How do we gather our data
With millions of scientific papers already published, gathering all the facts about medical treatments is a complex and expensive task. To tackle this task, we use automated process to gain and edit our databases.
Some of these processes include tools for extracting specific data features about treatments (for example – brand names, medical condition definitions, and when was a treatment first released to the market) from open online resources.
However, the main process, that yields the most important data in CureFacts database, involves Natural Processing Language (NLP). This proprietary semi-automatic NLP tool enables us to extract and edit information about the evidence, effectiveness and safety of various medical treatment – from thousands of systematic reviews. It is done in a semi-automatic fashion, after which the results (data and texts) are reviewed and edited by experts. This way, we can assure that the gathered information is precise and clear.
The use of NLP technology has many advantages. NLP speeds up the creation of our databases and saves costs and efforts. It also brings uniformity to the extraction process, which combined with a manual (human) inspection increases the reliability of the information that CureFacts provides. In addition, since the NLP process is designed and calibrated specifically for CureFacts unique needs, it adds to our intellectual property.