Canadian Partnership Against Cancer

CPAC

Canadian Partnership Against Cancer

CPAC

CPAC is Canada’s central source of evidence-based information on cancer-related issues. The site shares and disseminates relevant knowledge from public sources, institutions and policies focused on cancer and chronic disease prevention. The Prevention Policies Directory helps specialists and decision makers navigate prevention policies in relation to e.g., risk factors and local legislation.

Challenge

It required considerable resources to manually keep track of current updates of the many thousands of documents from hundreds of different sources. Therefore, CPAC invited 3 agencies to bid on automating the process. CPH Cloud won the tender and the task of automating data collection and rationalizing the manual processes for maintaining The Prevention Policies Directory.

Solution

The solution covers three main areas. Collection of documents, validation of data and dissemination of content. The first two have been automated and thus optimized and simplified but the latter, still involves a manual effort from CPAC’s experts.

The approximately 300 selected sources are monitored by pre-programmed robots that collect new releases and updated documents. The robots deliver the collected data to our ICE (Intelligent Content Extraction) system, which checks and evaluates the collected data for validity, relevance, repetitions, etc. before they are passed on to the last manual process

Result

CPAC’s Prevention Policies Directory has been very successfully serviced by our robots and ICE identifies and collects approx. 1,100 documents a year, which would otherwise have required manual research efforts, and updates could very well have been unidentified in a manual process.

Our deliveries

Strategy & solution design

Configuration and maintenance of Dexi scraber robots

Development and operation of Intelligent Content Extraction setup (ICE)

Machine Learning project