Icon

Similarity in trend

Similarity in trend
Generate trend columns for the stores in the data Compare the trend columns of the control stores to the test store in terms of R-squared. Generate a new feature Counter that indicates the similarity ofthe trend column of a control store to the test store. Add static features before similarity search This workflow finds the most similar control series to a test series in terms ofthe trend and a few static features Test: 10018Control: 10068, 10118, 10168, 10218 Read weekly_transactions.xlsOnly trendadd store ID to columnsRegExBy row ID test storein all iterationsR2Check visuallystable features sq_feet, avg_month_salesBased on R2,sq_feet,avg_month_sales0 for the closesttop: greatest R20 for the teststore Excel Reader Group Loop Start AggregationGranularity Timestamp Alignment Decompose Signal RowID Column Filter Column Rename(Regex) String Manipulation(Variable) Loop End (ColumnAppend) Column ListLoop Start Numeric Scorer Row Filter Normalizer Line Plot (Plotly) Loop End (ColumnAppend) Transpose GroupBy Joiner (Labs) Normalizer Similarity Search Counter Generation Sorter Normalizer Counter Generation The same time rangefor all stores Only 5 storesfor the demo Generate trend columns for the stores in the data Compare the trend columns of the control stores to the test store in terms of R-squared. Generate a new feature Counter that indicates the similarity ofthe trend column of a control store to the test store. Add static features before similarity search This workflow finds the most similar control series to a test series in terms ofthe trend and a few static features Test: 10018Control: 10068, 10118, 10168, 10218 Read weekly_transactions.xlsOnly trendadd store ID to columnsRegExBy row ID test storein all iterationsR2Check visuallystable features sq_feet, avg_month_salesBased on R2,sq_feet,avg_month_sales0 for the closesttop: greatest R20 for the teststoreExcel Reader Group Loop Start AggregationGranularity Timestamp Alignment Decompose Signal RowID Column Filter Column Rename(Regex) String Manipulation(Variable) Loop End (ColumnAppend) Column ListLoop Start Numeric Scorer Row Filter Normalizer Line Plot (Plotly) Loop End (ColumnAppend) Transpose GroupBy Joiner (Labs) Normalizer Similarity Search Counter Generation Sorter Normalizer Counter Generation The same time rangefor all stores Only 5 storesfor the demo

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
Created by: Maarit Laukkanen
Created at: 2020-12-08
On NodePit since: 2021-09-17
Last update: 2024-04-27
Created with KNIME version: v4.3.0
Tags: trendseasonalitymatch pairs

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!