Hub
NodeNode / Learner

Potential Density Initializer

KNIME Labs Active Learning Score Density
Drag & drop
Like
Copy short link

Creates a potential based density model of a n-dimensional vector space.
In order to calculate the density at a specific row in the table, its immediate neighborhood in the vector space is considered. The closer and more numerous the neighbors the higher is the density. The size of the neighborhood is controlled by the Radius Alpha parameter and a larger value will result in a larger neighborhood. For more details see the PBAC paper by Cebron et. al.
If the node fails due to memory problems, this is usually due to a too large radius Alpha which results in many rows with a lot of neighbors.

Node details

Input ports
  1. Type: Table
    Data
    Table to build a density model for.
Output ports
  1. Type: Density Scorer Model
    Density Scorer Model
    Density scorer model.

Extension

The Potential Density Initializer node is part of this extension:

Related workflows & nodes

  1. Active Learning with Basic SVM Model
    Active learning Potential density Uncertainty
    +12
    This workflow uses a simple example to demonstrate one possible structure for an active l…
    knime > Examples > 04_Analytics > 12_Active_Learning > 04_Active_Learning_with_basic_SVM
    knime
  2. Active Learning with Basic SVM Model
    Active learning Potential density Uncertainty
    +12
    This workflow uses a simple example to demonstrate one possible structure for an active l…
    ngohienduong > Public > 04_Active_Learning_with_basic_SVM
    ngohienduong
  3. Active Learning with Body Mass Index Heuristic
    Active learning Potential density Uncertainty
    +14
    This workflow uses a simple example to demonstrate one possible structure for an active l…
    paolotamag > Public > 04_Active_Learning_with_Body_Mass_Index
    paolotamag
  4. Active Learning - Exploration / Exploitation Score
    PBAC Active learning Human-in-the-loop
    +13
    This workflows shows examples of Active Learning with the Exploration vs Exploitation str…
    knime > Examples > 04_Analytics > 12_Active_Learning > 01_Active_Learning_PBCA_default
    gab1one
  5. Active Learning - Exploration / Exploitation Score
    PBAC Active learning Human-in-the-loop
    +13
    This workflows shows examples of Active Learning with the Exploration vs Exploitation str…
    noyankara > Public > 01_Active_Learning_PBCA_default
    noyankara