# Node Description

# Similarity Collapse Node

The Similarity Collapse node takes a super-set of Products and collapses them into a smaller set of aggregated Products. At the same time, the Similarity Collapse node creates a corresponding Output Correlation Matrix based upon the smaller set of aggregated Products. In this way, a large market can be simulated through a smaller set of representative Products.

Upstream of the Similarity Collapse node would typically be a Similarity Family node or a Correlation Segmentation node. Either upstream node can define how the super-set of Products should be allocated into a smaller set of Product Families. Products having the same Brand, Store, Location, Category, Platform can be grouped into Families. Furthermore the upstream Similarity Family node may require that Products also be sufficiently similar to each other in order to be grouped together.

Downstream of the Similarity Collapse node would typically be a Matrix Distributions node followed by a Tune Market node. The Matrix Distributions node would convert the Output Correlation Matrix from the Similarity Collapse node into a set of Unit Distributions. The Tune Market node would then create a Willingness To Pay (WTP) Matrix that define the preferences of all Virtual Customers in the Market.

*This Premium Node is not available as part of the free Community Edition. Premium Nodes help clean and connect real-world data to Market Simulations, and provide advanced Market Science analysis. Note that these descriptions are often deliberately vague.*

# Downloads

# Similarity Collapse

The Similarity Collapse node takes a super-set of Products and collapses them into a smaller set of aggregated Products. At the same time, the Similarity Collapse node creates a corresponding Output Correlation Matrix based upon the smaller set of aggregated Products. In this way, a large Market can be simulated through a smaller set of representative Products.

## Inputs

#### Input Product Array

The super-set of Products from which the smaller set of Product Families will be collapsed.

#### Product Similarity Rankings

The *optional* list of all Product-to-Product rankings. That is, from the Customers who looked at Product01 how did they rank Product02. Alternatively, from the Customers who looked at Product01 what is the probability that they also looked at Product02 (sorted to provide a rank order).

## Node

#### Configuration

The Similarity Collapse node can aggregate Products together by Brand, Store, Location, Family, Category or Platform. In addition, the ‘Market’ option causes the Similarity Collapse node to collapse all of the Products in the entire Market down into a single Product.

## Outputs

#### Output Product Array

The Output Product Array contains the subset of Products collapsed from the Input Product Array. Note how the number of Products in this example has dropped from 115 to the more manageable 57.

#### Product Correlation Matrix

A symmetrical matrix reflecting the correlation between each collapsed Product and each of the other collapsed Products in the Market.