header
Image from OpenLibrary

Semi-supervised classification using natural-based computation / Shahira Shaaban Azab Ahmed ; Supervised Mohamed Farouk Abdelhady , Hesham Ahmed Hefny

By: Contributor(s): Material type: TextTextLanguage: English Publication details: Cairo : Shahira Shaaban Azab Ahmed , 2017Description: 146 Leaves : charts , facsimiles ; 30cmOther title:
  • التصنيف شبه الإشرافي باستخدام الحوسبة المستوحاة من الطبيعة [Added title page title]
Subject(s): Available additional physical forms:
  • Issued also as CD
Dissertation note: Thesis (Ph.D.) - Cairo University - Institute of Statistical Studies and Research - Department of Computer and Information Science Summary: This Thesis presents a cluster-and-label model using PSO to optimize the cluster centroid. In addition, labeled data are used to label cluster and guide clustering process. In some domains, the number of clusters in semi-supervised classification is unknown as in the Automatic Knowledgebase Construction. This thesis proposes an algorithm 2ESPSO3 to detect the number of clusters in the dataset by using PSO to optimize silhouette score. Then, the detected numbers of clusters are used in exploratory semi-supervised classification tasks with an unanticipated cluster
Tags from this library: No tags from this library for this title. Log in to add tags.
Star ratings
    Average rating: 0.0 (0 votes)
Holdings
Item type Current library Home library Call number Copy number Status Barcode
Thesis Thesis قاعة الرسائل الجامعية - الدور الاول المكتبة المركزبة الجديدة - جامعة القاهرة Cai01.18.02.Ph.D.2017.Sh.S (Browse shelf(Opens below)) Not for loan 01010110075111000
CD - Rom CD - Rom مخـــزن الرســائل الجـــامعية - البدروم المكتبة المركزبة الجديدة - جامعة القاهرة Cai01.18.02.Ph.D.2017.Sh.S (Browse shelf(Opens below)) 75111.CD Not for loan 01020110075111000

Thesis (Ph.D.) - Cairo University - Institute of Statistical Studies and Research - Department of Computer and Information Science

This Thesis presents a cluster-and-label model using PSO to optimize the cluster centroid. In addition, labeled data are used to label cluster and guide clustering process. In some domains, the number of clusters in semi-supervised classification is unknown as in the Automatic Knowledgebase Construction. This thesis proposes an algorithm 2ESPSO3 to detect the number of clusters in the dataset by using PSO to optimize silhouette score. Then, the detected numbers of clusters are used in exploratory semi-supervised classification tasks with an unanticipated cluster

Issued also as CD

There are no comments on this title.

to post a comment.