Wiki Categories

Model Evaluation


The term association can create some misunderstandings since many people confuse it with if-then statements that help show the probability of relationships between data items and Entity- association. There are two possible definitions. In data science, the first definition is used more than the second one. Let's take a closer look.

1. Association Rule: This is a rule-based machine learning method for identifying relations between dataset variables. Based on the concept of strong rules, association methods can discover regularities between products in large-scale transaction data recorded by point-of-sale (POS) systems, like in supermarkets. For example, if you say “onions “ and “potatoes”, I can make an association and predict that the main subject that can be “burger” .

In real life, the sales data of a supermarket would indicate that if a customer buys onions and potatoes together, he is also likely buy hamburger meat products. This information can be used as a basis for decisions about marketing activities such as promotional pricing or product placements.

2. But what about Entity-Association (EA) in a relation model?

What does EA mean ?

The Entity-Association model (EA) is also frequently referred to as Entity-Relationship and sometimes Entity-Relationship-Attribute. The EA model proposes concepts (mainly entities, associations and attributes) to describe a set of data relating to a defined domain in order to integrate them into a Database (DB). The relational model is poor in semantic representation capacity.

However, the designer needs some semantic information that the EA model allows him to describe. It is difficult to model a domain in a form directly manipulated by a DBMS. One or more intermediate models can be extremely useful, with the EA model being the most common. To better understand the association rule let's define some important terms.

Entity: An entity is a concrete or abstract object that can be recognized distinctly. An entity type consists of a set of entities that have the same characteristics. Be warned that we often use the word entity instead of the word type-entity by misusing of the language.

Examples of type-entity: Person, Automobile, Region.

Examples of entities: My car, Las Vegas, Jerusalem.


Explorium delivers the end-game of every data science process - from raw, disconnected data to game-changing insights, features, and predictive models. Better than any human can.
Request a demo
Get started with Explorium External Data Cloud Start for free