Automatically assigned DDC number:

Manually assigned DDC number: 00631

Title: Parallel Classification by Feature Partitioning


Subject: Huseyin Simit¸ci H. Altay Guvenir Parallel Classification by Feature Partitioning

Description: This work presents a parallel method for learning from examples using parallel feature partitioning (PFP). Feature partitioning (FP) is an inductive, incremental and supervised learning method proposed by S¸irin and Guvenir [1]. PFP assigns feature dimensions to separate nodes. Learning in PFP is accomplished by storing the objects separately in each feature dimension as disjoint partitions of values. Every node expands a partition, which is initially a point in the feature dimension through generalization. The CFP algorithm specializes a partition by subdividing it into sub-partitions. PFP is implemented in the PCFP (Parallel Classification by Feature Partitioning) algorithm. PCFP is tested in six different domains, and results are compared with CFP of S¸irin and Guvenir [1]. feature 1 feature 2 feature k host feature node 2 partitions for feature 2 ...... .... Figure 1: Topology of the feature nodes and the host node 1 Introduction This work presents a parallel method for learni...

Contributor: The Pennsylvania State University CiteSeer Archives

Publisher: unknown

Date: 1994-02-01

Format: ps



Language: en

Rights: unrestricted


<?xml   version="1.0"   encoding="UTF-8"?>


      <rec   ID="SELF"   Type="SELF"   CiteSeer_Book="SELF"   CiteSeer_Volume="SELF"   Title="Parallel   Classification   by   Feature   Partitioning"   />