More and more people are looking to new types of tools and solutions for deeper insights into data than traditional statistics. There is great need to provide better services based on diagnosis, recommendation, exploration, estimation, optimization, etc. It is imperative to introduce more “intelligence” into
the systems in order to create value and enhance competitiveness.
The new age of “wearables” (smart watches, “glasses”, bio-sensors and new sensors in mobiles) requires intelligent processing to become truly meaningful. The availability of large amounts of data makes it possible to quickly train the models for prediction, classification, recognizing patterns in texts, etc.
But are current models sufficient? Intelligent systems at the forefront are often combinations of different types of AI, extraction features that span over multiple data sources with highly variable quality. Another important aspect is the ability to utilize feedback, such as from example users, which allow solutions to be automatically optimized and adapted to changing conditions. There will be many exciting practical examples from AI projects
from major U.S. companies such as eBay, Vodafone (one of the largest telecom operators) and Safeway (U.S. equivalent to ICA).
UCTfO1Sn5MXXPasEl3ZS4GGA
source
