are defined by the Machine Learning algorithm we apply in the analytics project.

are the three different types of measures we need to compute an analytics problem.

can be used to understand the scope of the problem, or how complex it is.

describe a problem or problem space we want to address through analytics.

the result of the analytics project must be measurable in Return-on-Investment.

we expect analytics to give us insight into the problem that enables us to do something to help solve the problem.

the insight we get should also be something that we have the capacity to use or accomplish in some way.

the pattern should give us relevant information about the problem that isn't obvious.

the pattern should predict or foresee exact measures we need to implement in order to solve the considered problem.

managing data as part of a firm's IT Management function.

capturing, cleaning, and transforming of unstructured social media and web data.

extracting patterns from data to improve decision-making.

the use of big data technologies to store and process big, unstructured data sets.

tackling questions related to data ethics and regulation.

anomaly or outlier detection, e.g., to extract patterns that identify strange or abnormal events, such as fraudulent insurance claims.

association-rule mining, e.g., to identify products that are frequently bought together.

interrelation rules, e.g., to understand the linkage between customer and production orders.

classification rules or prediction, e.g., in order to identify spam e-mails.

clustering, e.g., to understand customer segmentation.

correlation laws, e.g., to predict how well customers will appreciate the new design of a product.