guidelines for accessible content of web pages as well as formalization of their rules and subsequent validation of these rules.

Workshop Mensch-Computer-Vernetzung Web Accessibility Gestaltung von Webseiten um eine Nutzung mit vielen verschiedenen Zugangsmgeräten (Scrennreader, Bildschirmlupe) und unter verschiedenen technischen Bedingungen (langsame Anbindung, veraltete Software, keine Farbe) weitestgehend uneingeschränkt zu ermöglichen.

Med hjälp av Internet har världens befolkning kunnat utforska oändligt många tjänster. Denna utveckling har dock lett till barriärer för personer med funktionsnedsättning. På grund av det har World Wide Web Consortium (W3C) tillsammans med Web Accessibility Initiative (WAI) skapat Web Content Accessibility Guidelines (WCAG). WCAG är riktlinjer skapade för att göra webbsidor mer tillgängliga för alla, inklusive människor med funktionsnedsättning. I Sverige finns lagen för digital offentlig service (DOS-lagen). För att upprätthålla denna lag krävs det att offentliga aktörer följer WCAG:s riktlinjer upp till och med nivå AA.Syftet med denna studie är att utifrån WCAG 2.1 få en förståelse över vilka tillgänglighetsproblem som är vanligast och om problemen korrelerar med använt webbramverk. Även de hinder som eventuellt kan finnas kring arbetet runt WCAG är av intresse. Detta för att ge underlag för framtida strategiska beslut för svenska offentliga aktörer för ett lyckat arbete med att uppfylla WCAG 2.1. En kvantitativ undersökning på 30 olika webbplatser har utförts med hjälp av ett automatiskt verktyg och en manuell undersökning av startsidorna. Semistrukturerade intervjuer utfördes med 4 olika offentliga aktörer för att kunna svara på om det finns hinder i arbetet med att följa WCAG 2.1.Resultatet beskriver de tillgänglighetsproblem som finns på de olika webbplatserna. Av kriterierna som undersöktes automatiskt var de vanligaste överträdda [1.3.1] som menar på att man ska använda HTML-element på ett korrekt sätt så innehåll kan presenteras, [4.1.1] som ska se till att koden valideras samt [4.1.2] som menar på att man ska se till att anpassade komponenter fungerar i hjälpmedel. Den manuella undersökningen visade de vanligaste överträdda kriterierna var [1.4.10] som menar på att man ska skapa en flexibel layout som fungerar vid olika skärmstorlekar och [3.3.3] som menar på att man ska kunna ge förslag på hur inmatningsfel kan rättas till vid olika sökfunktioner. Resultatet pekade även på att en korrelation mellan valt webbramverk och antalet webbtillgänglighetproblem inte existerade. De semistrukturerade intervjuerna kunde fastställa att det fanns en del utmaningar vid arbetet runt WCAG 2.1. De största orsakerna var kopplade till budget och resurser men också okunskap och inställning till WCAG. En annan utmaning var även kopplad till kompetens vid nyrekrytering. / With the help of Internet the world’s population have had the opportunity to explore the different services Internet has to offer. Although this amazing development has been advantageous to many people it has created barriers for people with different disabilities. Because of this the World Wide Web Consortium (W3C) together with the Web Accessibility Initiative (WAI) have created Web Content Accessibility Guidelines (WCAG). WCAG are guidelines for making web pages more accessible for everyone, including people with disabilities. In Sweden there is an existing law called the law of digital public service. This law demands public authorities to have accessible web pages and follow the WCAG guidelines up to level AA.The purpose of this study is, from the perspective of WCAG 2.1, to get an understanding of which accessibility problems that are most common and if there is a correlation between these problems and used web frameworks. The challenges that eventually exist around the work with WCAG 2.1 are also of interest. The purpose of this is to give basis for future strategic decision-making for Swedish public authorities. The methods that were being used during the study were both qualitative and quantitative. The quantitative research on 30 different web sites was executed with the help of an automatic testing tool and manual research of the start pages. Semi structured interviews were conducted with 4 different public authorities to find answers if there were any difficulties with the implementation of WCAG 2.1.The result describes the accessibility problems that exist on the different webpages. Of all the criterions evaluated in the automatic research, the most violated criterions were [1.3.1] which means to use HTML elements correctly so that content can be presented, [4.1.1] to make sure that the code validates and [4.1.2] to make sure that custom components work in assistive devices. The manual research showed that the most common criterions were [1.4.10] which means to create a flexible layout that works at different screen sizes and [3.3.3] which means that is should be possible to give suggestions on how errors can be corrected in different search functions. The result gave indications that there was no correlation between web framework used and amount of accessibility problems. The semi structured interviews concluded that there were many difficulties with the implementation of WCAG. The biggest reasons were connected to budget, resources, and attitude towards WCAG. Another challenge was connected to the amount of competence on WCAG when recruiting new employees.