◆ In the following descriptions, the letter following each “Task” (e.g., Task A) corresponds to the codes (A–Q) listed under “Outsourcing task(s)” in the Contact Us form.
Information collection must be conducted continuously throughout all stages of a project — from planning and execution to completion and evaluation.
Typically, project members (experts, researchers, technical staff) gather documents and data from online resources and conduct literature surveys by carefully reviewing scholarly publications.
In particular, since scholarly articles include knowledge-based information (insights) obtained by the authors, conducting literature surveys to obtain the latest information relevant to research and development projects is an essential task.
However, due to the vast amount of information that must be reviewed, projects often require significant time, human resources, and costs for information collection, making it one of the major bottlenecks in project implementation.
With advances in computing and analytical techniques, Natural Language Processing (NLP) technologies have become available, enabling computers to process and comprehend text.
However, many commonly used NLP tools, while capable of processing large volumes of documents at high speed, still face issues with accuracy and output formats.
For example, they are unable to extract summarized information on key terms, merely outputting the entire sentences containing the key terms, making it difficult for users to instantly grasp the relevant information.
To address this limitation, WGI has developed its proprietary AI-powered text mining technology (A2K technology) to efficiently and accurately collect gene functional information.
The table below shows a comparison between manual reviews and A2K/LA2K technologies.
From this comparison, it is clear that utilizing A2K/LA2K analyses can significantly enhance project performance.
Moreover, applying A2K/LA2K to large-scale literature datasets enables rapid, accurate, and cost-effective generation of summaries and statistical results (e.g., frequent terms, co-occurring terms) related to key information.
This not only accelerates the understanding of the latest information but also enables users to easily identify key documents (such as essential literature sources) that should be carefully read by experts.
