Page 171 - Building Big Data Applications

170 Building Big Data Applications

prior processing stored in a knowledge repository (a NoSQL or DBMS-like data-
base) along with the algorithms for machine learning.
3. The data is then processed through the hypothesis workflows
4. The output from a hypothesis and predictive mining exercises are sent to the
knowledge repository as a collection with meta-tags for search criteria and associ-
ated user geographic and demographic information as much available.
5. Process the outputs of hypothesis to outputs for further analysis or presentation to
users.
Examples of real-life implementations of machine learning:

IBM Watson
Amazon recommendation engine
Yelp ratings
Analysis of astronomical data
Human speech recognition
Stream analytics
Credit card fraud
Electronic trading fraud
Google robotedriven vehicles
Predict stock rates
Genome classification
Using semantic libraries, metadata and master data along with the data collected
from each iterative processing, enriches the capabilities of the algorithms to detect better
patterns and predict better outcomes.

FIGURE 9.7.1 User searches for movie.