Lau Tak Pang. / Thesis submitted in: October 2006. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (leaves 110-122). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.v / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Motivation and Major Contributions --- p.1 / Chapter 1.1.1 --- Chinese Readability Analysis --- p.1 / Chapter 1.1.2 --- Web Readability Analysis --- p.3 / Chapter 1.2 --- Thesis Chapter Organization --- p.6 / Chapter 2 --- Related Work --- p.7 / Chapter 2.1 --- Readability Assessment --- p.7 / Chapter 2.1.1 --- Assessment for Text Document --- p.8 / Chapter 2.1.2 --- Assessment for Web Page --- p.13 / Chapter 2.2 --- Support Vector Machine --- p.14 / Chapter 2.2.1 --- Characteristics and Advantages --- p.14 / Chapter 2.2.2 --- Applications --- p.16 / Chapter 2.3 --- Chinese Word Segmentation --- p.16 / Chapter 2.3.1 --- Difficulty in Chinese Word Segmentation --- p.16 / Chapter 2.3.2 --- Approaches for Chinese Word Segmentation --- p.17 / Chapter 3 --- Chinese Readability Analysis --- p.20 / Chapter 3.1 --- Chinese Readability Factor Analysis --- p.20 / Chapter 3.1.1 --- Systematic Analysis --- p.20 / Chapter 3.1.2 --- Feature Extraction --- p.30 / Chapter 3.1.3 --- Limitation of Our Analysis and Possible Extension --- p.32 / Chapter 3.2 --- Research Methodology --- p.33 / Chapter 3.2.1 --- Definition of Readability --- p.33 / Chapter 3.2.2 --- Data Acquisition and Sampling --- p.34 / Chapter 3.2.3 --- Text Processing and Feature Extraction . --- p.35 / Chapter 3.2.4 --- Regression Analysis using Support Vector Regression --- p.36 / Chapter 3.2.5 --- Evaluation --- p.36 / Chapter 3.3 --- Introduction to Support Vector Regression --- p.38 / Chapter 3.3.1 --- Basic Concept --- p.38 / Chapter 3.3.2 --- Non-Linear Extension using Kernel Technique --- p.41 / Chapter 3.4 --- Implementation Details --- p.42 / Chapter 3.4.1 --- Chinese Word Segmentation --- p.42 / Chapter 3.4.2 --- Building Basic Chinese Character / Word Lists --- p.47 / Chapter 3.4.3 --- Pull Sentence Detection --- p.49 / Chapter 3.4.4 --- Feature Selection Using Genetic Algorithm --- p.50 / Chapter 3.5 --- Experiments --- p.55 / Chapter 3.5.1 --- Experiment 1: Evaluation on Chinese Word Segmentation using the LMR-RC Tagging Scheme --- p.56 / Chapter 3.5.2 --- Experiment 2: Initial SVR Parameters Searching with Different Kernel Functions --- p.61 / Chapter 3.5.3 --- Experiment 3: Feature Selection Using Genetic Algorithm --- p.63 / Chapter 3.5.4 --- Experiment 4: Training and Cross-validation Performance using the Selected Feature Subset --- p.67 / Chapter 3.5.5 --- Experiment 5: Comparison with Linear Regression --- p.74 / Chapter 3.6 --- Summary and Future Work --- p.76 / Chapter 4 --- Web Readability Analysis --- p.78 / Chapter 4.1 --- Web Page Readability --- p.79 / Chapter 4.1.1 --- Readability as Comprehension Difficulty . --- p.79 / Chapter 4.1.2 --- Readability as Grade Level --- p.81 / Chapter 4.2 --- Web Site Readability --- p.83 / Chapter 4.3 --- Experiments --- p.85 / Chapter 4.3.1 --- Experiment 1: Web Page Readability Analysis -Comprehension Difficulty --- p.87 / Chapter 4.3.2 --- Experiment 2: Web Page Readability Analysis -Grade Level --- p.92 / Chapter 4.3.3 --- Experiment 3: Web Site Readability Analysis --- p.98 / Chapter 4.4 --- Summary and Future Work --- p.101 / Chapter 5 --- Conclusion --- p.104 / Chapter A --- List of Symbols and Notations --- p.107 / Chapter B --- List of Publications --- p.110 / Bibliography --- p.113

This thesis focuses on two important aspects of non-native reading comprehension: text readability assessment, which estimates the reading difficulty of a given text for L2 learners, and learner summarisation assessment, which evaluates the quality of learner summaries to assess their reading comprehension. We approach both tasks as supervised machine learning problems and present automated assessment systems that achieve state-of-the-art performance. We first address the task of text readability assessment for L2 learners. One of the major challenges for a data-driven approach to text readability assessment is the lack of significantly-sized level-annotated data aimed at L2 learners. We present a dataset of CEFR-graded texts tailored for L2 learners and look into a range of linguistic features affecting text readability. We compare the text readability measures for native and L2 learners and explore methods that make use of the more plentiful data aimed at native readers to help improve L2 readability assessment. We then present a summarisation task for evaluating non-native reading comprehension and demonstrate an automated summarisation assessment system aimed at evaluating the quality of learner summaries. We propose three novel machine learning approaches to assessing learner summaries. In the first approach, we examine using several NLP techniques to extract features to measure the content similarity between the reading passage and the summary. In the second approach, we calculate a similarity matrix and apply a convolutional neural network (CNN) model to assess the summary quality using the similarity matrix. In the third approach, we build an end-to-end summarisation assessment model using recurrent neural networks (RNNs). Further, we combine the three approaches to a single system using a parallel ensemble modelling technique. We show that our models outperform traditional approaches that rely on exact word match on the task and that our best model produces quality assessments close to professional examiners.

With this work I wanted to examine how to improve readability of battlefields in top-down gamesfor VR by increasing the visual distinction between units. I analyzed three successful PC gameswithin the genre and created three different user tests to come up with my results and conclusions.The results from my VR test is compared to my non-VR tests to see what one has to do to reachthe same level of readability or better in VR. It turns out you need to put more time in to yourassets for improving readability in VR than without VR on a non-stereo display.

Accounting as a tool for communicating relevant information to multiple users must overcome issues and barriers of presenting such information. One of the issues is its understandability, once lack of understanding can impact decisions. However, understandability is not easily measured. Previous studies use readability metrics as proxies for understandability. They understand that these concepts are related to some extent, but while readability focuses on the text, understandability includes the reader characteristics. Thus, the main purpose of this research was to verify how readability impacts the understandability of the information from the notes to the financial statements. I also verified whether the distance between them was due to reporting language. The specific goals were: (i) to verify whether different readability levels have impact on the notes to the financial statements\' understandability; (ii) to verify whether the understandability levels are due to accounting specific terms; (iii) to verify if those notes\' understandability and readability are different for different languages; and (iv) to verify if the levels of readability and understandability changed after OCPC 07. I performed two different types of research methods, archival and experimental research. The archival research applied mean tests to 44 firms that presented their notes both in Portuguese and in English from 2012 to 2015. I studied the notes of Financial Instruments and Provisions. I used indexes of intelligibility as proxies to understandability, calculated by a linguistic software. I found that, for most indexes where firms had worse levels of readability, they showed better levels of intelligibility, which goes against the expected. My results also show that with few exceptions, the readability and intelligibility indexes are different for Portuguese and English. Moreover, I found that there was no change after the adoption of OCPC 07 on both readability and intelligibility indexes. In the experimental research, I applied a self-created instrument based on the MIT test to 112 students in Brazil of graduate and undergraduate courses related to business, to verify how readability impacts on understandability, by manipulating readability components of the Flesch index (number of words per sentence and number of syllables per word). I controlled for some characteristics of each participant and the technicality of accounting language. I used an ANCOVA to test the difference between treatments and a Mediation Analysis to verify the impact of the glossary. I found no difference in understanding through different levels of readability, while some participant\'s characteristics - such as previous understanding skills, level of education, courses and the type of school - showed relevance in explaining understandability. The evidence also suggests that using a glossary to assist the reader affected understandability, but mostly for those who have less prior knowledge regarding notes to the financial statements. These results allow me to conclude readability metrics are not good proxies to measure understandability in Accounting, glossary assists on reader\'s understandability, language has an impact over readability and understandability, and readability and understandability did not change with the OCPC 07. / A Contabilidade como ferramenta para comunicar informação relevante a múltiplos usuários, deve superar problemas e barreiras na apresentação dessa informação. Um dos problemas é sua compreensibilidade, dado que a falta de compreensão pode impactar as decisões. Contudo, compreensibilidade não é mensurada facilmente. Estudos anteriores usam métricas de legibilidade como proxies de compreensibilidade. Eles entendem que esses conceitos estão relacionados em certa medida, mas enquanto legibilidade foca no texto, compreensibilidade inclui características do leitor. Assim, o objetivo principal dessa pesquisa foi verificar como a legibilidade impacta na compreensibilidade das informações das Notas Explicativas. Foi verificado também se a distância entre elas se deve à língua do relatório. Os objetivos específicos eram: (i) verificar se diferentes níveis de legibilidade impactam a compreensibilidade das notas explicativas; (ii) verificar se os níveis de compreensibilidade dependem de termos específicos contábeis; (iii) verificar se a compreensibilidade e a legibilidade das notas são diferente em diferentes línguas; e (iv) verificar se a compreensibilidade e a legibilidade das notas mudaram depois da OCPC 07. Realizaram-se dois métodos de pesquisas diferentes, um de arquivo e outro experimental. A pesquisa de arquivo aplicou testes de média em 44 empresas abertas que apresentaram suas notas em português e em inglês de 2012 a 2015. Foram estudadas notas de Instrumentos Financeiros e Provisões. Utilizaram-se índices de inteligibilidade como proxies para compreensibilidade, calculados por um software linguístico. Os resultados mostram que a maior parte das empresas que possuíam os piores níveis de legibilidade, apresentaram os melhores níveis de inteligibilidade, o que é contrário ao esperado. Os resultados também mostram que, com algumas exceções, os índices de legibilidade e compreensibilidade são diferentes em português e inglês. Ademais, descobriu-se que não houve mudança após a adoção da OCPC 07 em ambos os índices de legibilidade e compreensibilidade. No experimento, foi aplicado um instrumento criado pela autora, baseado no teste MIT, a 112 estudantes no Brasil de cursos relacionados a negócios de graduação e pós-graduação, para verificar como a legibilidade impacta na compreensibilidade, manipulando os componentes de legibilidade do índice Flesch (número de palavras por frase e número de sílabas por palavra). Controlou-se algumas características de cada participante e pela tecnicidade da linguagem contábil. Usou-se a ANCOVA para testar a diferença entre tratamentos e a Análise de Mediação para verificar o impacto de um glossário. Não foram encontradas diferenças de compreensão entre os diferentes níveis de legibilidade, enquanto que algumas características dos participantes - como habilidades de compreensão prévias, nível educacional, curso e tipo de universidade - se mostraram relevantes para explicar compreensibilidade. Os resultados também sugerem que o uso de um glossário para ajudar o leitor afetou compreensibilidade, mas mais para aqueles que possuíam menor conhecimento prévio sobre Notas Explicativas. Os resultados permitem concluir que as métricas de legibilidade não são boas proxies para mensurar compreensibilidade em Contabilidade, o glossário auxilia na compreensibilidade do leitor, a língua impacta legibilidade e compreensibilidade e legibilidade e compreensibilidade não mudaram com a OCPC 07.

Creating easy-to-read text is an issue that has traditionally been solved with manual work. But with advancing research in natural language processing, automatic systems for text simplification are being developed. These systems often need training data that is parallel aligned. For several years, simple Wikipedia has been the main source for this data. In the current study, several readability measures has been tested on a popular simplification corpus. A selection of guidelines from simple Wikipedia has also been operationalized and tested. The results imply that the following of guidelines are not greater in simple Wikipedia than in standard Wikipedia. There are however differences in the readability measures. The syntactical structures of simple Wikipedia seems to be less complex than those of standard Wikipedia. A continuation of this study would be to examine other readability measures and evaluate the guidelines not covered within the current work.

<p><strong><strong>Uppsatsens titel:</strong></strong></p><p>Lättläst – för alla?</p><p><strong>Key words: </strong></p><p>Easy-to-Read (Lättläst)<em>, </em>Printed matter (Trycksak)<em>, </em>Readability (Läsbarhet)<em>, </em>Legibility (Läslighet), Reading value (Läsvärde).</p><p><strong>Syfte: </strong></p><p>Vårt syfte med examensarbetet var att med text och layout framställa ett lättläst material, i tryckt form, anpassat till målgruppen.</p><p><strong>Frågeställningar:</strong></p><p>Går det att framställa ett lättläst och tydligt informationsmaterial som fungerar för alla?</p><p>Om det inte går att framställa ett informationsmaterial som fungerar för alla, behövs i stället olika versioner för att alla ska förstå? I så fall hur många versioner?</p><p>Hur ska materialet anpassas till målgruppen när det gäller språkutformning och layout?</p><p><strong>Teori:</strong></p><p> </p><p>Vårt arbetssätt följde den hermeneutiska spiralen. Vi gick från förförståelse och tolkning genom litteratursökning, tidigare forskning och analysering av material och målgrupp, till erfarenheter genom utprovningar. Därefter reviderade vi materialet och vidgade förståelsen genom nya utprovningar och fick återigen ny erfarenhet. Vi fick på så sätt påtaglig kunskap om målgruppens krav och behov.</p><p><strong>Metod:</strong><strong> </strong></p><p>Vi valde att arbeta både kvantitativt och kvalitativt, genom att utföra enkätundersökning samt intervju. Vi använde oss av båda tillvägagångssätten eftersom de kompletterade varandra. De metoder som vi använde oss av var litteratursökning, analysering av material samt målgruppsanalys (intervju- och enkätmetodik). Vi arbetade efter en iterativ metod, det vill säga fortlöpande utprovningar. Under utprovningarna använde vi "tänka högt"-metoden, prestationstest samt synpunkter från experter.</p><p><strong><strong>Slutsats:</strong></strong></p><p>Vår slutsats var att det gick att framställa ett lättläst och tydligt informationsmaterial som fungerade för alla i vår målgrupp, oavsett ålder, kön och bakgrund. Genom utprovningarna framkom det att samtliga av testpersonerna föredrog vår lättlästa version och ingen ansåg att texten var för lätt. Tvärtom var alla positiva till förändringarna i materialet eftersom de förstod informationen och tyckte att innehållet var relevant. Vi skapade ett lättläst material genom att till exempel använda konkreta ord, korta meningar utan bisatser, ordförklaringar, luftig layout samt överskådligt innehåll med tydliga och kortfattade rubriker.</p><p> </p>

<p>Is it possible to trace differences in the syntax used in various newspapers and how these differences influence the readability? Earlier studies confirm this and show that it is possible to make a wider distinction between the languages used in for example a broadsheet compared to the language used in a tabloid. In this study, both sentence length and sentence complexity of a broadsheet and a tabloid with a similar political stance were examined in order to find out if it is possible to show differences in readability between the two newspapers.</p><p>The articles used in this study are on-line articles and have thus been taken from a search on the internet. In order to obtain adequate research material, ten articles from each newspaper have been used. Five articles from each newspaper website are news articles whereas the remaining five were taken from the culture pages. Regarding sentence length, the average of each article has been calculated. When it comes to sentence complexity, ratios of simple, complex, and compound sentences have been investigated.</p><p>The analysis revealed that it is possible to show that there are not any substantial differences in sentence length and sentence complexity between the examined newspapers. However, in contrast to the hypothesis of this study, the examined articles in the tabloid consisted of longer sentences and more complex sentence constructions which, according to earlier research, would be an indication of a more formal language which probably has an effect on readability. Since both examined newspapers are supposed to support the Conservative party, it is, with the result of this study, possible to claim that both newspapers have the same targeted audience.</p>

<p>The use for a readability classification model is mainly as an integrated part of an information retrieval system. By matching the user's demands of readability to the documents with the corresponding readability, the classification model can further improve the results of, for example, a search engine. This thesis presents a new solution for classification into readability levels for Swedish. The results from the thesis are a number of classification models. The models were induced by training a Support Vector Machines classifier on features that are established by previous research as good measurements of readability. The features were extracted from a corpus annotated with three readability levels. Natural Language Processing tools for tagging and parsing were used to analyze the corpus and enable the extraction of the features from the corpus. Empirical testings of different feature combinations were performed to optimize the classification model. The classification models render a good and stable classification. The best model obtained a precision score of 90.21\% and a recall score of 89.56\% on the test-set, which is equal to a F-score of 89.88.</p> / <p>Uppsatsen beskriver utvecklandet av en klassificeringsmodell för Svenska texter beroende på dess läsbarhet. Användningsområdet för en läsbaretsklassificeringsmodell är främst inom informationssökningssystem. Modellen kan öka träffsäkerheten på de dokument som anses relevanta av en sökmotor genom att matcha användarens krav på läsbarhet med de indexerade dokumentens läsbarhet. Resultatet av uppsatsen är ett antal modeller för klassificering av text beroende på läsbarhet. Modellerna har tagits fram genom att träna upp en Support Vector Machines klassificerare, på ett antal särdrag som av tidigare forskning har fastslagits vara goda mått på läsbarhet. Särdragen extraherades från en korpus som är annoterad med tre läsbarhetsnivåer. Språkteknologiska verktyg för taggning och parsning användes för att möjliggöra extraktionen av särdragen. Särdragen utvärderades empiriskt i olika särdragskombinationer för att optimera modellerna. Modellerna testades och utvärderades med goda resultat. Den bästa modellen hade en precision på 90,21 och en recall på 89,56, detta ger en F-score som är 89,88. Uppsatsen presenterar förslag på vidareutveckling samt potentiella användningsområden.</p>

<p>The thesis presents findings from five readability studies performed on mobile devices. The dynamic Rapid Serial Visual Presentation (RSVP) format has been enhanced with regard to linguistic adaptation and segmentation as well as eye movement modeling. The novel formats have been evaluated against other common presentation formats including Paging, Scrolling, and Leading in latin-square balanced repeated-measurement studies with 12-16 subjects. Apart from monitoring Reading speed, Comprehension, and Task load (NASA-TLX), Eye movement tracking has been used to learn more about how the text presentation affects reading.</p><p>The Page format generally offered best readability. Reading on a mobile phone decreased reading speed by 10% compared to reading on a Personal Digital Assistant (PDA), an interesting finding given that the display area of the mobile phone was 50% smaller. Scrolling, the most commonly used presentation format on mobile devices today, proved inferior to both Paging and RSVP. Leading, the most widely known dynamic format, caused very unnatural eye movements for reading. This seems to have increased task load, but not affected reading speed to a similar extent. The RSVP format displaying one word at time was found to reduce eye movements significantly, but contrary to common claims, this resulted in decreased reading speed and increased task load. In the last study, Predictive Text Presentation (PTP) was introduced. The format is based on RSVP and combines linguistic chunking and adaptation with eye movement modeling to achieve a reading experience that can rival traditional text presentation.</p><p>It is explained why readability on mobile devices is important, how it may be evaluated in an efficient and yet reliable manner, and PTP is pinpointed as the format with greatest potential for improvement. The methodology used in the evaluations and the shortcomings of the studies are discussed. Finally, a hyper-graeco-latin-square experimental design is proposed for future evaluations.</p>

The thesis presents findings from five readability studies performed on mobile devices. The dynamic Rapid Serial Visual Presentation (RSVP) format has been enhanced with regard to linguistic adaptation and segmentation as well as eye movement modeling. The novel formats have been evaluated against other common presentation formats including Paging, Scrolling, and Leading in latin-square balanced repeated-measurement studies with 12-16 subjects. Apart from monitoring Reading speed, Comprehension, and Task load (NASA-TLX), Eye movement tracking has been used to learn more about how the text presentation affects reading. The Page format generally offered best readability. Reading on a mobile phone decreased reading speed by 10% compared to reading on a Personal Digital Assistant (PDA), an interesting finding given that the display area of the mobile phone was 50% smaller. Scrolling, the most commonly used presentation format on mobile devices today, proved inferior to both Paging and RSVP. Leading, the most widely known dynamic format, caused very unnatural eye movements for reading. This seems to have increased task load, but not affected reading speed to a similar extent. The RSVP format displaying one word at time was found to reduce eye movements significantly, but contrary to common claims, this resulted in decreased reading speed and increased task load. In the last study, Predictive Text Presentation (PTP) was introduced. The format is based on RSVP and combines linguistic chunking and adaptation with eye movement modeling to achieve a reading experience that can rival traditional text presentation. It is explained why readability on mobile devices is important, how it may be evaluated in an efficient and yet reliable manner, and PTP is pinpointed as the format with greatest potential for improvement. The methodology used in the evaluations and the shortcomings of the studies are discussed. Finally, a hyper-graeco-latin-square experimental design is proposed for future evaluations.