Information Analysis and Retrieval
Scientific Literature Mining (LitMiner)
How can scientific researchers hope to know about all of the latest advances and new discoveries in their field, given that more than 40,000 scholarly articles are published in the scientific literature every month? How can they be sure of finding all of the relevant knowledge “hidden” in journal articles?
Even with the advent of massive numerical and structural databases, the scientific literature still holds the newest information and the intelligence surrounding the data. The problem is that researchers cannot hope to read all the articles relevant to their field of study if they are also to conduct research.
In response to this pressing challenge, the NRC Institute for Information Technology (NRC-IIT), in collaboration with the NRC Institute for Biological Sciences (NRC-IBS), the Canada Institute for Scientific and Technical Information (CISTI), the Samuel Lunenfeld Institute and Blueprint International, is developing a unified collection of text and language processing tools to solve the real information needs of genomic and proteomic scientists.
In the short-term, the goal is to save researchers time by letting computers assume some of the tasks. In the longer term, the goal is to support hypothesis formation in ways that are not possible with the current organization of the literature.
The LitMiner project is currently in its first stage, which is to integrate several existing text tools into a proof-of-concept prototype. More elaborate scenarios of use will be possible from that prototype, which will ultimately lead to more useful systems.
The research is being conducted in both in text processing and bio-informatics. Most of the tools being combined in LitMiner are machine learning, information retrieval or text mining algorithms, either new or based on novel modifications of existing algorithms.
While the application to the scientific literature is driving the further development of these algorithms, the research is important in its own right.
