Jia-Wen Liu

Learn More
Many of the existing machine-learning approaches to speech summarization cast important sentence selection as a two-class classification problem and have shown empirical success for a wide variety of summarization tasks. However, the imbalanced-data problem sometimes results in a trained speech summarizer with unsatisfactory performance. On the other hand,(More)
The purpose of extractive speech summarization is to automatically select a number of indicative sentences or paragraphs (or audio segments) from the original spoken document according to a target summarization ratio and then concatenate them to form a concise summary. Much work on extractive summarization has been initiated for developing machine-learning(More)
Aziridine-containing compounds have been of interest as anticancer agents since late 1970s. The design, synthesis and study of triaziquone (TZQ) analogues with the aim of obtaining compounds with enhanced efficacy and reduced toxicity are an ongoing research effort in our group. A series of bis-type TZQ derivatives has been prepared and their cytotoxic(More)
We aimed to evaluate the interference of 50 Hz extremely low frequency electromagnetic field (ELF-EMF) occupational exposure on the neurobehavior tests of workers performing tour-inspection close to transformers and distribution power lines. Occupational short-term “spot” measurements were carried out. 310 inspection workers and 300 logistics staff were(More)
The trajectory of incoming missile and shrapnel are analyzed to make sure that the incoming missile would be within the distribute range of bullets dispersed by the shrapnel. In the case of the center of bullet group hits the target, the miss distance is optimized with the angle of fire and direction angle as the optimization variables in MATLAB. An(More)
  • 1