Hierarchical clustering of a Finnish newspaper article collection with graded relevance assessments

Search facilitated with agglomerative hierarchical clustering methods was studied in a collection of Finnish newspaper articles (N = 53,893). To allow quick experiments, clustering was applied to a sample (N = 5,000) that was reduced with principal components analysis. The dendrograms were heuristically cut to find an optimal partition, whose clusters were compared with each of the 30 queries to retrieve the best-matching cluster. The four-level relevance assessment was collapsed into a binary one by (A) considering all the relevant and (B) only the highly relevant documents relevant, respectively. Single linkage (SL) was the worst method. It created many tiny clusters, and, consequently, searches enabled with it had high precision and low recall. The complete linkage (CL), average linkage (AL), and Ward's methods (WM) returned reasonably-sized clusters typically of 18–32 documents. Their recall (A: 27–52%, B: 50–82%) and precision (A: 83–90%, B: 18–21%) was higher than and comparable to those of the SL clusters, respectively. The AL and WM clustering had 1–8% better effectiveness than nearest neighbor searching (NN), and SL and CL were 1–9% less efficient that NN. However, the differences were statistically insignificant. When evaluated with the liberal assessment A, the results suggest that the AL and WM clustering offer better retrieval ability than NN. Assessment B renders the AL and WM clustering better than NN, when recall is considered more important than precision. The results imply that collections in the highly inflectional and agglutinative languages, such as Finnish, may be clustered as the collections in English, provided that documents are appropriately preprocessed.     

Books are silent ambassadors: The Noma award for the translation of Japanese literature

Summary Now more than ever good translations are proving to be vitally important to fulfilling Shoichi Noma’s mission of advancing good relationships between cultures. With the emergence of dozens of world-class Japanese writers, the translators of Japanese literature have acquired a new significance and a new importance. No longer are they required to simply handle the language like an exotic and precious commodity but they must prepare a writer’s work to compete with literature from around the globe and let others understand the issues at their very heart of the Japanese people. Indeed, the selection committee’s choice of Murakami’s Wind-Up Bird Chronicle is an apt example of how literature at its very best can span cultural divides to promote mutual understanding of very complicated and important world issues. Written for Japanese to help them understand the devastation of war, it is a rare and timely book that confronts Japan’s experience head on. Ironically, Noma’s idea that books are silent ambassadors, an idea born from agony and defeat, so many years later is helping the world come together towards a unified vision of peace and understanding. “Literature has been described as a means of exchange between the hearts of people. Literary translation, then, is the building of bridges so that all peoples of the world can interact in this way.” Ye Weiqu, Chinese Institute of Social Studies.     

Analysis of Statistical Question Classification for Fact-Based Questions

Question classification systems play an important role in question answering systems and can be used in a wide range of other domains. The goal of question classification is to accurately assign labels to questions based on expected answer type. Most approaches in the past have relied on matching questions against hand-crafted rules. However, rules require laborious effort to create and often suffer from being too specific. Statistical question classification methods overcome these issues by employing machine learning techniques. We empirically show that a statistical approach is robust and achieves good performance on three diverse data sets with little or no hand tuning. Furthermore, we examine the role different syntactic and semantic features have on performance. We find that semantic features tend to increase performance more than purely syntactic features. Finally, we analyze common causes of misclassification error and provide insight into ways they may be overcome.