| Summary: | 碩士 === 國立成功大學 === 資訊工程學系碩博士班 === 95 === With the decoding of human genome sequences, a large amount of research about the genetic sequence data has been published. The improved biological technology, such as DNA microarrays and the high-throughput experiments for molecular biology, has produced a huge scale of data about gene information and such abundant knowledge is almost recorded in the plaintext format. Due to the rapid growing rate of biomedical knowledge, the concern of effective information retrieval has arisen. For biomedical scientists, it’s an important issue to understand the transcriptional regulation of genes under different conditions. The major challenge comes from the complexity of the gene regulatory systems.
The biomedical literatures contain a huge amount of gene-related data, including the transcriptional regulation between the regulators and the target genes. Although the biomedical literatures and papers contain rich resources, it takes lots of efforts for the researchers to obtain the relationships from the tremendous literatures. Therefore, efficient processing of the large scale of resources is needed. Our goal is to develop an intelligent gene regulatory mining system for extracting the informative sentences containing the regulation relationships from biomedical literatures. Our experiments show that the derived system attains a stable and prominent performance and is useful to extract the gene regulatory information from literatures.
|
|---|
