How does Facebook know who is in the picture I shared?

Photo Credit: https://forbiddenplanet.com/1917-star-wars-kotobukiya-statue-c-3po-r2-d2/

Photo Credit: https://forbiddenplanet.com/1917-star-wars-kotobukiya-statue-c-3po-r2-d2/

Ever posted a photo to Facebook and wondered how names of all the people in the photo were accurately suggested? Or, perhaps you are trying to understand the technology behind cars that can drive themselves? I’ll let you in on a little secret…No, there isn’t a little elf deep within the depths of Facebook’s database, nor is there an actual C-3PO-looking robot that has been programmed to drive motorized vehicles. It’s Machine Learning, and it is transforming businesses within all industries.

Machine Learning is a term that has been gaining lots of buzz within the past few years. This relatively new term is describing algorithms written that can adapt when encountering new data. Machine Learning has tremendous potential to transform companies, whether it's for creating self-driving cars, suggesting products for consumers to purchase from an online store, or redacting sensitive information from public documents preventing identity theft. Machine Learning is the key to why Extract Systems can automate document handling, data indexing, and redacting of unstructured documents

Making predictions based on Machine Learning

Making predictions is the most common application of all Machine Learning tools. Here are some examples of prediction problems that businesses might face:

  • Making personalized recommendations for customers
  • Forecasting long-term customer loyalty
  • Anticipating the future performance of employees
  • Rating the credit risk of loan applicants
  • Suggesting sensitive data to redact on public documents

These settings share some common features. For one, they are all complex environments, where the right decision might depend on a lot of variables (which means they require “wide” data). They also have some outcome to validate the results of a prediction – like whether someone clicks on a recommended item, whether a customer makes another purchase, or whether the suggested data to redact is validated as being accurate.

Machine Learning and Extract Systems

By using Machine Learning, Extract can accurately automate workflows, and make predictions based on the software’s adaptive learning from previous experiences. Extract’s software learns from previous documents that have been accurately redacted, or indexed and can suggest additional data within documents that should be given a closer look. Human verifiers will then be notified with color coded signifiers, indicating how accurate the software is that the data should be indexed or redacted.

Let’s have a deeper discussion about how Extract’s intelligent software can help your organization achieve faster results.  


About the Author: Tera Madigan

As a designer of experiences, Tera strives to bridge the gap between the user's needs, the physical world, and innovative technology by creating intuitive and engaging cross-channel experiences. By gaining valuable insight from people's motivations, behaviors, and attitudes, she enjoys analyzing findings to quickly iterate on ideas to help develop amazing products and experiences that meet real human needs--needs that they may never have thought about. She is an experienced Creative Leader who has a strong record of building and implementing successful branding, digital marketing, and eCommerce strategic plans. Tera has first-hand experience with starting and leading programs that have substantially increased awareness and driven large growth in eCommerce, UX, Web/App Design, and Digital Marketing for leading brands. She has a strong talent for understanding "the big picture" and leading large diversified and cross-functional teams in a large corporate setting as well as a fast paced, nimble and agile startup environment.