Skip to main content
Log in

Multiview granular data analytics based on three-way concept analysis

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

Multiview granular data analytics reflects various aspects of the knowledge embodied in data by multiple granular structures. It has been investigated in many topics closely related to granular computing, especially those researches involving formal concept analysis. Three-way concept analysis has demonstrated its usefulness for knowledge discovery in formal contexts, since it can extract positive information and negative information between objects and attributes simultaneously. Taking advantage of it, we propose a concrete model of multiview granular data analytics based on three-way concept analysis. Firstly, two hexagons of trisections in three-way concept analysis are presented. The hexagons reveal two different trisection forms in existing three-way concept lattice models. These models are then accordingly grouped into two classes, namely, orthopair-based weak tri-partition model and weak tri-covering model. Secondly, interval-set-based weak tri-partition model of three-way concept lattices is designed by reformulating the knowledge ordering of interval sets. More specifically, sufficiency-possibility three-way concept lattices and necessity-dual three-way concept lattices are defined on the basis of different combinations of modal-style operators. Finally, the transformation methods among various types of three-way concept lattices are explored by analyzing their relationships. Further interpretations of the hidden semantics in these relationships are also given in terms of trisection.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15

Similar content being viewed by others

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

References

  1. Wille R (1982) Restructuring lattice theory: an approach based on hierarchies of concepts. In: Rival I (ed) Ordered sets. Reidel Publishing Company, Dordrecht–Boston, pp 445–470

  2. Ganter B, Wille R (1999) Formal concept analysis: mathematical foundations. Springer-Verlag, Berlin

    Book  MATH  Google Scholar 

  3. Zhang W, Wei L, Qi J (2005) Attribute reduction theory and approach to concept lattice. Sci China Ser F Inf Sci 48(6):713–726. https://doi.org/10.1360/122004-104

    Article  MathSciNet  MATH  Google Scholar 

  4. Qin K, Lin H, Jiang Y (2020) Local attribute reductions of formal contexts. Int J Mach Learn Cybernet 11(1):81–93. https://doi.org/10.1007/s13042-019-00956-z

    Article  Google Scholar 

  5. Burmeister P, Holzer R Ganter B, Mineau GW (eds) (2000) On the treatment of incomplete knowledge in formal concept analysis. Springer, Berlin

  6. Yao Y (2020) Three-way granular computing, rough sets, and formal concept analysis. Int J Approx Reason 116:106–125. https://doi.org/10.1016/j.ijar.2019.11.002

    Article  MathSciNet  MATH  Google Scholar 

  7. Xie J, Yang M, Li J, Zheng Z (2018) Rule acquisition and optimal scale selection in multi-scale formal decision contexts and their applications to smart city. Futur Gener Comput Syst 83:564–581. https://doi.org/10.1016/j.future.2017.03.011

    Article  Google Scholar 

  8. Loia V, Orciuoli F, Pedrycz W (2018) Towards a granular computing approach based on formal concept analysis for discovering periodicities in data. Knowl-Based Syst 146:1–11. https://doi.org/10.1016/j.knosys.2018.01.032

    Article  Google Scholar 

  9. Chen J, Zheng H, Wei L, Wan Z, Ren R, Li J, Li H, Bian W, Gao M, Bai Y (2020) Factor diagnosis and future governance of dangerous goods accidents in China’s ports. Environ Pollut 257:113582. https://doi.org/10.1016/j.envpol.2019.113582

    Article  Google Scholar 

  10. Yao Y (2012) An outline of a theory of three-way decisions. In: Yao J, Yang Y, Slowinski R, Greco S, Li H, Mitra S, Polkowski L (eds) Rough sets and current trends in computing, vol. 7413 of lecture notes in computer science. Springer, Berlin, pp 1–17

  11. Qi J, Wei L, Yao Y (2014) Three-way formal concept analysis. In: Miao D, Pedrycz W, Slezak D, Peters G, Hu Q, Wang R (eds) Rough sets and knowledge technology, vol. 8818 of lecture notes in computer science. Springer International Publishing, Cham, pp 732–741

  12. Qi J, Qian T, Wei L (2016) The connections between three-way and classical concept lattices. Knowl-Based Syst 91(1):143–151. https://doi.org/10.1016/j.knosys.2015.08.006

    Article  Google Scholar 

  13. Ren R, Wei L (2016) The attribute reductions of three-way concept lattices. Knowl-Based Syst 99:92–102. https://doi.org/10.1016/j.knosys.2016.01.045https://doi.org/10.1016/j.knosys.2016.01.045

    Article  Google Scholar 

  14. Wei L, Liu L, Qi J, Qian T (2020) Rules acquisition of formal decision contexts based on three-way concept lattices. Inf Sci 516:529–544. https://doi.org/10.1016/j.ins.2019.12.024

    Article  MathSciNet  MATH  Google Scholar 

  15. Qian T, Wei L, Qi J (2017) Constructing three-way concept lattices based on apposition and subposition of formal contexts. Knowl-Based Syst 116:39–48. https://doi.org/10.1016/j.knosys.2016.10.033https://doi.org/10.1016/j.knosys.2016.10.033

    Article  Google Scholar 

  16. Yang S, Lu Y, Jia X, Li W (2020) Constructing three-way concept lattice based on the composite of classical lattices. Int J Approx Reason 121:174–186. https://doi.org/10.1016/j.ijar.2020.03.007

    Article  MathSciNet  MATH  Google Scholar 

  17. Hu Q, Qin K, Yang L (2022) The updating methods of object-induced three-way concept in dynamic formal contexts. Appl Intell. https://doi.org/10.1007/s10489-022-03646-6

  18. Yu H, Li Q, Cai M (2018) Characteristics of three-way concept lattices and three-way rough concept lattices. Knowl-Based Syst 146:181–189. https://doi.org/10.1016/j.knosys.2018.02.007

    Article  Google Scholar 

  19. Huang C, Li J, Mei C, Wu W (2017) Three-way concept learning based on cognitive operators: an information fusion viewpoint. Int J Approx Reason 83:218–242. https://doi.org/10.1016/j.ijar.2017.01.009https://doi.org/10.1016/j.ijar.2017.01.009

    Article  MathSciNet  MATH  Google Scholar 

  20. Li J, Huang C, Qi J, Qian Y, Liu W (2017) Three-way cognitive concept learning via multi-granularity. Inf Sci 378(1):244–263. https://doi.org/10.1016/j.ins.2016.04.051

    Article  MATH  Google Scholar 

  21. Yao Y (2017) Interval sets and three-way concept analysis in incomplete contexts. Int J Mach Learn Cybernet 8(1):3–20. https://doi.org/10.1007/s13042-016-0568-1

    Article  MathSciNet  Google Scholar 

  22. Wang Z, Wei L, Qi J, Qian T (2020) Attribute reduction of SE-ISI concept lattices for incomplete contexts. Soft Comput 24(20):15143–15158. https://doi.org/10.1007/s00500-020-05271-2https://doi.org/10.1007/s00500-020-05271-2

    Article  MATH  Google Scholar 

  23. Qi J, Wei L, Ren R (2021) 3-way concept analysis based on 3-valued formal contexts. Cognitive Computation. https://doi.org/10.1007/s12559-021-09899-6https://doi.org/10.1007/s12559-021-09899-6

  24. He X, Wei L, She Y (2018) L-fuzzy concept analysis for three-way decisions: basic definitions and fuzzy inference mechanisms. Int J Mach Learn Cybernet 9:1857–1867. https://doi.org/10.1007/s13042-018-0857-yhttps://doi.org/10.1007/s13042-018-0857-y

    Article  Google Scholar 

  25. Bartl E, Konecny J (2019) L-concept lattices with positive and negative attributes: modeling uncertainty and reduction of size. Inf Sci 472:163–179. https://doi.org/10.1016/j.ins.2018.08.057

    Article  MathSciNet  MATH  Google Scholar 

  26. Zhao X, Miao D, Fujita H (2021) Variable-precision three-way concepts in L-contexts. Int J Approx Reason 130:107–125. https://doi.org/10.1016/j.ijar.2020.11.005

    Article  MathSciNet  MATH  Google Scholar 

  27. Zhao X, Miao D (2022) Isomorphic relationship between L-three-way concept lattices. Cogn Comput. https://doi.org/10.1007/s12559-021-09902-0https://doi.org/10.1007/s12559-021-09902-0

  28. Chen X, Qi J, Zhu X, Wang X, Wang Z (2020) Unlabelled text mining methods based on two extension models of concept lattices. Int J Mach Learn Cybern 11 (2):475–490. https://doi.org/10.1007/s13042-019-00987-6

    Article  Google Scholar 

  29. Shivhare R, Cherukuri AK (2017) Three-way conceptual approach for cognitive memory functionalities. Int J Mach Learn Cybern 8(1):21–34. https://doi.org/10.1007/s13042-016-0593-0

    Article  Google Scholar 

  30. Yuan K, Xu W, Li W, Ding W (2022) An incremental learning mechanism for object classification based on progressive fuzzy three-way concept. Inf Sci 584:127–147.

  31. Zadeh LA (1979) Fuzzy sets and information granularity. In: Gupta N, Ragade R, Yager R (eds) Advances in fuzzy set theory and applications, North Holland, Amsterdam, pp 3–18

  32. Zadeh LA (1997) Toward a theory of fuzzy information granulation and its centrality in human reasoning and fuzzy logic. Fuzzy Sets Syst 90(2):111–127. https://doi.org/10.1016/S0165-0114(97)00077-8https://doi.org/10.1016/S0165-0114(97)00077-8

    Article  MathSciNet  MATH  Google Scholar 

  33. Yao Y (2016) A triarchic theory of granular computing. Granul Comput 1:145–157. https://doi.org/10.1007/s41066-015-0011-0

    Article  Google Scholar 

  34. Fujita H, Li T, Yao Y (2016) Advances in three-way decisions and granular computing. Knowl-Based Syst 91:1–3. https://doi.org/10.1016/j.knosys.2015.10.026

    Article  Google Scholar 

  35. Yao Y (2018) Three-way decision and granular computing. Int J Approx Reason 103:107–123. https://doi.org/10.1016/j.ijar.2018.09.005https://doi.org/10.1016/j.ijar.2018.09.005

    Article  MATH  Google Scholar 

  36. Fujita H, Gaeta A, Loia V, Orciuoli F (2018) Resilience analysis of critical infrastructures: a cognitive approach based on granular computing. IEEE Trans Cybern 49(5):1835–1848. https://doi.org/10.1109/TCYB.2018.2815178

    Article  Google Scholar 

  37. Liu D, Yang X, Li T (2020) Three-way decisions: beyond rough sets and granular computing. Int J Mach Learn Cybern 11(5):989–1002. https://doi.org/10.1007/s13042-020-01095-6

    Article  Google Scholar 

  38. Pedrycz W (2021) Granular computing: Fundamentals and system modeling. In: C L, Wu M, Pedrycz W (eds) Developments in advanced control and intelligent automation for complex systems. Studies in systems, decision and control, vol 329. Springer, Cham, pp 167–192. https://doi.org/10.1007/978-3-030-62147-6_7

  39. Yang X, Zhang Y, Fujita H, Liu D, Li T (2020) Local temporal-spatial multi-granularity learning for sequential three-way granular computing. Inf Sci 541:75–97. https://doi.org/10.1016/j.ins.2020.06.020

    Article  MathSciNet  MATH  Google Scholar 

  40. Düntsch I, Gediga G (2002) Modal-style operators in qualitative data analysis. In: Proceedings of the 2002 IEEE international conference on data mining, pp 155–162

  41. Yao Y, Słowiṅski R, Komorowski J, Grzymała-Busse JW Tsumoto S (ed) (2004) A comparative study of formal concept analysis and rough set theory in data analysis, vol 3066. Springer, Berlin

  42. Chen Y, Yao Y (2008) A multiview approach for intelligent data analysis based on data operators. Inf Sci 178(1):1–20. https://doi.org/10.1016/j.ins.2007.08.011

    Article  MathSciNet  MATH  Google Scholar 

  43. Ciucci D (2011) Orthopairs: a simple and widely used way to model uncertainty. Fund Inform 108(3):287–304. https://doi.org/10.3233/FI-2011-424

    MathSciNet  MATH  Google Scholar 

  44. Ciucci D (2016) Orthopairs and granular computing. Granul Comput 1 (3):159–170. https://doi.org/10.1007/s41066-015-0013-y

    Article  Google Scholar 

  45. Qian T, Wei L, Qi J (2019) A theoretical study on the object (property) oriented concept lattices based on three-way decisions. Soft Comput 23(19):9477–9489. https://doi.org/10.1007/s00500-019-03799-6

    Article  MATH  Google Scholar 

  46. Zhi H, Qi J, Qian T, Wei L (2019) Three-way dual concept analysis. Int J Approx Reason 114:151–165. https://doi.org/10.1016/j.ijar.2019.08.010

    Article  MathSciNet  MATH  Google Scholar 

  47. Zhao X, Miao D, Hu B (2020) On relationship between three-way concept lattices. Inf Sci 538:396–414. https://doi.org/10.1016/j.ins.2020.06.007

    Article  MathSciNet  MATH  Google Scholar 

  48. Rowley J (2007) The wisdom hierarchy: representations of the DIKW hierarchy. J Inf Sci 33 (2):163–180. https://doi.org/10.1177/0165551506070706

    Article  Google Scholar 

  49. Yao Y (2020) Tri-level thinking: models of three-way decision. Int J Mach Learn Cybern 11:947–959. https://doi.org/10.1007/s13042-019-01040-2

    Article  Google Scholar 

  50. Yao Y (2009) Interval sets and interval-set algebras. In: 2009 8th IEEE international conference on cognitive informatics, pp 307–314. https://doi.org/10.1109/COGINF.2009.5250723

  51. Dubois D, Prade H (2012) From blanché’s hexagonal organization of concepts to formal concept analysis and possibility theory. Log Univers 6:149–169. https://doi.org/10.1007/s11787-011-0039-0

    Article  MathSciNet  MATH  Google Scholar 

  52. Béziau J-Y (2012) The power of the hexagon. Log Univers 6:1–43. https://doi.org/10.1007/s11787-012-0046-9

    Article  MathSciNet  MATH  Google Scholar 

  53. Moore RE, Kearfott RB, Cloud MJ (2009) Introduction to interval analysis. Society for Industrial and Applied Mathematics, Philadelphia

    Book  MATH  Google Scholar 

  54. Zhi H, Qi J (2022) Common-possible concept analysis: a granule description viewpoint. Appl Intell 52(3):2975–2986. https://doi.org/10.1007/s10489-021-02499-9

    Article  Google Scholar 

  55. Wei L, Wang Z, Qi J, Ren R (2022) Necessity-possibility semi-three-way concept (in Chinese). J Southwest China Normal Univ (Nat Sci Ed) 47(4):12–20. https://doi.org/10.13718/j.cnki.xsxb.2022.04.002

    Google Scholar 

  56. Vormbrock B, Wille R (2005) Semiconcept and protoconcept algebras: the basic theorems. In: Ganter B, Stumme G, Wille R (eds) Formal concept analysis. Lecture notes in computer science, vol 3626. Springer, Berlin, pp 34–48

  57. Yang D, Deng T, Fujita H (2020) Partial-overall dominance three-way decision models in interval-valued decision systems. Int J Approx Reason 126:308–325. https://doi.org/10.1016/j.ijar.2020.08.014

    Article  MathSciNet  MATH  Google Scholar 

  58. Xu Y, Li B (2022) Multiview sequential three-way decisions based on partition order product space. Inf Sci 600:401–430. https://doi.org/10.1016/j.ins.2022.04.007

    Article  Google Scholar 

  59. Ye J, Zhan J, Ding W, Fujita H (2022) A novel three-way decision approach in decision information systems. Inf Sci 584:1–30. https://doi.org/10.1016/j.ins.2021.10.042

    Article  Google Scholar 

  60. Yang X, Chen Y, Fujita H, Liu D, Li T (2022) Mixed data-driven sequential three-way decision via subjective-objective dynamic fusion. Knowl-Based Syst 237:107728. https://doi.org/10.1016/j.knosys.2021.107728

    Article  Google Scholar 

  61. Yao Y (2019) Three-way conflict analysis: Reformulations and extensions of the Pawlak model. Knowl-Based Syst 180:26–37. https://doi.org/10.1016/j.knosys.2019.05.016

    Article  Google Scholar 

  62. Doignon J-P, Falmagne J-C (1985) Spaces for the assessment of knowledge. Int J Man-Mach Stud 23(2):175–196. https://doi.org/10.1016/S0020-7373(85)80031-6

    Article  MATH  Google Scholar 

  63. Zhi H, Qi J, Qian T, Ren R (2020) Conflict analysis under one-vote veto based on approximate three-way concept lattice. Inf Sci 516:316–330. https://doi.org/10.1016/j.ins.2019.12.065

    Article  MathSciNet  Google Scholar 

  64. Lang G, Luo J, Yao Y (2020) Three-way conflict analysis: a unification of models based on rough sets and formal concept analysis. Knowl-Based Syst 194:105556. https://doi.org/10.1016/j.knosys.2020.105556

    Article  Google Scholar 

  65. Sun W, Li J, Ge X, Lin Y (2021) Knowledge structures delineated by fuzzy skill maps. Fuzzy Sets Syst 407:50–66. https://doi.org/10.1016/j.fss.2020.10.004

    Article  MathSciNet  MATH  Google Scholar 

  66. Zhou Y, Li J, Wang H, Sun W (2022) Skills and fuzzy knowledge structures. J Intell Fuzzy Syst 42(3):2629–2645. https://doi.org/10.3233/JIFS-212018

    Article  Google Scholar 

Download references

Acknowledgments

This work is partially supported by the National Natural Science Foundation of China (Nos. 12171392, 62006190, 61772021, 12101478), China Scholarship Council (No. 202006970030), and the Natural Science Basic Research Program of Shaanxi (Program No. 2021JM-141).

Author information

Authors and Affiliations

Authors

Corresponding authors

Correspondence to Jianjun Qi or Ling Wei.

Ethics declarations

Ethics Approval

This article does not contain any studies with human participants or animals performed by the author.

Conflict of Interests

The authors have no relevant financial or non-financial interests to disclose.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix: Proofs

Appendix: Proofs

We just present the proofs of Proposition 4, Lemma 1, Theorems 2, 6, 8 and 10, and the other results can be proved in a similar way.

1.1 Proof of Theorem 2

Proof

We only need to prove statements (i) to (iii), since (iv) to (vi) can be proved dually. If X = Ø, then statements (i) and (ii) hold naturally based on the definition of possibility operator. Let X be a nonempty object subset, then we can know that XØ, \(X^{\overline \diamond }\not =\O \). If there exists a xG such that X = {gGgI = xI}, then \(X^{\diamondsuit }\cap X^{\overline \diamondsuit }=\O \) holds. Hence, we have that \(X^{*}\cup X^{\overline *}\)=\(X^{\diamondsuit c}\cup X^{\overline \diamondsuit c}=(X^{\diamondsuit }\cap X^{\overline \diamondsuit })^{c}=M\). It follows that \(X^{\diamondsuit }\cap (X^{*}\cup X^{\overline *})\not =\O \), \(X^{\overline \diamondsuit }\cap (X^{*}\cup X^{\overline *})\not =\O \), \(X^{\diamondsuit }\cup X^{\overline \diamondsuit }\cup (X^{*}\cup X^{\overline *})=M\). A complementary case is that there is no such element x, then we have that \(X^{\diamondsuit }\cap X^{\overline \diamondsuit }\not =\O \). Thus, \(X^{\diamondsuit }\cap X^{\overline \diamondsuit }\subseteq X^{\diamondsuit }\) and \(X^{\diamondsuit }\cap X^{\overline \diamondsuit }\subseteq X^{\overline \diamondsuit }\) hold. It follows that \(X^{\diamondsuit }\cup X^{\overline \diamondsuit }\cup (X^{*}\cup X^{\overline *})=X^{\diamondsuit }\cup X^{\overline \diamondsuit }\cup (X^{\diamondsuit }\cap X^{\overline \diamondsuit })^{c}=M\). Therefore, statements (i), (ii) and (iii) hold. □

1.2 Proof of Proposition 4

Proof

(i) For any X ∈ 2G, \(X^{\vartriangle \triangledown }=[X^{\Box }, X^{\#}]^{\triangledown }=X^{\Box \diamondsuit }\cup X^{\#\#}\subseteq X\) holds, since \(X^{\Box \diamondsuit }\subseteq X\) and \(X^{\#\#}\subseteq X\). We also have that \([\underline A, \overline A]^{\triangledown \vartriangle }=(\underline A^{\diamondsuit }\cup \overline A^{\#})^{\vartriangle }=[(\underline A^{\diamondsuit }\cup \overline A^{\#})^{\Box }, (\underline A^{\diamondsuit }\cup \overline A^{\#})^{\#}]\) for any interval set \([\underline A, \overline A]\in \mathbb {I}(2^{M})\). Based on Proposition 1, we can obtain that \( \underline A^{\diamondsuit \Box }\cup \overline A^{\#\Box }\subseteq (\underline A^{\diamondsuit }\cup \overline A^{\#})^{\Box }\), and \((\underline A^{\diamondsuit }\cup \overline A^{\#})^{\#}\subseteq \underline A^{\diamondsuit \#}\cap \overline A^{\#\#}\). Since \(\underline A\subseteq \underline A^{\diamondsuit \Box }\) and \( \overline A^{\#\#}\subseteq \overline A\), then we have that \(\underline A\subseteq (\underline A^{\diamondsuit }\cup \overline A^{\#})^{\Box }\) and \((\underline A^{\diamondsuit }\cup \overline A^{\#})^{\#}\subseteq \overline A\). It follows that \([\underline A, \overline A]\sqsubseteq [\underline A, \overline A]^{\triangledown \vartriangle }\).

(ii) Let X1,X2 be two object subsets satisfying \(X_{1}\subseteq X_{2}\). Then, we have that \(X_{1}^{\Box }\subseteq X_{2}^{\Box }\) and \(X_{2}^{\#}\subseteq X_{1}^{\#}\). It follows that \(X_{1}^{\vartriangle }\sqsubseteq X_{2}^{\vartriangle }\). For any two interval sets \([\underline {A_{1}},\overline {A_{1}}], [\underline {A_{2}},\overline {A_{2}}]\), \([\underline {A_{1}},\overline {A_{1}}]\sqsubseteq [\underline {A_{2}},\overline {A_{2}}]\) implies that \(\underline {A_{1}}\subseteq \underline {A_{2}}\subseteq \overline {A_{2}}\subseteq \overline {A_{1}}\) or \([\underline {A_{2}},\overline {A_{2}}]=[\O ,\O ]\). If \([\underline {A_{2}},\overline {A_{2}}]=[\O ,\O ]\), then \([\underline {A_{1}},\overline {A_{1}}]^{\triangledown }\subseteq [\underline {A_{2}},\overline {A_{2}}]^{\triangledown }\) holds based on the definition of OEND-operators. If \(\underline {A_{1}}\subseteq \underline {A_{2}}\subseteq \overline {A_{2}}\subseteq \overline {A_{1}}\) holds, then we have that \(\underline {A_{1}}^{\diamondsuit }\subseteq \underline {A_{2}}^{\diamondsuit }\) and \(\overline {A_{1}}^{\#}\subseteq \overline {A_{2}}^{\#}\). It follows that \([\underline {A_{1}},\overline {A_{1}}]^{\triangledown }=\underline {A_{1}}^{\diamondsuit }\cup \overline {A_{1}}^{\#} \subseteq \underline {A_{2}}^{\diamondsuit }\cup \overline {A_{2}}^{\#}=[\underline {A_{2}},\overline {A_{2}}]^{\triangledown }\).

(iii) For any X ∈ 2G, to prove that \(X^{\vartriangle \triangledown \vartriangle }=X^{\vartriangle }\), we need to prove \(X^{\vartriangle \triangledown \vartriangle }\subseteq X^{\vartriangle }\) and \(X^{\vartriangle \triangledown \vartriangle }\supseteq X^{\vartriangle }\) (that can be proved by (i) directly) hold simultaneously. From (i), we know that \(X\supseteq X^{\vartriangle \triangledown }\), and we can also obtain \(X^{\vartriangle }\supseteq X^{\vartriangle \triangledown \vartriangle }\) based on (ii). Thus, \(X^{\vartriangle \triangledown \vartriangle }=X^{\vartriangle }\) holds. Similarly, we can prove that \([\underline A, \overline A]^{\triangledown }=[\underline {A}, \overline {A}]^{\triangledown \vartriangle \triangledown }\).

(iv) Let X1,X2 be two object subsets. We have that \((X_{1}\cap X_{2})^{\vartriangle }=[(X_{1}\cap X_{2})^{\Box },(X_{1}\cap X_{2})^{\#}]=[X_{1}^{\Box }\cap X_{2}^{\Box }, X_{1}^{\#}\cup X_{2}^{\#}]=[X_{1}^{\Box }, X_{1}^{\#}]\sqcap [X_{2}^{\Box }, X_{2}^{\#}]\)=\(X_{1}^{\vartriangle }\sqcap X_{2}^{\vartriangle }\). Supposing that \([\underline {A_{1}},\overline {A_{1}}], [\underline {A_{2}},\overline {A_{2}}]\in \mathbb {I}(2^{M})\) are two interval sets, we obatain that \(([\underline {A_{1}},\overline {A_{1}}]\sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown }=[\underline {A_{1}}\cup \underline {A_{2}}, \overline {A_{1}}\cap \overline {A_{2}}]^{\triangledown }=(\underline {A_{1}}\cup \underline {A_{2}})^{\diamondsuit }\cup (\overline {A_{1}}\cap \overline {A_{2}})^{\#}=(\underline {A_{1}}^{\diamondsuit }\cup \underline {A_{2}}^{\diamondsuit })\cup (\overline {A_{1}}^{\#}\cup \overline {A_{2}}^{\#})=(\underline {A_{1}}^{\diamondsuit }\cup \overline {A_{1}}^{\#})\cup (\underline {A_{2}}^{\diamondsuit }\cup \overline {A_{2}}^{\#})=[\underline {A_{1}},\overline {A_{1}}]^{\triangledown }\cup [\underline {A_{2}},\overline {A_{2}}]^{\triangledown }\).

(v) Let X1,X2 be two object subsets. It follows that \((X_{1}\cup X_{2})^{\vartriangle }=[(X_{1}\cup X_{2})^{\Box },(X_{1}\cup X_{2})^{\#}]\sqsupseteq [X_{1}^{\Box }\cup X_{2}^{\Box }, X_{1}^{\#}\cap X_{2}^{\#}]=[X_{1}^{\Box }, X_{1}^{\#}]\sqcup [X_{2}^{\Box }, X_{2}^{\#}]=X_{1}^{\vartriangle }\sqcup X_{2}^{\vartriangle }\). □

1.3 Proof of Theorem 6

Proof

We only prove the equation \((X_{1}, [\underline {A_{1}},\overline {A_{1}}]) \vee (X_{2}, [\underline {A_{2}},\overline {A_{2}}])= (X_{1} \cup X_{2}, ([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle })\). The equation of infimum of any two OEND-concepts can be proved dually.

First, we prove that the right of the equation is an OEND-concept. That is, we need to prove \( (X_{1} \cup X_{2})^{\vartriangle }=([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle }\) and \(([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle \triangledown }=X_{1} \cup X_{2}\). We first prove that \( (X_{1} \cup X_{2})^{\vartriangle }=([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle }\). Since \((X_{1}, [\underline {A_{1}},\overline {A_{1}}])\) and \((X_{2}, [\underline {A_{2}},\overline {A_{2}}])\) are two OEND-concepts, then \(X_{1}^{\vartriangle }=[\underline {A_{1}},\overline {A_{1}}], [\underline {A_{1}},\overline {A_{1}}]^{\triangledown }= X_{1}\), and \(X_{2}^{\vartriangle }=[\underline {A_{2}},\overline {A_{2}}], [\underline {A_{2}},\overline {A_{2}}]^{\triangledown }= X_{2}\) hold. Hence, \( (X_{1} \cup X_{2})^{\vartriangle }=([\underline {A_{1}},\overline {A_{1}}]^{\triangledown }\cup [\underline {A_{2}},\overline {A_{2}}]^{\triangledown })^{\vartriangle }=([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle }\) holds based on (iii) in Proposition 4. Now, we prove that \(([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle \triangledown }=X_{1} \cup X_{2}\). Since \(X_{1} \cup X_{2}=[\underline {A_{1}}, \overline {A_{1}}]^{\triangledown } \cup [\underline {A_{2}},\overline {A_{2}}]^{\triangledown }=([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown }\), then we can know that \(([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle \triangledown }=([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown }=X_{1} \cup X_{2}\) holds. These equations verify the fact that \((X_{1} \cup X_{2}, ([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle })\) is an OEND-concept.

Next, we prove that \((X_{1} \cup X_{2}, ([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle })\) is the least upper bound of \((X_{1}, [\underline {A_{1}},\overline {A_{1}}])\) and \((X_{2}, [\underline {A_{2}},\overline {A_{2}}])\). Evidently, we can know that \((X_{1} \cup X_{2}, ([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle })\) is an upper bound of \((X_{1}, [\underline {A_{1}},\overline {A_{1}}])\) and \((X_{2}, [\underline {A_{2}},\overline {A_{2}}])\). If there is an OEND-concept \((X,[\underline A, \overline A])\) being an upper bound of \((X_{1}, [\underline {A_{1}},\overline {A_{1}}])\) and \((X_{2}, [\underline {A_{2}},\overline {A_{2}}])\), then we have \(X_{1}\subseteq X\) and \(X_{2}\subseteq X\), then \(X_{1}\cup X_{2}\subseteq X\). It indicates that \((X_{1} \cup X_{2}, ([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle })\leqslant (X,[\underline A, \overline A])\). Therefore, we can say that \((X_{1} \cup X_{2}, ([\underline {A_{1}}, \overline {A_{1}}] \sqcup [\underline {A_{2}},\overline {A_{2}}])^{\triangledown \vartriangle })\) is the least upper bound of \((X_{1}, [\underline {A_{1}},\overline {A_{1}}])\) and \((X_{2}, [\underline {A_{2}},\overline {A_{2}}])\). □

1.4 Proof of Theorem 8

Proof

We first prove the equivalence of (i) to (iv), and the equivalence of (v) to (viii) can be obtained in a similar way. The equivalence of (i) and (ii) is proved in Reference [46], then we need to prove that (i)⇔(iii) and (i)⇔(iv) hold simultaneously. Only the proof of (i)⇔(iii) is presented, since (i)⇔(iv) can be proved in a similar way.

$$ \begin{array}{@{}rcl@{}} &&(X,(A,B))\in \text{OECL}(G,M,I) \\ &\overset{\text{Definition}~4}{\Longleftrightarrow}& X^{\lessdot_{*}}=(X^{*},X^{\overline *})=(A,B), (A,B)^{\gtrdot_{*}}=A^{*}\cap B^{\overline *}=X\\ &\overset{\text{Proposition}~2}{\Longleftrightarrow}& (X^{*},X^{\diamondsuit c})=(A,B), A^{*}\cap B^{c\Box}=X \\ &\Longleftrightarrow& [X^{*},X^{\diamondsuit }]=[A,B^{c}], A^{*}\cap B^{c\Box}=X \\ &\overset{\text{Definition}~5}{\Longleftrightarrow} & X^{\triangleleft}=[A,B^{c}], [A,B^{c}]^{\triangleright}=X\\ &\overset{\text{Definition}~6}{\Longleftrightarrow} &(X,[A,B^{c}])\in \text{OEL}_{\text{SP}}(G,M,I). \end{array} $$

The proof is finished by proving (i) and (v) are equivalent.

$$ \begin{array}{@{}rcl@{}} &&(X,(A,B))\in \text{OECL}(G,M,I) \\ &\overset{\text{Definition}~4}{\Longleftrightarrow} & X^{\lessdot_{*}}=(X^{*},X^{\overline *})=(A,B), (A,B)^{\gtrdot_{*}}=A^{*}\cap B^{\overline *}=X\\ &\overset{\text{Proposition}~2}{\Longleftrightarrow} & (X^{c\overline\Box},X^{c\Box})=(A,B), A^{\overline\diamondsuit c}\cap B^{\diamondsuit c}=X \\ &\Longleftrightarrow& (X^{c\Box},X^{c\overline\Box})=(B,A), B^{\diamondsuit}\cap A^{\overline\diamondsuit}=X^{c}\\ &\overset{\text{Definition}~3}{\Longleftrightarrow} & X^{c\lessdot_{\Box}}=(B,A), (B,A)^{\gtrdot_{\diamondsuit}}=X^{c}\\ &\overset{\text{Definition}~4}{\Longleftrightarrow} &(X^{c},(B,A))\in \text{OEOL}(G,M,I). \end{array} $$

1.5 Proof of Lemma 1

Proof

For any X ∈ 2G, \(X^{\lessdot _{*}\gtrdot _{*}}=X^{\lessdot _{\diamondsuit }\gtrdot _{\Box }}\) has been proved in Reference [47]. Now, we prove the remaining equations. It can be known that \(X^{\lessdot _{*}\gtrdot _{*}}=(X^{*},X^{\overline *})^{\gtrdot _{*}}=X^{**}\cap X^{\overline *\overline *}=X^{**}\cap X^{\diamondsuit c\overline *}=X^{**}\cap X^{\diamondsuit \Box }=X^{\triangleleft \triangleright }\) and \(X^{\lessdot _{*}\gtrdot _{*}}=(X^{*},X^{\overline *})^{\gtrdot _{*}}= X^{\overline *\overline *}\cap X^{**}=X^{\overline *\overline *}\cap X^{\overline \diamondsuit c *}=X^{\overline *\overline *}\cap X^{\overline \diamondsuit \overline \Box }=X^{\overline \triangleleft \overline \triangleright }\) hold based on Proposition 2, Definitions 3 and 5. It follows that \(X^{\lessdot _{*}\gtrdot _{*}}=X^{\triangleleft \triangleright }=X^{\overline \triangleleft \overline \triangleright }\). Similarly, we have that \(X^{c\lessdot _\#\gtrdot _\#}=X^{c\lessdot _{\Box }\gtrdot _{\diamondsuit }}=X^{c\vartriangle \triangledown }=X^{c\overline \vartriangle \overline \triangledown }\). On the basis of Proposition 2, it can be obtained that \(X^{\lessdot _{*}\gtrdot _{*}}=X^{**}\cap X^{\overline *\overline *}=X^{c\overline \Box \overline \diamondsuit c}\cap X^{c\Box \diamondsuit c}=(X^{c\overline \Box \overline \diamondsuit }\cup X^{c\Box \diamondsuit })^{c}=X^{c\lessdot _{\Box }\gtrdot _{\diamondsuit } c}\) holds. Therefore, we have that \(X^{\lessdot _{*}\gtrdot _{*}}=X^{\lessdot _{\diamondsuit }\gtrdot _{\Box }}=X^{\triangleleft \triangleright }=X^{\overline \triangleleft \overline \triangleright }= X^{c\lessdot _\#\gtrdot _\# c}=X^{c\lessdot _{\Box }\gtrdot _{\diamondsuit }c}=X^{c\vartriangle \triangledown c}=X^{c\overline \vartriangle \overline \triangledown c}\). □

1.6 Proof of Theorem 10

Proof

We only prove statements (i). (ii) can be proved dually. We first prove the isomorphic relation among OECL(G,M,I), OELSP(G,M,I), OELNSP(G,M,I), and OEPL(G,M,I), the isomorphic relation among another four kinds of OE-concept lattices can obtained in a simlar way. The isomorphic relation between OECL(G,M,I) and OEPL(G,M,I) has been proved in Reference [47], then we only need to prove that OECL(G,M,I)≅OELSP(G,M,I) ≅OELNSP(G,M,I).

Let f : OECL(G,M,I)→OELSP(G,M,I) be a mapping defined by

$$f((X,(A,B)))=(X,[A,B^{c}]).$$

This mapping is evidently a bijection based on Theorem 8. Then, we only need to prove that f is ∨-preserving and ∧-preserving.

Assume (X, (A,B)) and (Y, (E,F)) are two OEC-concepts, then we have that

$$ \begin{array}{@{}rcl@{}} &&f((X,(A,B))\vee(Y,(E,F)))\\ &\overset{\text{Theorem}~1}{=}&f(((X\cup Y)^{\lessdot_{*}\gtrdot_{*}},(A\cap E,B\cap F))) \\ &\overset{\text{Assumption}}{=}&((X\cup Y)^{\lessdot_{*}\gtrdot_{*}},[A\cap E,(B\cap F)^{c}]) \\ &\overset{\text{Lemma}~1}{=}&((X\cup Y)^{\triangleleft\triangleright},[A\cap E, B^{c}\cup F^{c}])\\ &\overset{\text{Theorem}~4}{=}&(X,[A,B^{c}])\vee (Y,[E,F^{c}]) \\ &\overset{\text{Assumption}}{=}&f((X,(A,B)))\vee f((Y,(E,F))). \end{array} $$

It shows that f is ∨-preserving. Dually, we can prove that f((X, (A,B)) ∧ (Y, (E,F))) = f((X, (A,B))) ∧ f((Y, (E,F))). The equation states that f is also ∧-preserving. It follows that f is an isomorphic mapping between OECL(G,M,I) and OELSP(G,M,I). OECL(G,M,I) ≅OELNSP(G,M,I) can be obtained similarly.

We finish the proof by proving \(\text {OECL}(G,M,I)\overset {anti}{\cong }\text {OEOL}(G,M,I)\). Let g : OECL(G,M,I)→OEOL(G,M,I) be a mapping defined by

$$g((X,(A,B)))=(X^{c},(B,A)).$$

This mapping is evidently a bijection based on Theorem 8. Then, we only need to prove that g is anti-∨-preserving and anti-∧-preserving.

Assume (X, (A,B)) and (Y, (E,F)) are two OEC-concepts, then we have that

$$ \begin{array}{@{}rcl@{}} &&g((X,(A,B))\vee(Y,(E,F)))\\ &\overset{\text{Theorem}~1}{=}&g(((X\cup Y)^{\lessdot_{*}\gtrdot_{*}},(A\cap E,B\cap F))) \\ &\overset{\text{Assumption}}{=}&((X\cup Y)^{\lessdot_{*}\gtrdot_{*}c},(B\cap F, A\cap E)) \\ &\overset{\text{Lemma}~1}{=}&((X^{c}\cap Y^{c})^{\lessdot_{\Box}\gtrdot_{\diamondsuit}},(B\cap F, A\cap E))\\ &\overset{\text{Theorem}~1}{=}&(X^{c},(B,A))\wedge (Y,(E,F)) \\ &\overset{\text{Assumption}}{=}&g((X,(A,B)))\wedge g((Y,(E,F))). \end{array} $$

It shows that g is anti-∨-preserving. Dually, we can prove that g((X, (A,B)) ∧ (Y, (E,F))) = g((X, (A,B))) ∨ g((Y, (E,F))). The equation states that g is anti-∧-preserving. It follows that g is an anti-isomorphic mapping between OECL(G,M,I) and OEOL(G,M,I). □

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, Z., Qi, J., Shi, C. et al. Multiview granular data analytics based on three-way concept analysis. Appl Intell 53, 14645–14667 (2023). https://doi.org/10.1007/s10489-022-04145-4

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-022-04145-4

Keywords

Navigation