MINAS: multiclass learning algorithm for novelty detection in data streams

详细信息    查看全文

• 作者：Elaine Ribeiro de Faria…
• 关键词：Novelty detection ; Data streams ; Multiclass classification ; Concept evolution
• 刊名：Data Mining and Knowledge Discovery
• 出版年：2016
• 出版时间：May 2016
• 年：2016
• 卷：30
• 期：3
• 页码：640-680
• 全文大小：1,454 KB
• 参考文献：Aggarwal CC, Han J, Wang J, Yu PS (2003) A framework for clustering evolving data streams. In: Procedings of the 29th conference on very large data bases (VLDB’03), pp 81–92
  Al-Khateeb T, Masud MM, Khan L, Aggarwal C, Han J, Thuraisingham B (2012a) Stream classification with recurring and novel class detection using class-based ensemble. In: Proceedings of the IEEE 12th international conference on data mining (ICDM ’12), pp 31–40
  Al-Khateeb TM, Masud MM, Khan L, Thuraisingham B (2012b) Cloud guided stream classification using class-based ensemble. In: Proceedings of the 2012 IEEE 5th international conference on computing (CLOUD’12), pp 694–701
  Bifet A, Holmes G, Pfahringer B, Kranen P, Kremer H, Jansen T, Seidl T (2010) MOA: massive online analysis, a framework for stream classification and clustering. J Mach Learn Res 11:44–50
  Farid DM, Rahman CM (2012) Novel class detection in concept-drifting data stream mining employing decision tree. In: 7th international conference on electrical computer engineering (ICECE’ 2012), pp 630–633
  Faria ER, Gama J, Carvalho ACPLF (2013) Novelty detection algorithm for data streams multi-class problems. In: Proceedings of the 28th symposium on applied computing (SAC’13), pp 795–800
  Faria ER, Goncalves IJCR, Gama J, Carvalho ACPLF (2013) Evaluation methodology for multiclass novelty detection algorithms. In: 2nd Brazilian conference on intelligent systems (BRACIS’13), pp 19–25
  Farid DM, Zhang L, Hossain A, Rahman CM, Strachan R, Sexton G, Dahal K (2013) An adaptive ensemble classifier for mining concept drifting data streams. Exp Syst Appl 40(15):5895–5906CrossRef
  Frank A, Asuncion A (2010) UCI machine learning repository. http://​archive.​ics.​uci.​edu/​ml . Accessed 20 Aug 2015
  Gama J (2010) Knowledge discovery from data streams, vol 1, 1st edn. CRC press chapman hall, AtlantaCrossRef MATH
  Hayat MZ, Hashemi MR (2010) A DCT based approach for detecting novelty and concept drift in data streams. In: Proceedings of the international conference on soft computing and pattern recognition (SoCPaR), pp 373–378
  Krawczyk B, Woźniak M (2013) Incremental learning and forgetting in one-class classifiers for data streams. In: Proceedings of the 8th international conference on computer recognition systems (CORES’ 13), advances in intelligent systems and computing vol 226, pp 319–328
  Liu J, Xu G, Xiao D, Gu L, Niu X (2013) A semi-supervised ensemble approach for mining data streams. J Comput 8(11):2873–2879
  Lloyd SP (1982) Least squares quantization in PCM. IEEE Trans Inf Theory 28(2):129–137MathSciNet CrossRef MATH
  MacQueen JB (1967) Some methods of classification and analysis of multivariate observations. In: Proceedings of the 5th Berkeley symposium on mathematical statistics and probability, pp 281–297
  Masud M, Gao J, Khan L, Han J, Thuraisingham BM (2011) Classification and novel class detection in concept-drifting data streams under time constraints. IEEE Trans Knowl Data Eng 23(6):859–874CrossRef
  Masud MM, Chen Q, Khan L, Aggarwal CC, Gao J, Han J, Thuraisingham BM (2010) Addressing concept-evolution in concept-drifting data streams. In: Proceedings of the 10th IEEE international conference on data mining (ICDM’10), pp 929–934
  Naldi M, Campello R, Hruschka E, Carvalho A (2011) Efficiency issues of evolutionary k-means. Appl Soft Comput 11:1938–1952CrossRef
  Perner P (2008) Concepts for novelty detection and handling based on a case-based reasoning process scheme. Eng Appl Artif Intell 22:86–91CrossRef
  Rusiecki A (2012) Robust neural network for novelty detection on data streams. In: Proceedings of the 11th international conference on artificial intelligence and soft computing—volume part I (ICAISC’12), pp 178–186
  Spinosa EJ, Carvalho ACPLF, Gama J (2009) Novelty detection with application to data streams. Intell Data Anal 13(3):405–422
  Vendramin L, Campello R, Hruschka E (2010) Relative clustering validity criteria: a comparative overview. Stat Anal Data Min 3:209–235MathSciNet
  Zhang T, Ramakrishnan R, Livny M (1996) BIRCH: an efficient data clustering method for very large databases. In: Proceedings of the ACM SIGMOD international conference on management of data, pp 103–114
  
• 作者单位：Elaine Ribeiro de Faria (1)
  André Carlos Ponce de Leon Ferreira Carvalho (2)
  João Gama (3)
  
  1. Faculty of Computer Science, Federal University of Uberlândia, Uberlândia, Brazil
  2. Institute of Mathematics and Computer Science, University of São Paulo, São Carlos, Brazil
  3. Laboratory of Artificial Intelligence and Decision Support (LIAAD), University of Porto, Porto, Portugal
  
• 刊物类别：Computer Science
• 刊物主题：Data Mining and Knowledge Discovery
  Computing Methodologies
  Artificial Intelligence and Robotics
  Statistics
  Statistics for Engineering, Physics, Computer Science, Chemistry and Geosciences
  Information Storage and Retrieval
  
• 出版者：Springer Netherlands
• ISSN：1573-756X

文摘

Data stream mining is an emergent research area that aims at extracting knowledge from large amounts of continuously generated data. Novelty detection (ND) is a classification task that assesses if one or a set of examples differ significantly from the previously seen examples. This is an important task for data stream, as new concepts may appear, disappear or evolve over time. Most of the works found in the ND literature presents it as a binary classification task. In several data stream real life problems, ND must be treated as a multiclass task, in which, the known concept is composed by one or more classes and different new classes may appear. This work proposes MINAS, an algorithm for ND in data streams. MINAS deals with ND as a multiclass task. In the initial training phase, MINAS builds a decision model based on a labeled data set. In the online phase, new examples are classified using this model, or marked as unknown. Groups of unknown examples can be used later to create valid novelty patterns (NP), which are added to the current model. The decision model is updated as new data come over the stream in order to reflect changes in the known classes and allow the addition of NP. This work also presents a set of experiments carried out comparing MINAS and the main novelty detection algorithms found in the literature, using artificial and real data sets. The experimental results show the potential of the proposed algorithm.