As we spend more and more time on the Internet, engaging in activities from emailing, to typing up documents, to uploading images, all of the data that embodies who we are as an individual becomes lost in a sea of unorganized information. In attempt to organize it, Microsoft has built “Lifebrowser,” a prototype curation software that mines through your personal timeline of online activities and displays it a digestible and searchable timeline.
The demand for curating the massive volume of a company’s data, called “big data,” is increasing among the likes of Apple, Google and Facebook. When a company’s data overwhelms its in-house management tools, it usually takes it to a big data firm. The big data company will sift through the exorbitant information and create the necessary organizational structure and tools to make sense of the noise. Lifebrowser is a scaled down and personal solution.
You can think of it as a smart search engine of your virtual proclivities. Lifebrowser’s machine-learning algorithm will crawl through all of your online and offline activities, find only what it discerns as significant events in your life, and save the information in chronological order for later perusal.
For example, determining the significance of a photograph on your life goes beyond identifying the file name and creation date. According to MIT’s Technology Review, a machine-vision algorithm can discern the number of individuals within a photograph and ascertain whether the picture was captured indoors or outdoors. The algorithm, in determining the file’s significance, will go as far as to factor in the number of photographs taken within a session.
To determine the relationship between you and the recipient of an email, Lifebrowser will crawl through your calendar, where it may find a meeting date, time and location to appropriately categorize an email.
When the software is stumped, it will prompt you to simply decide whether a file is or isn’t significant. Over time, the algorithm will adapt to understand files that are of significance to you.
While an algorithm is decisively not human, Microsoft’s researcher Eric Horvitz believes algorithms can be personal. “You always think that machine learning is kind of cold,” Microsoft researcher, Horvitz, told Technology Review. “This is showing that a model is not only learning about how I think, it’s also very warmly understanding what it means to capture humanity.”
While we tend to ignore the happenstances of the past, there are moments when an email that you sent a month ago with an important attachment remains elusive, or you may want to relive a moment in time. Lifebrowser, using its sophisticated search engine, will allow users to search and quickly cut through the noise to bring up all relevant documents pertaining to your search query.
The design of the software was built to mimic how human beings recall information. New research by researchers from Indiana Unviersity and the University of Warwick has revealed that humans recall information in patches akin to the way bees forage for pollen. Horvitz hopes that its prototype algorithm and timeline design, in development even before Facebook’s founding, will be made available to the public. Is anyone else thinking Windows 9?
- Deep learning vs. machine learning: what's the difference between the two?
- Don’t be fooled by dystopian sci-fi stories: A.I. is becoming a force for good
- ‘Mind-reading’ A.I. produces a description of what you’re thinking about
- Can an algorithm be racist? Spotting systemic oppression in the age of Google
- Ubisoft’s ‘Sam’ A.I. wants to help you play video games