Text mining: generating hypotheses from MEDLINE
Journal of the American Society for Information Science and Technology
Using statistical and knowledge-based approaches for literature-based discovery
Journal of Biomedical Informatics
| Hi-index | 0.00 |
The existing evidence shows that Chinese disjointed literature-based knowledge discovery is largely limited to traditional Chinese medical literature instead of the entire medical literature available in Chinese. Some limitations exist in current research when put it into practice, for many reasons, including the data all coming from a single source. During the process of knowledge discovery, the problem of excessively large intermediate and target concept sets still needs attention. This Chinese medical disjointed literature-based knowledge discovery system was constructed by applying research to the concepts of extraction, pruning and sorting algorithms in an endeavour to solve the existing problems associated with disjointed literature-based knowledge discovery systems. The system was tested and verified in accordance with the classic magnesium-migraine hypothesis of Dr Swanson. Future improvements and development of the system are also proposed.