Text Complexity and Text Simplification in the Crisis Management Domain

Due to the fact that emergency situations can lead to substantial losses, both financial and in terms of human lives, it is essential that texts used in a crisis situation be clearly understandable. This thesis is concerned with the study of the complexity of the crisis management sub-language and with methods to produce new, clear texts and to rewrite pre-existing crisis management documents which are too complex to be understood. By doing this, this interdisciplinary study makes several contributions to the crisis management field. First, it contributes to the knowledge of the complexity of the texts used in the domain, by analysing the presence of a set of written language complexity issues derived from the psycholinguistic literature in a novel corpus of crisis management documents. Second, since the text complexity analysis shows that crisis management documents indeed exhibit high numbers of text complexity issues, the thesis adapts to the English language controlled language writing guidelines which, when applied to the crisis management language, reduce its complexity and ambiguity, leading to clear text documents. Third, since low quality of communication can have fatal consequences in emergency situations, the proposed controlled language guidelines and a set of texts which were re-written according to them are evaluated from multiple points of view. In order to achieve that, the thesis both applies existing evaluation approaches and develops new methods which are more appropriate for the task. These are used in two evaluation experiments – evaluation on extrinsic tasks and evaluation of users’ acceptability. The evaluations on extrinsic tasks (evaluating the impact of the controlled language on text complexity, reading comprehension under stress, manual translation, and machine translation tasks) 4 Text Complexity and Text Simplification in the Crisis Management domain show a positive impact of the controlled language on simplified documents and thus ensure the quality of the resource. The evaluation of users’ acceptability contributes additional findings about manual simplification and helps to determine directions for future implementation. The thesis also gives insight into reading comprehension, machine translation, and cross-language adaptability, and provides original contributions to machine translation, controlled languages, and natural language generation evaluation techniques, which make it valuable for several scientific fields, including Linguistics, Psycholinguistics, and a number of different sub-fields of NLP. Text Complexity and Text Simplification in the Crisis Management domain 5 Dedication and Acknowledgements The greatest good you can do for another is not just to share your riches but to reveal to him his own. (Benjamin Disraeli) This thesis is dedicated to my parents, the nuclear physicists Prof. Dr. Galina Maneva and Prof. Dr. Petar Temnikov, to whom I am deeply grateful for having inspired me to enter the scientific profession and for their extensive moral and scientific support in times that were difficult for me. I gratefully acknowledge the support of many people who were important to me in the conduct of the research reported in this thesis. Their names and the reasons that I am grateful to them follow below. First of all, I would like to express my gratitude to my Director of Studies, Prof. Dr. Ruslan Mitkov, for having allowed me to fulfil my dream of working in the computational linguistics field, for having exposed me to extensive computational linguistics knowledge thanks to my work for the Journal of Natural Language Engineering and the active scientific environment in his group, and for providing precious comments regarding my final thesis. Secondly, I would like to express my gratefulness to the other members of my supervisory team—Dr. Le An Ha and Mr. Richard Evans— for their constant availability and for giving me valuable lessons in structuring and clearly expressing the ideas in my thesis. Their contributions to my research were invaluable. 6 Text Complexity and Text Simplification in the Crisis Management domain Next, a very big “thank you!” goes to all of the great, experienced researchers who helped me with lots of useful discussions and clever ideas during the course of my thesis, and whom I consider as being my unofficial supervisors, such as: Dr. Constantin Orasan, Dr. Anke Buttner, Prof. Dr. Albena Vassileva, Dr. Lucia Specia, and Dr. Sobha Lalitha Devi. Their input was essential, because of the wide interdisciplinary character of my thesis. Among them, I would like to express my special gratitude to Prof. Dr. Albena Vassileva, who created a masters course in Computational Linguistics in the French philology department in Sofia University and made it accessible to other fields, by this allowing me to come to know the wonderful field of Computational Linguistics. Without her, this thesis would not be possible. She, and the other professors (like Dr. Kiril Simov, Dr. Petya Osenova, Dr. Svetla Boycheva and Dr. Denitsa Daynovska) leading that course, made me fall in love with this discipline. I would like also to express a big “thank you” to Prof. Dr. Galia Angelova, for having given me the precious opportunity to come into close touch with the field of Computational Linguistics by involving me in one of the most prestigious NLP conferences—RANLP—and for having shown me support during the whole of my Ph.D. studies. Special thanks go to Dr. Ralf Steinberger for having given me a strong push in the beginning of my career as a computational linguistics trainee, and to Dr. Katerina Gachevska, who gave me insights regarding how to organize the effective writing of my thesis and many other useful things. I would also like to express my enormous gratitude to a very special person, Dr. Kevin Bretonnel Cohen, for having had the patience and perseverance to read, proofread, and provide useful scientific comments about all of the chapters of my dissertation and for having urged me not to give Text Complexity and Text Simplification in the Crisis Management domain 7 up and to keep writing. I would also like to thank Mrs. Alison Carminke and Mrs. Erin Stokes for proofreading several of my papers that contributed to this thesis. Additionally, I would like to thank very much the members of my research group for their precious ideas before my viva, and particularly specific colleagues and fellow Ph.D. students for their availability and their support during the course of my thesis, such as Iustina Ilisei, Iustin Dornescu, Andrea Varga, Dr. Raphael Rubino (in particular for his programming help with Python scripts), Ivelina Nikolova, Natalia Konstantinova (for her support in the Journal of Natural Language Engineering work), Dr. Georgiana Puscasu, and Dr. Laura Hasler. Special thanks also go to Ms. Ruslana Margova, who always replied promptly and with enthusiasm to ideas for our shared research. I would also like to mention some important people from my past: my grandmother, Dr. Cvetana Maneva; my grandfather, Mihail Manev; my elementary school teacher, Marina Genina, and my favorite high school teacher, Prof. Tonina Fancello, for having shown me the nice sides of knowledge and for inspiring me to follow that light. Last but not least, I would like to thank my exboyfriend Mihail Milushev, for having helped me fight my fear of programming by showing me how interesting Perl is and for having taught me to program in Perl. And finally, I would like to say a big “thank you” to all of the people who were close to me in the last years (among which my friends and my brother) for the patience, understanding, friendship, and support that they have given me. They urged me not to give up and to “keep pushing” for all these years. You know who you are. THANK YOU! 8 Text Complexity and Text Simplification in the Crisis Management domain Text Complexity and Text Simplification in the Crisis Management domain 9