k-means clustering and hierarchical clustering (Plantae)

Knowledge graph centered on k-means clustering and hierarchical clustering (Plantae) with 26 nodes and 77 connections. Top connected: Aedes communis, vascular plants, mycorrhizal fungi, Vaccinium, Agoseris glauca.

Method synopsis

Alternative species groupings using k-means and hierarchical agglomerative clustering based on multivariate trait data. Compared explanatory power and species composition to traditional functional groups.

Synthesized from method descriptions across 1 paper using this protocol.

Procedure from a recent peer-reviewed implementation

Steps below were extracted from the most recent peer-reviewed implementation of this protocol in the corpus — Traditional plant functional groups explain variation in economic but not size related traits across the tundra biome (2018), Global ecology and biogeography. Implementations in other papers (listed below) may differ.

1. Create post hoc trait-based classifications
   Grouped species using k-means clustering (k-means) and hierarchical agglomerative clustering (HCA). K-means clustering employs a top-down approach, assigning species to groups based on multivariate distance from group means. HCA clustering employs a bottom-up approach, iteratively combining groups with similar traits. Used Euclidean distance and Ward's criterion to measure linkage for HCA.
   Quantities: Selected 4-cluster solution for both methods to correspond with the number of traditional functional groupsDuration: Not specifiedConditions: Used Euclidean distance for both clustering methods
   Equipment: R package 'vegan'
2. Assess species composition similarity between grouping methods
   Calculated the maximum possible number of consistently categorized species amongst grouping methods and estimated the relative abundance of consistently grouped species within the ITEX database using the most recent year for all plots and aggregating at the site level.
   Quantities: Calculated similarity percentages between traditional functional groups and post hoc classificationsDuration: Not specifiedConditions: Used ITEX database with most recent year for each plot
   Equipment: ITEX database (Polar Data Catalogue: CCIN 10786)

Typical Equipment

• R package 'vegan'

Output Measurements

• species cluster assignments
• variance explained

Papers Using This Protocol (1)

Species Studied With This Protocol

Places Where Used