‘More Data, More Problems’ or ‘More Data, More Possibilities?’

‘More Data, More Problems’ or ‘More Data, More Possibilities?’
May 28, 2015 Kristofer


With the release of the new NetClean Analyze DI feature Analyze Relations, Johann Hofmann, explains how letting novel tools aid investigators in finding relations between digital files and building visual maps is key to solving cases.

The human visual system performs all kinds of amazing jobs, from finding constellations in the night sky or picking out a person in a crowd, to making a rapid assessment of a room you just entered. How do our eyes and brains so easily assess a situation? How can we separate a friendly carnival from a riot by just looking at a picture?

Already as toddlers our brains are experts at making sense of what we see. As computer vision expert and Director of Stanford AI Lab, Fei Fei Li, explains, humans can look at an image and understand relations, emotions, actions and intentions. Meanwhile our most advanced computers still struggle to even recognize objects. To them photos are just a two dimensional collection of pixels.

However, unlike the human brain, computers have an enormous capacity to process and sort through large volumes of data. They can, for example, filter out the few photos with the exact same pattern among thousands of digital files. Or find all the photos taken at a certain time, date or place in a database with millions of images.

With the proliferation of digital media, investigators face the challenge of manually reviewing and analysing millions of digital files containing visual data. Although more digital files mean more potential evidence, it is time consuming, frustrating and nearly impossible to make sense of such large quantities of information.

However, technology can help our visual capabilities along the way, by separating pertinent from non-pertinent material, actively making connection between millions of images and helping our brains to find patterns and clues. Therefore thelatest development of Analyze DI is focused on actively finding and suggesting connections between images, and building visual maps. As the user throws a suspect image to Analyze Relations and navigates his or her way through multiple iterations of suggested relations, new leads are revealed.

The result is not just superior investigations, but also a constantly growing knowledge base. Each clue revealed can also be a crucial piece of the puzzle in another investigation. Our belief and hope is that with this feature more data will not mean more problems, but instead more possibilities.

Read more about Analyze DI 15.1 and Analyze Relations here.

Watch the video “Beyond the grid”, with the vision of Analyze Relations.