This thesis deals with text simplification in Czech, in particular with lexical simplification. Several strategies of complex word identification, substitution generation and substitution ranking are implemented and evaluated. Substitution generation is attempted both in a dictionary-based manner and in an embedding- based manner. Some experiments involving people are also presented, the experiments aim at gaining an in- sight into perceived simplicity/complexity and its factors. The experiments conducted and evaluated include sentence pair comparison and manual text simplification. Both the evaluation results of various strategies and the outcomes of experiments involving humans are described and some future work is suggested. 1

Introduction: This thesis was submitted by Matthew Shardlow to the University of Manchester for the degree of Doctor of Philosophy (PhD) in the year 2015. Lexical simplification is the practice of automatically increasing the readability and understandability of a text by identifying problematic vocabulary and substituting easy to understand synonyms. This work describes the research undertaken during the course of a 4-year PhD. We have focused on the pipeline of operations which string together to produce lexical simplifications. We have identified key areas for research and allowed our results to influence the direction of our research. We have suggested new methods and ideas where appropriate. Objectives: We seek to further the field of lexical simplification as an assistive technology. Although the concept of fully-automated error-free lexical simplification is some way off, we seek to bring this dream closer to reality. Technology is ubiquitous in our information-based society. Ever-increasingly we consume news, correspondence and literature through an electronic device. E-reading gives us the opportunity to intervene when a text is too difficult. Simplification can act as an augmentative communication tool for those who find a text is above their reading level. Texts which would otherwise go unread would become accessible via simplification. Contributions: This PhD has focused on the lexical simplification pipeline. We have identified common sources of errors as well as the detrimental effects of these errors. We have looked at techniques to mitigate the errors at each stage of the pipeline. We have created the CW Corpus, a resource for evaluating the task of identifying complex words. We have also compared machine learning strategies for identifying complex words. We propose a new preprocessing step which yields a significant increase in identification performance. We have also tackled the related fields of word sense disambiguation and substitution generation. We evaluate the current state of the field and make recommendations for best practice in lexical simplification. Finally, we focus our attention on evaluating the effect of lexical simplification on the reading ability of people with aphasia. We find that in our small-scale preliminary study, lexical simplification has a nega- tive effect, causing reading time to increase. We evaluate this result and use it to motivate further work into lexical simplification for people with aphasia.

In this study automatic lexical simplification via synonym replacement in Swedish was investigated using three different strategies for choosing alternative synonyms: based on word frequency, based on word length, and based on level of synonymy. These strategies were evaluated in terms of standardized readability metrics for Swedish, average word length, proportion of long words, and in relation to the ratio of errors (type A) and number of replacements. The effect of replacements on different genres of texts was also examined. The results show that replacement based on word frequency and word length can improve readability in terms of established metrics for Swedish texts for all genres but that the risk of introducing errors is high. Attempts were made at identifying criteria thresholds that would decrease the ratio of errors but no general thresholds could be identified. In a final experiment word frequency and level of synonymy were combined using predefined thresholds. When more than one word passed the thresholds word frequency or level of synonymy was prioritized. The strategy was significantly better than word frequency alone when looking at all texts and prioritizing level of synonymy. Both prioritizing frequency and level of synonymy were significantly better for the newspaper texts. The results indicate that synonym replacement on a one-to-one word level is very likely to produce errors. Automatic lexical simplification should therefore not be regarded a trivial task, which is too often the case in research literature. In order to evaluate the true quality of the texts it would be valuable to take into account the specific reader. A simplified text that contains some errors but which fails to appreciate subtle differences in terminology can still be very useful if the original text is too difficult to comprehend to the unassisted reader.

This thesis presents the development and evaluation of context-aware Lexical Simplification (LS) systems for the Swedish language. In total three versions of LS models, LäsBERT, LäsBERT-baseline, and LäsGPT, were created and evaluated on a newly constructed Swedish LS evaluation dataset. The LS systems demonstrated promising potential in aiding audiences with reading difficulties by providing context-aware word replacements. While there were areas for improvement, particularly in complex word identification, the systems showed agreement with human annotators on word replacements. The effects of fine-tuning a BERT model for substitution generation on easy-to-read texts were explored, indicating no significant difference in the number of replacements between fine-tuned and non-fine-tuned versions. Both versions performed similarly in terms of synonymous and simplifying replacements, although the fine-tuned version exhibited slightly reduced performance compared to the baseline model. An important contribution of this thesis is the creation of an evaluation dataset for Lexical Simplification in Swedish. The dataset was automatically collected and manually annotated. Evaluators assessed the quality, coverage, and complexity of the dataset. Results showed that the dataset had high quality and a perceived good coverage. Although the complexity of the complex words was perceived to be low, the dataset provides a valuable resource for evaluating LS systems and advancing research in Swedish Lexical Simplification. Finally, a more transparent and reader-empowering approach to Lexical Simplification isproposed. This new approach embraces the challenges with contextual synonymy and reduces the number of failure points in the conventional LS pipeline, increasing the chancesof developing a fully meaning-preserving LS system. Links to different parts of the project can be found here: The Lexical Simplification dataset: https://github.com/emilgraichen/SwedishLSdataset The lexical simplification algorithm: https://github.com/emilgraichen/SwedishLexicalSimplifier

Electronic health records (EHRs) are used in Sweden's healthcare systems to store patients' medical information. Patients in Sweden have the right to access and read their health records. Unfortunately, the language used in EHRs is very complex and presents a challenge for readers who lack medical knowledge. Simplifying the language used in EHRs could facilitate the transfer of information between medical staff and patients. This project investigates the possibility of generating Swedish medical synonyms automatically. These synonyms are intended to be used in future systems for lexical simplification that can enhance the readability of Swedish EHRs and simplify medical terminology. Current publicly available Swedish corpora that provide synonyms for medical terminology are insufficient in size to be utilized in a system for lexical simplification. To overcome the obstacle of insufficient corpora, machine learning models are trained to generate synonyms and terms that convey medical concepts in a more understandable way. With the purpose of establishing a foundation for analyzing complex medical terms, a simple mechanism for Complex Word Identification (CWI) is implemented. The mechanism relies on matching strings and substrings from a pre-existing corpus containing hand-curated medical terms in Swedish. To find a suitable strategy for generating medical synonyms automatically, seven different machine learning models are queried for synonym suggestions for 50 complex sample terms. To explore the effect of different input data, we trained our models on different datasets with varying sizes. Three of the seven models are based on BERT and four of them are based on Word2Vec. For each model, results for the 50 complex sample terms are generated and raters with medical knowledge are asked to assess whether the automatically generated suggestions could be considered synonyms. The results vary between the different models and seem to be connected to the amount and quality of the data they have been trained on. Furthermore, the raters involved in judging the synonyms exhibit great disagreement, revealing the complexity and subjectivity of the task to find suitable and widely accepted medical synonyms. The method and models applied in this project do not succeed in creating a stable source of suitable synonyms. The chosen BERT approach based on Masked Language Modelling cannot reliably generate suitable synonyms due to the limitation of generating one term per synonym suggestion only. The Word2Vec models demonstrate some weaknesses due to the lack of context consideration. Despite the fact that the current performance of our models in generating automatic synonym suggestions is not entirely satisfactory, we have observed a promising number of accurate suggestions. This gives us reason to believe that with enhanced training and a larger amount of input data consisting of Swedish medical text, the models could be improved and eventually effectively applied.