The quality of question classification is vital for a practical question-answering system. This paper proposes a transfer learning method based on generating virtual data for zero-shot questions. The ...basic idea is to exploit the commonality and difference between zero annotated questions and large enough annotated questions to generate virtual training data for zero annotated questions, thereby relieving the problem of data imbalance and improving performance of question classifier. Concretely, we first apply a template-based generator to generate basic virtual samples, then use them to train an encoder-decoder based generator to generate large enough virtual data. Finally, the real samples and virtual ones are used to train a supervised question classifier. Experiments show that the proposed method improves the overall classification performance both for English and Chinese data sets. Especially, the classification performance of zero annotated questions increased significantly, from 7.46% to 59.34% for English and from 1.96% to 42.67% for Chinese, and the generated virtual data has minute impact on the performance of large annotated question test set.
Although the underlying mechanism is not well understood, there is considerable evidence that the constellation of cognitive factors known as 'spatial aptitude' influences users' performance in ...information spaces. Evidence of the effect in the computer science literature is contradictory: some studies show that techniques, which support users with lower aptitude, retard performance by those with higher aptitude. We have investigated the effect of the visualization subfactor in a real-world navigation task using location menu breadcrumbs and Dillon's IMRD task. We compared the navigational styles and success rates in an answer seeking task using both standard and menu breadcrumbs in a large website. The higher aptitude group was significantly more efficient and used the Back button less than the lower aptitude group. We discuss implications for explaining why spatial aptitude affects success with hypertext, the potential for practical application, and ongoing follow-up work.