Unstructured Data Analytics

CHAPTER 12The Future of Analytics

Imagination is the beginning of creation. You imagine what you desire, you will what you imagine, and at last you create what you will.

—George Bernard Shaw

In the past years we all witnessed a lot of blandishments around artificial intelligence (AI). The phenomenon of advanced analytics and artificial intelligence is not new, but the pace of recent development is. The main factors driving this acceleration are:

• The progression of machine learning, deep learning, and cognitive analytics algorithms.
• The enhanced computing capacity, enabling fast training of large and complex models.
  • Graphics processing units (GPUs), originally designed for video games, have been repurposed for high-speed data and algorithms crunching at speeds many times faster than traditional processor chips.
  • Silicon-level advances beyond GPUs are emerging, such as tensor processing units (TPUs), which are now aggregated in scalable data centers and are accessible through the cloud. Leading tech giants like Google are rethinking their computational architecture to run on TPUs, which are 15 to 30 times faster and 30 to 80 percent more power efficient than CPUs and GPUs.
• The access to massive amounts of data generated in billions through daily creation of images, online click streams, voice and video, mobile locations, wearables, sensors, and smart machines, embedded in the Internet of Things and clouds.

The combination of these breakthroughs led to the creation of Google ...