Information Retrieval from Dutch Historical Corpora

Preface Writing a thesis is often regarded as rather solitary labour. During my research I learned that the opposite is true, since writing this thesis would not have been possible without the support of several people who I would like to acknowledge in this preface. First and foremost I would like to express my gratitude to the members of my thesis committee. I would like to thank Prof. dr. H.J. van den Herik, the chair of my thesis committee, both for his valuable comments on my thesis and especially for stimulating my interest in scientific research. I acknowledge Prof. dr. A.M.J.A. Berkvens for arranging contacts with the domain experts, for commenting on my thesis, and for his interest in my research. I am sincerely grateful to Dr. F.J. Wiesman, my daily advisor, for his guidance and enthusiastic support during the writing of my thesis. Without his valuable comments my thesis would not have been written in its present form. I would like to thank Dr. I.G. Sprinkhuizen-Kuyper for answering many of my questions, especially on mathematics and statistics, and for commenting on my thesis. Moreover, I am greatly indebted to the domain experts, Prof. dr. who participated in the various experiments which were part of my research. Without their cooperation, the research could not have been conducted as it was. I am grateful to my parents for their continuous support and interest in my education, especially during the writing of this thesis. They were the first to teach me motivation, self-discipline and aiming at perfection; all these characteristics I tried to pursue when writing this thesis. Finally, I would like to give special thanks to Ron for his support and encouragement. Although our interests lie in different scientific disciplines, we had many interesting discussions which greatly inspired me during my research. i ii PREFACE Summary The thesis focuses on a separate area in the field of information retrieval: information retrieval from historical corpora. The goal of the research is to identify the bottlenecks of information retrieval from historical corpora and to solve these bottlenecks. A collection of Dutch and Belgian law texts from the 16th and 17th century serves as test corpus. The research consists of three phases: bottleneck identification, solution determination and system design, and system evaluation and comparison. In the first phase (bottleneck identification), an experiment is conducted to identify the bottlenecks of information retrieval from historical corpora. …