<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.2d1 20170631//EN" "JATS-journalpublishing1.dtd">
<ArticleSet>
  <Article>
    <Journal>
      <PublisherName>ijesm</PublisherName>
      <JournalTitle>International Journal of Engineering, Science and</JournalTitle>
      <PISSN>I</PISSN>
      <EISSN>S</EISSN>
      <Volume-Issue>Volume 6, Issue 4</Volume-Issue>
      <PartNumber/>
      <IssueTopic>Multidisciplinary</IssueTopic>
      <IssueLanguage>English</IssueLanguage>
      <Season>August 2017</Season>
      <SpecialIssue>N</SpecialIssue>
      <SupplementaryIssue>N</SupplementaryIssue>
      <IssueOA>Y</IssueOA>
      <PubDate>
        <Year>-0001</Year>
        <Month>11</Month>
        <Day>30</Day>
      </PubDate>
      <ArticleType>Engineering, Science and Mathematics</ArticleType>
      <ArticleTitle>NOVEL FRAMEWORK FOR DATA STREAMS CLASSIFICATION APPROACH BY DETECTING RECURRING FEATURE CHANGE IN FEATURE EVOLUTION AND FEATURE`S CONTRIBUTION IN CONCEPT DRIFT</ArticleTitle>
      <SubTitle/>
      <ArticleLanguage>English</ArticleLanguage>
      <ArticleOA>Y</ArticleOA>
      <FirstPage>99</FirstPage>
      <LastPage>105</LastPage>
      <AuthorList>
        <Author>
          <FirstName>Ms. Ritu                                                         Dr. Bhawna Suri                             Dr. P. S.</FirstName>
          <LastName>Kulkarni</LastName>
          <AuthorLanguage>English</AuthorLanguage>
          <Affiliation/>
          <CorrespondingAuthor>N</CorrespondingAuthor>
          <ORCID/>
        </Author>
      </AuthorList>
      <DOI/>
      <Abstract>Data stream classification poses many challenges, most of which are not addressed by the state –of-the-art. These are infinite length, concept drift, concept evolution, feature evolution. Data streams are assumed to be infinite in length, which necessitates single pass incremental learning techniques. Concept drift occurs in a data stream when the underlying concept changes over time. Most existing data stream classification techniques address only the infinite length and concept drift problems. However, to the best of our knowledge, no drift detection method provides insights into which features are involved in the concept drift, which is potentially valuable information. For example, if a feature is contributing to a concept drift it can be assumed that the feature may have become either more or less relevant for the concept encoded in the stream after the drift. This knowledge about a feature’s contribution to concept drift could be used to develop an efficient real-time feature selection method that does not require examining the entire feature space for online feature selection. Given a data stream, we want to improve data stream classification accuracy in dynamic feature space by using optimal and dynamic feature selection technique and also address the problem of detecting recurring feature change and feature’s contribution in concept drift. This paper proposes a framework for the development of real time feature selection in which we will improve upon the classification accuracy to select the features.</Abstract>
      <AbstractLanguage>English</AbstractLanguage>
      <Keywords>real time feature selection in which we will improve  upon the classification accuracy to select the features.</Keywords>
      <URLs>
        <Abstract>https://ijesm.co.in/ubijournal-v1copy/journals/abstract.php?article_id=3293&amp;title=NOVEL FRAMEWORK FOR DATA STREAMS CLASSIFICATION APPROACH BY DETECTING RECURRING FEATURE CHANGE IN FEATURE EVOLUTION AND FEATURE`S CONTRIBUTION IN CONCEPT DRIFT</Abstract>
      </URLs>
      <References>
        <ReferencesarticleTitle>References</ReferencesarticleTitle>
        <ReferencesfirstPage>16</ReferencesfirstPage>
        <ReferenceslastPage>19</ReferenceslastPage>
        <References/>
      </References>
    </Journal>
  </Article>
</ArticleSet>