Learning with Drift Detection View Full Text


Ontology type: schema:Chapter      Open Access: True


Chapter Info

DATE

2004

AUTHORS

João Gama , Pedro Medas , Gladys Castillo , Pedro Rodrigues

ABSTRACT

Most of the work in machine learning assume that examples are generated at random according to some stationary probability distribution. In this work we study the problem of learning when the distribution that generate the examples changes over time. We present a method for detection of changes in the probability distribution of examples. The idea behind the drift detection method is to control the online error-rate of the algorithm. The training examples are presented in sequence. When a new training example is available, it is classified using the actual model. Statistical theory guarantees that while the distribution is stationary, the error will decrease. When the distribution changes, the error will increase. The method controls the trace of the online error of the algorithm. For the actual context we define a warning level, and a drift level. A new context is declared, if in a sequence of examples, the error increases reaching the warning level at example kw, and the drift level at example kd. This is an indication of a change in the distribution of the examples. The algorithm learns a new model using only the examples since kw. The method was tested with a set of eight artificial datasets and a real world dataset. We used three learning algorithms: a perceptron, a neural network and a decision tree. The experimental results show a good performance detecting drift and with learning the new concept. We also observe that the method is independent of the learning algorithm. More... »

PAGES

286-295

References to SciGraph publications

Book

TITLE

Advances in Artificial Intelligence – SBIA 2004

ISBN

978-3-540-23237-7
978-3-540-28645-5

Identifiers

URI

http://scigraph.springernature.com/pub.10.1007/978-3-540-28645-5_29

DOI

http://dx.doi.org/10.1007/978-3-540-28645-5_29

DIMENSIONS

https://app.dimensions.ai/details/publication/pub.1051660541


Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
Incoming Citations Browse incoming citations for this publication using opencitations.net

JSON-LD is the canonical representation for SciGraph data.

TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

[
  {
    "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
    "about": [
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0801", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Artificial Intelligence and Image Processing", 
        "type": "DefinedTerm"
      }, 
      {
        "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
        "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
        "name": "Information and Computing Sciences", 
        "type": "DefinedTerm"
      }
    ], 
    "author": [
      {
        "affiliation": {
          "alternateName": "University of Porto", 
          "id": "https://www.grid.ac/institutes/grid.5808.5", 
          "name": [
            "LIACC, University of Porto, Rua Campo Alegre 823, 4150, Porto, Portugal", 
            "Fac. Economics, University of Porto"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Gama", 
        "givenName": "Jo\u00e3o", 
        "id": "sg:person.011001067722.17", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011001067722.17"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Porto", 
          "id": "https://www.grid.ac/institutes/grid.5808.5", 
          "name": [
            "LIACC, University of Porto, Rua Campo Alegre 823, 4150, Porto, Portugal"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Medas", 
        "givenName": "Pedro", 
        "id": "sg:person.015627462141.57", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015627462141.57"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Aveiro", 
          "id": "https://www.grid.ac/institutes/grid.7311.4", 
          "name": [
            "LIACC, University of Porto, Rua Campo Alegre 823, 4150, Porto, Portugal", 
            "University of Aveiro"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Castillo", 
        "givenName": "Gladys", 
        "id": "sg:person.015237077151.36", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015237077151.36"
        ], 
        "type": "Person"
      }, 
      {
        "affiliation": {
          "alternateName": "University of Porto", 
          "id": "https://www.grid.ac/institutes/grid.5808.5", 
          "name": [
            "LIACC, University of Porto, Rua Campo Alegre 823, 4150, Porto, Portugal"
          ], 
          "type": "Organization"
        }, 
        "familyName": "Rodrigues", 
        "givenName": "Pedro", 
        "id": "sg:person.011726504540.41", 
        "sameAs": [
          "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011726504540.41"
        ], 
        "type": "Person"
      }
    ], 
    "citation": [
      {
        "id": "sg:pub.10.1007/3-540-59286-5_74", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1009791448", 
          "https://doi.org/10.1007/3-540-59286-5_74"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1023/a:1007661119649", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1018650476", 
          "https://doi.org/10.1023/a:1007661119649"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/bf00116900", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1025237168", 
          "https://doi.org/10.1007/bf00116900"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "sg:pub.10.1007/bf00116900", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1025237168", 
          "https://doi.org/10.1007/bf00116900"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.2307/1390807", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1069468261"
        ], 
        "type": "CreativeWork"
      }, 
      {
        "id": "https://doi.org/10.3233/ida-2004-8305", 
        "sameAs": [
          "https://app.dimensions.ai/details/publication/pub.1107700015"
        ], 
        "type": "CreativeWork"
      }
    ], 
    "datePublished": "2004", 
    "datePublishedReg": "2004-01-01", 
    "description": "Most of the work in machine learning assume that examples are generated at random according to some stationary probability distribution. In this work we study the problem of learning when the distribution that generate the examples changes over time. We present a method for detection of changes in the probability distribution of examples. The idea behind the drift detection method is to control the online error-rate of the algorithm. The training examples are presented in sequence. When a new training example is available, it is classified using the actual model. Statistical theory guarantees that while the distribution is stationary, the error will decrease. When the distribution changes, the error will increase. The method controls the trace of the online error of the algorithm. For the actual context we define a warning level, and a drift level. A new context is declared, if in a sequence of examples, the error increases reaching the warning level at example kw, and the drift level at example kd. This is an indication of a change in the distribution of the examples. The algorithm learns a new model using only the examples since kw. The method was tested with a set of eight artificial datasets and a real world dataset. We used three learning algorithms: a perceptron, a neural network and a decision tree. The experimental results show a good performance detecting drift and with learning the new concept. We also observe that the method is independent of the learning algorithm.", 
    "editor": [
      {
        "familyName": "Bazzan", 
        "givenName": "Ana L. C.", 
        "type": "Person"
      }, 
      {
        "familyName": "Labidi", 
        "givenName": "Sofiane", 
        "type": "Person"
      }
    ], 
    "genre": "chapter", 
    "id": "sg:pub.10.1007/978-3-540-28645-5_29", 
    "inLanguage": [
      "en"
    ], 
    "isAccessibleForFree": true, 
    "isPartOf": {
      "isbn": [
        "978-3-540-23237-7", 
        "978-3-540-28645-5"
      ], 
      "name": "Advances in Artificial Intelligence \u2013 SBIA 2004", 
      "type": "Book"
    }, 
    "name": "Learning with Drift Detection", 
    "pagination": "286-295", 
    "productId": [
      {
        "name": "dimensions_id", 
        "type": "PropertyValue", 
        "value": [
          "pub.1051660541"
        ]
      }, 
      {
        "name": "doi", 
        "type": "PropertyValue", 
        "value": [
          "10.1007/978-3-540-28645-5_29"
        ]
      }, 
      {
        "name": "readcube_id", 
        "type": "PropertyValue", 
        "value": [
          "f615d80f00c226b529c41e73c8add21d30df65264c76c2270aa1220024511c2c"
        ]
      }
    ], 
    "publisher": {
      "location": "Berlin, Heidelberg", 
      "name": "Springer Berlin Heidelberg", 
      "type": "Organisation"
    }, 
    "sameAs": [
      "https://doi.org/10.1007/978-3-540-28645-5_29", 
      "https://app.dimensions.ai/details/publication/pub.1051660541"
    ], 
    "sdDataset": "chapters", 
    "sdDatePublished": "2019-04-16T08:28", 
    "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
    "sdPublisher": {
      "name": "Springer Nature - SN SciGraph project", 
      "type": "Organization"
    }, 
    "sdSource": "s3://com-uberresearch-data-dimensions-target-20181106-alternative/cleanup/v134/2549eaecd7973599484d7c17b260dba0a4ecb94b/merge/v9/a6c9fde33151104705d4d7ff012ea9563521a3ce/jats-lookup/v90/0000000363_0000000363/records_70066_00000002.jsonl", 
    "type": "Chapter", 
    "url": "https://link.springer.com/10.1007%2F978-3-540-28645-5_29"
  }
]
 

Download the RDF metadata as:  json-ld nt turtle xml License info

HOW TO GET THIS DATA PROGRAMMATICALLY:

JSON-LD is a popular format for linked data which is fully compatible with JSON.

curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-28645-5_29'

N-Triples is a line-based linked data format ideal for batch operations.

curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-28645-5_29'

Turtle is a human-readable linked data format.

curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-28645-5_29'

RDF/XML is a standard XML format for linked data.

curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/978-3-540-28645-5_29'


 

This table displays all metadata directly associated to this object as RDF triples.

114 TRIPLES      23 PREDICATES      32 URIs      20 LITERALS      8 BLANK NODES

Subject Predicate Object
1 sg:pub.10.1007/978-3-540-28645-5_29 schema:about anzsrc-for:08
2 anzsrc-for:0801
3 schema:author Na7d599d814ed4a73aa40e4e95a723e6b
4 schema:citation sg:pub.10.1007/3-540-59286-5_74
5 sg:pub.10.1007/bf00116900
6 sg:pub.10.1023/a:1007661119649
7 https://doi.org/10.2307/1390807
8 https://doi.org/10.3233/ida-2004-8305
9 schema:datePublished 2004
10 schema:datePublishedReg 2004-01-01
11 schema:description Most of the work in machine learning assume that examples are generated at random according to some stationary probability distribution. In this work we study the problem of learning when the distribution that generate the examples changes over time. We present a method for detection of changes in the probability distribution of examples. The idea behind the drift detection method is to control the online error-rate of the algorithm. The training examples are presented in sequence. When a new training example is available, it is classified using the actual model. Statistical theory guarantees that while the distribution is stationary, the error will decrease. When the distribution changes, the error will increase. The method controls the trace of the online error of the algorithm. For the actual context we define a warning level, and a drift level. A new context is declared, if in a sequence of examples, the error increases reaching the warning level at example kw, and the drift level at example kd. This is an indication of a change in the distribution of the examples. The algorithm learns a new model using only the examples since kw. The method was tested with a set of eight artificial datasets and a real world dataset. We used three learning algorithms: a perceptron, a neural network and a decision tree. The experimental results show a good performance detecting drift and with learning the new concept. We also observe that the method is independent of the learning algorithm.
12 schema:editor N53cfd6b5f1ec4f1a801a7a928b3c6bfb
13 schema:genre chapter
14 schema:inLanguage en
15 schema:isAccessibleForFree true
16 schema:isPartOf N52abfd3b1009409e93f25e766d555f6d
17 schema:name Learning with Drift Detection
18 schema:pagination 286-295
19 schema:productId N3869ef6dff35425683480b1c814cfa0d
20 N400454b985a743af930b71c59fb32582
21 N8bab0b8b92d14fcc994cf261df35f664
22 schema:publisher N3044d1c1a9a940daad36f5c94739e0d2
23 schema:sameAs https://app.dimensions.ai/details/publication/pub.1051660541
24 https://doi.org/10.1007/978-3-540-28645-5_29
25 schema:sdDatePublished 2019-04-16T08:28
26 schema:sdLicense https://scigraph.springernature.com/explorer/license/
27 schema:sdPublisher N3eb8426db3514889b2724d04098cf7ce
28 schema:url https://link.springer.com/10.1007%2F978-3-540-28645-5_29
29 sgo:license sg:explorer/license/
30 sgo:sdDataset chapters
31 rdf:type schema:Chapter
32 N19207c826d494995b6c104ceb97fd440 rdf:first sg:person.011726504540.41
33 rdf:rest rdf:nil
34 N3044d1c1a9a940daad36f5c94739e0d2 schema:location Berlin, Heidelberg
35 schema:name Springer Berlin Heidelberg
36 rdf:type schema:Organisation
37 N3869ef6dff35425683480b1c814cfa0d schema:name doi
38 schema:value 10.1007/978-3-540-28645-5_29
39 rdf:type schema:PropertyValue
40 N3d27ad04f64049aa889f632e5a5398ee schema:familyName Labidi
41 schema:givenName Sofiane
42 rdf:type schema:Person
43 N3eb8426db3514889b2724d04098cf7ce schema:name Springer Nature - SN SciGraph project
44 rdf:type schema:Organization
45 N400454b985a743af930b71c59fb32582 schema:name dimensions_id
46 schema:value pub.1051660541
47 rdf:type schema:PropertyValue
48 N47bc59fd4ff34298b6b2084f79d142ff schema:familyName Bazzan
49 schema:givenName Ana L. C.
50 rdf:type schema:Person
51 N52abfd3b1009409e93f25e766d555f6d schema:isbn 978-3-540-23237-7
52 978-3-540-28645-5
53 schema:name Advances in Artificial Intelligence – SBIA 2004
54 rdf:type schema:Book
55 N53cfd6b5f1ec4f1a801a7a928b3c6bfb rdf:first N47bc59fd4ff34298b6b2084f79d142ff
56 rdf:rest Nebf26156f37d459a8a396a8408cf4226
57 N8bab0b8b92d14fcc994cf261df35f664 schema:name readcube_id
58 schema:value f615d80f00c226b529c41e73c8add21d30df65264c76c2270aa1220024511c2c
59 rdf:type schema:PropertyValue
60 N92d0fa9b561341ac9dcd49d170b0ab92 rdf:first sg:person.015627462141.57
61 rdf:rest N9b0ebee6a0d248f580eba7b240a4913e
62 N9b0ebee6a0d248f580eba7b240a4913e rdf:first sg:person.015237077151.36
63 rdf:rest N19207c826d494995b6c104ceb97fd440
64 Na7d599d814ed4a73aa40e4e95a723e6b rdf:first sg:person.011001067722.17
65 rdf:rest N92d0fa9b561341ac9dcd49d170b0ab92
66 Nebf26156f37d459a8a396a8408cf4226 rdf:first N3d27ad04f64049aa889f632e5a5398ee
67 rdf:rest rdf:nil
68 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
69 schema:name Information and Computing Sciences
70 rdf:type schema:DefinedTerm
71 anzsrc-for:0801 schema:inDefinedTermSet anzsrc-for:
72 schema:name Artificial Intelligence and Image Processing
73 rdf:type schema:DefinedTerm
74 sg:person.011001067722.17 schema:affiliation https://www.grid.ac/institutes/grid.5808.5
75 schema:familyName Gama
76 schema:givenName João
77 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011001067722.17
78 rdf:type schema:Person
79 sg:person.011726504540.41 schema:affiliation https://www.grid.ac/institutes/grid.5808.5
80 schema:familyName Rodrigues
81 schema:givenName Pedro
82 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011726504540.41
83 rdf:type schema:Person
84 sg:person.015237077151.36 schema:affiliation https://www.grid.ac/institutes/grid.7311.4
85 schema:familyName Castillo
86 schema:givenName Gladys
87 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015237077151.36
88 rdf:type schema:Person
89 sg:person.015627462141.57 schema:affiliation https://www.grid.ac/institutes/grid.5808.5
90 schema:familyName Medas
91 schema:givenName Pedro
92 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.015627462141.57
93 rdf:type schema:Person
94 sg:pub.10.1007/3-540-59286-5_74 schema:sameAs https://app.dimensions.ai/details/publication/pub.1009791448
95 https://doi.org/10.1007/3-540-59286-5_74
96 rdf:type schema:CreativeWork
97 sg:pub.10.1007/bf00116900 schema:sameAs https://app.dimensions.ai/details/publication/pub.1025237168
98 https://doi.org/10.1007/bf00116900
99 rdf:type schema:CreativeWork
100 sg:pub.10.1023/a:1007661119649 schema:sameAs https://app.dimensions.ai/details/publication/pub.1018650476
101 https://doi.org/10.1023/a:1007661119649
102 rdf:type schema:CreativeWork
103 https://doi.org/10.2307/1390807 schema:sameAs https://app.dimensions.ai/details/publication/pub.1069468261
104 rdf:type schema:CreativeWork
105 https://doi.org/10.3233/ida-2004-8305 schema:sameAs https://app.dimensions.ai/details/publication/pub.1107700015
106 rdf:type schema:CreativeWork
107 https://www.grid.ac/institutes/grid.5808.5 schema:alternateName University of Porto
108 schema:name Fac. Economics, University of Porto
109 LIACC, University of Porto, Rua Campo Alegre 823, 4150, Porto, Portugal
110 rdf:type schema:Organization
111 https://www.grid.ac/institutes/grid.7311.4 schema:alternateName University of Aveiro
112 schema:name LIACC, University of Porto, Rua Campo Alegre 823, 4150, Porto, Portugal
113 University of Aveiro
114 rdf:type schema:Organization
 




Preview window. Press ESC to close (or click here)


...