Data Mining and Knowledge Discovery

, Volume 30, Issue 1, pp 1–46 | Cite as

Decomposition-by-normalization (DBN): leveraging approximate functional dependencies for efficient CP and tucker decompositions

  • Mijung KimEmail author
  • K. Selçuk Candan


For many multi-dimensional data applications, tensor operations as well as relational operations both need to be supported throughout the data lifecycle. Tensor based representations (including two widely used tensor decompositions, CP and Tucker decompositions) are proven to be effective in multi-aspect data analysis and tensor decomposition is an important tool for capturing high-order structures in multi-dimensional data. Although tensor decomposition is shown to be effective for multi-dimensional data analysis, the cost of tensor decomposition is often very high. Since the number of modes of the tensor data is one of the main factors contributing to the costs of the tensor operations, in this paper, we focus on reducing the modality of the input tensors to tackle the computational cost of the tensor decomposition process. We propose a novel decomposition-by-normalization scheme that first normalizes the given relation into smaller tensors based on the functional dependencies of the relation, decomposes these smaller tensors, and then recombines the sub-results to obtain the overall decomposition. The decomposition and recombination steps of the decomposition-by-normalization scheme fit naturally in settings with multiple cores. This leads to a highly efficient, effective, and parallelized decomposition-by-normalization algorithm for both dense and sparse tensors for CP and Tucker decompositions. Experimental results confirm the efficiency and effectiveness of the proposed decomposition-by-normalization scheme compared to the conventional nonnegative CP decomposition and Tucker decomposition approaches.


Tensor decomposition CP decomposition Tucker decomposition Parallel tensor decomposition Parallel CP decomposition Parallel Tucker decomposition 



This work is partially funded by NSF Grants #116394 “RanKloud: Data Partitioning and Resource Allocation Strategies for Scalable Multimedia and Social Media Analysis”.


  1. Allen GI (2012) Sparse higher-order principal components analysis. In: Proceedings of the 15th international conference on artificial intelligence and statistics (AISTATS)Google Scholar
  2. Andersson CA, Bro R (2000) The n-way toolbox for matlab. Chemom Intell Lab Syst 52(1):1–4.
  3. Antikainen J, Havel J, Josth JR, Herout A, Zemcik P, Hauta-Kasari M (2011) Nonnegative tensor factorization accelerated using gpgpu. IEEE Trans Parallel Distrib Syst 22(7):1135–1141CrossRefGoogle Scholar
  4. Bader BW, Kolda TG (2006) Efficient matlab computations with sparse and factored tensors. Technical Report SAND2006-7592, Sandia National LaboratoriesGoogle Scholar
  5. Bader BW, Kolda TG (2007) Matlab tensor toolbox version 2.2.
  6. Carroll J, Chang JJ (1970) Analysis of individual differences in multidimensional scaling via an n-way generalization of eckart-young decomposition. Psychometrika 35:283–319zbMATHCrossRefGoogle Scholar
  7. Chu W, Ghahramani Z (2009) Probabilistic models for incomplete multi-dimensional arrays. In: Proceedings of the 12th international conference on artificial intelligence and statisticsGoogle Scholar
  8. Elmasri R, Navathe SB (1994) Fundamentals of database systems, 2nd edn. Benjamin-Cummings, Redwood CityzbMATHGoogle Scholar
  9. Frank A, Asuncion A (2010) UCI machine learning repository.
  10. Garey MR, Johnson DS (1979) Computers and intractability: a guide to the theory of NP-completeness. W. H. Freeman, New YorkzbMATHGoogle Scholar
  11. Harshman RA (1970) Foundations of the PARAFAC procedure: models and conditions for an “explanatory” multi-modal factor analysis. UCLA Working Papers in Phonetics 16(1):84Google Scholar
  12. Hoff PD (2011) Hierarchical multilinear models for multiway data. Comput Stat Data Anal 55(1):530–543. doi: 10.1016/j.csda.2010.05.020 zbMATHMathSciNetCrossRefGoogle Scholar
  13. Huhtala Y, Kärkkäinen J, Porkka P, Toivonen H (1999) Tane: an efficient algorithm for discovering functional and approximate dependencies. Comput J 42(2):100–111zbMATHCrossRefGoogle Scholar
  14. Ilyas IF, Markl V, Haas PJ, Brown P, Aboulnaga (2004) A Cords: automatic discovery of correlations and soft functional dependencies. In: SIGMOD conference, pp. 647–658Google Scholar
  15. Karmarker N, Karp RM (1983) The differencing method of set partitioning. Technical report, BerkeleyGoogle Scholar
  16. Kim M, Candan KS (2011) Approximate tensor decomposition within a tensor-relational algebraic framework. In: Proceedings of the 20th ACM international conference on information and knowledge management, pp. 1737–1742 doi: 10.1145/2063576.2063827
  17. Kolda T, Sun J (2008) Scalable tensor decompositions for multi-aspect data mining. In: Proceedings of the 8th IEEE international conference on data mining, pp. 363–372. doi: 10.1109/ICDM.2008.89
  18. Kolda TG, Bader BW (2009) Tensor decompositions and applications. SIAM Rev 51(3):455–500. doi: 10.1137/07070111X zbMATHMathSciNetCrossRefGoogle Scholar
  19. Kolda TG, Bader BW, Kenny JP (2005) Higher-order web link analysis using multilinear algebra. In: Proceedings of the 5th IEEE international conference on data mining, pp. 242–249. doi: 10.1109/ICDM.2005.77
  20. Kruskal JB (1977) Three-way arrays: rank and uniqueness of trilinear decompositions, with application to arithmetic complexity and statistics. Linear Algebr Appl 18(2):95–138zbMATHMathSciNetCrossRefGoogle Scholar
  21. Lopes S, Petit JM, Lakhal L (2000) Efficient discovery of functional dependencies and armstrong relations. In: Proceedings of the 7th international conference on extending database technology: advances in database technology, EDBT ’00. Springer, London, pp. 350–364Google Scholar
  22. Mahoney MW, Maggioni M, Drineas P (2006) Tensor-cur decompositions for tensor-based data. In: Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining, pp. 327–336. doi: 10.1145/1150402.1150440
  23. Mangasarian OL, Wolberg WH (1990) Cancer diagnosis via linear programming. SIAM News 23(5):1–18Google Scholar
  24. Mannila H, Räihä KJ (1992) On the complexity of inferring functional dependencies. Discret Appl Math 40(2):237–243. doi: 10.1016/0166-218X(92)90031-5 zbMATHCrossRefGoogle Scholar
  25. Movielens dataset from grouplens research group (2013).
  26. Phan AH, Cichocki A (2011) Parafac algorithms for large-scale problems. Neurocomputing 74(11):1970–1984. doi: 10.1016/j.neucom.2010.06.030 CrossRefGoogle Scholar
  27. Rand WM (1971) Objective criteria for the evaluation of clustering methods. J Am Stat Assoc 66(336):846–850CrossRefGoogle Scholar
  28. Ruggles S, Sobek M (1997) Integrated public use microdata series: Version 2.0 minneapolis: historical census projects
  29. Sanchez E, Kowalski BR (1986) Generalized rank annihilation factor analysis. Anal Chem 58(2):496–499. doi: 10.1021/ac00293a054 CrossRefGoogle Scholar
  30. Sanchez E, Kowalski BR (1990) Tensorial resolution: a direct trilinear decomposition. J Chemom 4(1):29–45. doi: 10.1002/cem.1180040105 CrossRefGoogle Scholar
  31. Stoer M, Wagner F (1997) A simple min-cut algorithm. J ACM 44(4):585–591. doi: 10.1145/263867.263872 zbMATHMathSciNetCrossRefGoogle Scholar
  32. Sun J, Papadimitriou S, Lin CY, Cao N, Liu S, Qian W (2009) Multivis: content-based social network exploration through multi-way visual analysis. In: Proceedings SDM, vol 9. SIAM, pp. 1063–1074Google Scholar
  33. Sun J, Tao D, Papadimitriou S, Yu PS, Faloutsos C (2008) Incremental tensor analysis: theory and applications. ACM Trans Knowl Discov Data 2(3):11:1–11:37. doi: 10.1145/1409620.1409621 CrossRefGoogle Scholar
  34. Tang J, Zhang J, Yao L, Li J, Zhang L, Su Z (2008) Arnetminer: extraction and mining of academic social networks. In: Proceedings of the 14th ACM SIGKDD international conference on knowledge discovery and data mining. ACM, pp. 990–998Google Scholar
  35. Tsourakakis CE (2010) Mach: fast randomized tensor decompositions. In: Proceedings of the 10th SIAM International Conference on Data Mining, pp. 689–700Google Scholar
  36. Tucker L (1966) Some mathematical notes on three-mode factor analysis. Psychometrika 31(3):279–311. doi: 10.1007/BF02289464 MathSciNetCrossRefGoogle Scholar
  37. Wyss C, Giannella C, Robertson EL (2001) Fastfds: a heuristic-driven, depth-first algorithm for mining functional dependencies from relation instances: extended abstract. In: Proceedings of the Third International Conference on Data Warehousing and Knowledge Discovery, DaWaK ’01. Springer, London, pp 101–110Google Scholar
  38. Xu Z, Yan F, Qi A (2012) Infinite tucker decomposition: nonparametric bayesian models for multiway data analysis. In: ICML. Scholar
  39. Zhang Q, Berry M, Lamb B, Samuel T, Allen G, Nabrzyski J, Seidel E, van Albada G, Dongarra J, Sloot P (2009) A parallel nonnegative tensor factorization algorithm for mining global climate data, vol 5545. Springer, Berlin/Heidelberg, pp. 405–415Google Scholar
  40. Zhou G, He Z, Zhang Y, Zhao Q, Cichocki A (2009) Canonical polyadic decomposition: from 3-way to n-way. In: Eighth international conference on computational intelligence and security (CIS), pp 391–395. doi: 10.1109/CIS.2012.94

Copyright information

© The Author(s) 2015

Authors and Affiliations

  1. 1.Arizona State UniversityTempeUSA

Personalised recommendations