Category: Premium Nodes

The Ungroup Words node is designed to take a user selected column and Ungroup the Words found in each input String into separate rows. The results can be used to identify a Product Name, SKU Number, or Brand from a general Description of the Product. Both Chinese and English is currently supported.

The String Repair node is typically designed to look through an Input Product Array for the String columns selected by the user for clean up and repair. Dirty string problems can arise when crawling data from competitor websites.

The Brand Discover node is designed to take a long list of Brand names and intelligently group them into a Brand Dictionary. The Brand Dictionary can then be used by a downstream Brand Repair node to clean up and repair the Brand names found in an Input Product Array. Both Chinese and English is currently supported.

The Brand Repair node is designed to look through an Input Product Array for raw Brand names and match them against a cleaned Brand Dictionary. If a match is found then the raw Brand name will be replaced by the clean Brand name found in the Brand Dictionary. Both Chinese and English is currently supported.

The Correlation Repair node replaces missing Product Attributes with the Attributes found in highly correlated Products. For example, if the user wishes to repair the ‘Brand’ column, then all Products with missing Brand values are compared against similar Products having a Brand value.

The Tag Importance node correlates the Features, Benefits, Attributes and Consumer Sentiment that describe a Product with the sales performance of that Product. Each Tag is correlated against: Product Purchased, Product In Consideration Set, Product Ranking, Customer Willingness To Pay, Product Price, etc.

The Keyword Resistance node is designed to identify all of the Paths from the Keyword Search Results to the Product Buy Pages. The node then aggregates the total Customer Resistance to moving along any of those Paths to find the Product.

The Product Ranking node is designed to take a set of Products in a Market and determine how each Virtual Customer ranks each of those Products. Rankings are then used to build Cumulative Rank Histograms and Venn Diagrams, and are used to calculate Expected Value.

The Competitive Radar node determines the degree of Competitive Rivalry between Products in a Market. The competitive landscape can then be plotted in a scatter plot or bubble chart with the Focus Product located at the (0, 0) origin and each of the Competitive Rival Products located within a concentric circle up to a distance of 1.0 from that origin.

The Similarity Family node takes a super-set of Products and allocates those Products into a smaller set of Product Families. The Products are allocated in accordance to their mutual correlation as well as whether the Products have the same Brand, Store, Location, Category, and Platform.

The Correlation Segmentation node allocates all Products into Product Families based upon their mutual Correlation. Very similar Products, as perceived by Customers, should end up in the same Product Family, whereas different Products should end up in different Product Families.

The Store Match node finds the best match across each Family of Products sold by the Brand’s Master Distributor. The Matching Algorithm relies heavily upon text-matching within the Description, but the Correlation can also be used if the Input Similarity Rankings are provided.