Skip to main content

Advertisement

Log in

Two new kinds of protoconcepts based on three-way decisions model

  • Foundation, algebraic, and analytical methods in soft computing
  • Published:
Soft Computing Aims and scope Submit manuscript

Abstract

With the development of formal concept analysis, classical concept lattice cannot solve some problems in practice because of its strict conditions. Protoconcept, which is an extension of the formal concept, provides a new method of data processing. Three-way decisions is more in line with human cognition. Not like the decision of two ways, three-way cognition expands its scope of application. However, up to now, protoconcept is influenced by two-way decisions which leads to the restrictions of its applications. Hence, it is urgent to find a method to generalize protoconcept from the framework of two-way decisions. To give the solution to this urgent problem, the main contributions of this paper are the following: First, using the combination of protoconcept and three-way decisions, two models of three-way protoconcept are obtained. The algorithms for searching three-way protoconcept are built. Second, we investigate the constructions of three-way protoconcepts in double Boolean algebra. Third, the relationships between protoconcepts and three-way protoconcepts are found. Moreover, using examples that come from the real world, we explain the main results obtained above and show some applications. The two models for protoconcepts provided in this paper generalize the fields of thought from two-way decisions to three-way decisions. This follows that they are more in line with human thinking. Hence, the two models will certainly generalize the applications of protoconcept theory in the future.

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

Similar content being viewed by others

Data availability

Enquiries about data availability should be directed to the authors.

References

Download references

Funding

The authors have not disclosed any funding.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Hua Mao.

Ethics declarations

Conflict of interest

All authors declare that there is no conflict of interests regarding the publication of this manuscript.

Ethical approval

This manuscript does not contain any studies with human participants or animals performed by any of the authors. This manuscript is the authors’ original work and has not been published nor has it been submitted simultaneously elsewhere.

Additional information

Publisher's Note

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

Appendix

Appendix

1.Proof of Property 1

According to the definition of operators \(\sqcap ,\sqcup ,\wedge ,\vee \), we have

\(c\sqcap d=(E_{1} \cap E_{2},(E_{1} \cap E_{2})^{\lessdot })\),

\(c\sqcup d=(((M_{1},N_1) \cap (M_{2},N_2))^{ > rdot }, (M_{1},N_1) \cap (M_{2},N_2))\),

\(c\wedge d=(((M_{1},N_1)\cup (M_{2},N_2))^{ > rdot },(M_{1},N_1)\cup (M_{2},N_2))\),

\(c\vee d=(E_{1} \cup E_{2},(E_{1} \cup E_{2})^{\lessdot })\).

\(\bullet \) To prove \(c\sqcap d \le c,d \le c \sqcup d\).

  • We know \(E_1\cap E_2\subseteq E_1 \ \text {and} \ E_1 \cap E_2\subseteq E_2\) \({\mathop {\Longrightarrow }\limits ^{(C6),(C1)}}\) \((E_1\cap E_2)^{\lessdot } \supseteq E_1^{\lessdot }\cup E_2^{\lessdot }=(M_1,N_1)^{ > rdot \lessdot } \cup (M_2,N_2)^{ > rdot \lessdot } \supseteq (M_1,N_1) \cup (M_2,N_2) \supseteq (M_1,N_1)\) and \((E_1\cap E_2)^{\lessdot }\supseteq (M_2,N_2)\). Thus, \(c\sqcap d \le c,d \le c \sqcup d \).

\(\bullet \) To prove \(c\sqcap d \le c \wedge d.\)

  • By (C6) and (C1), we know \((E_{1} \cap E_{2})^{\lessdot }\supseteq E_{1}^{\lessdot }\cup E_{2}^{\lessdot }=(M_1,N_1)^{ > rdot \lessdot } \cup (M_2,N_2)^{ > rdot \lessdot } \supseteq (M_1,N_1)\cup (M_2,N_2)\).

  • According to (C5 and (C1), it holds \(((M_{1},N_1) \cup (M_{2},N_2))^{ > rdot }= (M_{1},N_1)^{ > rdot } \cap (M_{2},N_2)^{ > rdot }=E_1^{\lessdot > rdot } \cap E_2^{\lessdot > rdot } \supseteq E_1 \cap E_2 \) \({\mathop {\Longrightarrow }\limits ^{ \text {Lem}5}}\) \((E_{1} \cap E_{2},(E_{1} \cap E_{2})^{\lessdot })\le (((M_{1},N_1)\cup (M_{2},N_2))^ { > rdot },(M_{1},N_1)\cup (M_{2},N_2)) \) \(\Longrightarrow \) \(c\sqcap d \le c \wedge d\).

\(\bullet \) To prove \(c\vee d \le c \sqcup d.\)

  • According to (C5) and (C1), we obtain \((E_{1} \cup E_{2})^{\lessdot }=E_1^{\lessdot }\cap E_2^{\lessdot }=(M_1,N_1)^{ > rdot \lessdot }\cap (M_2,N_2)^{ > rdot \lessdot }\supseteq (M_1,N_1)\cap (M_2,N_2)\).

  • Then, by (C6) and (C1), we have \(((M_{1},N_1) \cap (M_{2},N_2))^{ > rdot }\supseteq E_1\cup E_2\) \({\mathop {\Longrightarrow }\limits ^{\text {Lem }5}}\) \((E_{1} \cup E_{2},(E_{1} \cup E_{2})^{\lessdot }) \le (((M_{1},N_1) \cap (M_{2},N_2))^{ > rdot }, (M_{1},N_1) \cap (M_{2},N_2))\) \(\Longrightarrow \) \(c \vee d \le c\sqcup d\). \(\square \)

2.Proof of Theorem 2

According to Definition 4, we should verify \(({\mathfrak {B}} (T),\sqcap ,\sqcup ,\lnot ,\lrcorner ,\bot ,\top )\) satisfying properties (1a), (1b), ..., (11a), (11b), and (12). We can easily prove these properties according to Definition 9, except (5a) and (12). Hence, the proofs of other properties are omitted. Suppose cdz are \((E_{1}, (M_{1},N_1))\), \((E_{2}, (M_2,N_2))\), and \((E_{3}, (M_{3},N_3))\), respectively.

\(\bullet \) To prove (5a).

  • We can acquire \(c \sqcap c=(E_1,E_1^{\lessdot })\) and \(c \sqcap (c \sqcup d)=(E_1\cap ((M_1,N_1)\cap (M_2,N_2))^{ > rdot },(E_1\cap (M_1,N_1)\cap (M_2,N_2))^{ > rdot })^{\lessdot })\).

  • If \((E_1,(M_1,N_1))\) is OE-protoconcept \({\mathop {\Longrightarrow }\limits ^{\text {Def } 8}}\) \((M_1,N_1)^{ > rdot }=E_1^{\lessdot > rdot }\) \({\mathop {\Longrightarrow }\limits ^{(C1)}}\) \(((M_1,N_1)\cap (M_2,N_2))^{ > rdot }\supseteq (M_1,N_1)^{ > rdot }=E_1^{\lessdot > rdot }\supseteq E_1\) \(\Longrightarrow \) \(c \sqcap (c \sqcup d)=(E_1,E_1^{\lessdot })=c \sqcap c\).

\(\bullet \) To prove (12).

  • \((E_1,(M_1,N_1))\) is OE-protoconcept \({\mathop {\Longrightarrow }\limits ^{\text {Prop }1}}\) \(E_1^{\lessdot > rdot }=(M_1,N_1)^{ > rdot }\) and \(E_1^{\lessdot }= (M_1,N_1)^{ > rdot \lessdot }\).

  • Due to \((c \sqcap c) \sqcup (c \sqcap c){\mathop {=}\limits ^{\text {Def }9}} (E_1^{\lessdot > rdot },E_1^{\lessdot })\) and \((c \sqcup c) \sqcap (c \sqcup c) {\mathop {=}\limits ^{\text {Def }9}} ((M_1,N_1)^{ > rdot },(M_1,N_1)^{ > rdot \lessdot })\) \({\mathop {\Longrightarrow }\limits ^{\text {Prop }1}}\) \((c \sqcap c) \sqcup (c \sqcap c)=(c \sqcup c) \sqcap (c \sqcup c)\). \(\square \)

3.Proof of Theorem 5

Let (EM) be a protoconcept. Then \((E,E^{\lessdot })\) is an OE-protoconcept by Definition 8.

\(\bullet \) To prove \(\varphi \) is \(\sqcap \)-preserving mapping.

  • Let \((E_1,M_1)\) and \((E_2,M_2)\) be any two protoconcepts. Then, by Definition 9, we can get that \(\varphi (E_1,M_1)\sqcap \varphi (E_2,M_2)=(E_1,E_1^{\lessdot })\sqcap (E_2,E_2^{\lessdot })=(E_1\cap E_2,(E_1\cap E_2)^{\lessdot })\) and \(\varphi ((E_1,M_1) \sqcap (E_2,M_2)) =\varphi ((E_1\cap E_2,(E_1 \cap E_2)^{\lessdot }))=(E_1\cap E_2,(E_1\cap E_2)^{\lessdot })\). Thus, we confirm that \(\varphi (E_1,M_1)\sqcap \varphi (E_2,M_2)\) = \(\varphi ((E_1,M_1) \sqcap (E_2,M_2))\). Then, we obtain \(\varphi \) is a \(\sqcap \)-preserving mapping.

\(\bullet \) To prove \(\varphi \) is \(\vee \)-preserving mapping.

  • Let \((E_1,M_1)\) and \((E_2,M_2)\) be any two protoconcepts. Then, using Definition 9, we can know that \(\varphi (E_1,M_1)\vee \varphi (E_2,M_2)= (E_1,E_1^{\lessdot }) \vee (E_2,E_2^{\lessdot })=(E_1\cup E_2,(E_1\cup E_2)^{\lessdot })\) and \(\varphi ((E_1,M_1) \vee (E_2,M_2)) =\varphi ((E_1\cup E_2,(E_1\cup E_2)^{\lessdot }))=(E_1\cup E_2,(E_1\cup E_2)^{\lessdot })\). Thus, \(\varphi (E_1,M_1)\vee \varphi (E_2,M_2)\) = \(\varphi ((E_1,M_1) \vee (E_2,M_2))\) holds. we get \(\varphi \) is a \(\vee \)-preserving mapping.

\(\bullet \) To prove \(\varphi \) is order-preserving mapping.

  • \((E_1,M_1)\le (E_2,M_2)\) \({\mathop {\Longrightarrow }\limits ^{\text {Lem }3}}\) \( E_1 \subseteq E_2\) \({\mathop {\Longrightarrow }\limits ^{\text {(C2)}}}\) \(E_1^{\lessdot } \supseteq E_2^{\lessdot }\) \({\mathop {\Longrightarrow }\limits ^{\text {Lem }3}}\) \((E_1,E_1^{\lessdot })\le (E_2,E_2^{\lessdot })\) \({\mathop {\Longrightarrow }\limits ^{\text {Theo }5}}\) \(\varphi (E_1,M_1)\le \varphi (E_2,M_2)\) \({\mathop {\Longrightarrow }\limits ^{\text {Def }11}}\) \(\varphi \) is an order-preserving mapping.

\(\square \)

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) 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

Mao, H., Ma, J. & Niu, Z. Two new kinds of protoconcepts based on three-way decisions model. Soft Comput 27, 11973–11984 (2023). https://doi.org/10.1007/s00500-023-08840-3

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00500-023-08840-3

Keywords

Navigation