Abstract
Uncertainty quantification of mass functions is a crucial and unsolved issue in belief function theory. Previous studies have mostly considered this problem from the perspective of viewing the belief function theory as an extension of probability theory. Recently, Yang and Han have developed a new distance-based total uncertainty measure directly and totally based on the framework of belief function theory so that there is no switch between the frameworks of belief function theory and probability theory in that measure. However, we have found some obvious deficiencies in Yang and Han’s uncertainty measure which could lead to counter-intuitive results in some cases. In this paper, an improved distance-based total uncertainty measure has been proposed to overcome the limitations of Yang and Han’s uncertainty measure. The proposed measure not only retains the desired properties of original measure, but also possesses higher sensitivity to the change of evidences. A number of examples and applications have verified the effectiveness and rationality of the proposed uncertainty measure.
Similar content being viewed by others
References
Abellán J (2011) Combining nonspecificity measures in Dempster–Shafer theory of evidence. Int J Gen Syst 40(6):611–622
Abellán J, Masegosa A (2008) Requirements for total uncertainty measures in Dempster–Shafer theory of evidence. Int J Gen Syst 37(6):733–747
Dempster AP (1967) Upper and lower probabilities induced by a multivalued mapping. Ann Math Stat 38 (2):325–339
Deng X, Hu Y, Deng Y, Mahadevan S (2014a) Environmental impact assessment based on D numbers. Expert Syst Appl 41(2):635–643
Deng X, Hu Y, Deng Y, Mahadevan S (2014b) Supplier selection using AHP methodology extended by D numbers. Expert Syst Appl 41(1):156–167
Deng X, Lu X, Chan FT, Sadiq R, Mahadevan S, Deng Y (2015) D-CFPR: D numbers extended consistent fuzzy preference relations. Knowl-Based Syst 73:61–68
Deng X, Han D, Dezert J, Deng Y, Shyr Y (2016a) Evidence combination from an evolutionary game theory perspective. IEEE Trans Cybern 46(9):2070–2082
Deng X, Liu Q, Deng Y (2016b) Matrix games with payoffs of belief structures. Appl Math Comput 273:868–879
Deng X, Liu Q, Deng Y, Mahadevan S (2016c) An improved method to construct basic probability assignment based on the confusion matrix for classification problem. Inf Sci 340–341:250–261
Deng Y (2015) Generalized evidence theory. Appl Intell 43(3):530–543
Deng Y (2016) Deng entropy. Chaos, Solitons & Fractals 91:549–553
Deng Y (2017) Fuzzy analytical hierarchy process based on canonical representation on fuzzy numbers. J Comput Anal Appl 22(2):201–228
Dezert J, Tchamova A (2014) On the validity of Dempster’s fusion rule and its interpretation as a generalization of bayesian fusion rule. Int J Intell Syst 29(3):223–252
Dubois D, Prade H (1985) A note on measures of specificity for fuzzy sets. Int J Gen Syst 10(4):279–283
Dubois D, Prade HM, Farreny H, Martin-Clouaire R, Testemale C (1988) Possibility Theory: An Approach to Computerized Processing of Uncertainty, vol 2. Plenum Press, New York
Duda RO, Hart PE, Stork DG (2012) Pattern Classification. John Wiley & Sons
Fan G, Zhong D, Yan F, Yue P (2016) A hybrid fuzzy evaluation method for curtain grouting efficiency assessment based on an AHP method extended by D numbers. Expert Syst Appl 44(1):289–303
Ferreira AA, Gonċalves M A, Laender AH (2012) A brief survey of automatic methods for author name disambiguation. Acm Sigmod Record 41(2):15–26
Fisher RA (1936) The use of multiple measurements in taxonomic problems. Ann Eugen 7(2):179–188
Fu C, Yang JB, Yang SL (2015) A group evidential reasoning approach based on expert reliability. Eur J Oper Res 246(3):886–893
Guo J (2016) A risk assessment approach for failure mode and effects analysis based on intuitionistic fuzzy sets and evidence theory. J Intell Fuzzy Syst 30(2):869–881
Han D, Dezert J, Duan Z (2016) Evaluation of probability transformations of belief functions for decision making. IEEE Trans Syst Man Cybern Syst Hum 46(1):93–108
Harmanec D, Klir GJ (1994) Measuring total uncertainty in Dempster-Shafer theory: A novel approach. Int J Gen Syst 22(4):405–419
Hartley RV (1928) Transmission of information. Bell Syst Tech J 7(3):535–563
Höhle U (1982) Entropy with respect to plausibility measures. In: Proceedings of the 12th IEEE International Symposium on Multiple-Valued Logic, pp 167–169
Jiang W, Wei B, Qin X, Zhan J, Tang Y (2016a) Sensor Data Fusion Based on a New Conflict Measure. Mathematical Problems in Engineering 2016, Article ID 5769061:11 pages. doi:10.1155/2016/5769061
Jiang W, Wei B, Xie C, Zhou D (2016b) An evidential sensor fusion method in fault diagnosis. Adv Mech Eng 8(3):1–7
Jiang W, Wei B, Zhan J, Xie C, Zhou D (2016c) A visibility graph power averaging aggregation operator: A methodology based on network analysis. Comput Ind Eng 101:260–268. doi:10.1016/j.cie.2016.09.009
Jiang W, Xie C, Wei B, Zhou D (2016d) A modified method for risk evaluation in failure modes and effects analysis of aircraft turbine rotor blades. Adv Mech Eng 8(4):1–16
Jiang W, Xie C, Zhuang M, Shou Y, Tang Y (2016e) Sensor data fusion with Z-numbers and its application in fault diagnosis. Sensors 16(9):1509. doi:10.3390/s16091509
Jiang W, Zhan J, Zhou D, Li X (2016f) A method to determine generalized basic probability assignment in the open world. Mathematical Problems in Engineering 2016:Article ID 3878,634
Jousselme AL, Liu C, Grenier D, Bossé É (2006) Measuring ambiguity in the evidence theory. IEEE Trans Syst Man Cybern Syst Hum 36(5):890–903
Karlsson A, Hammarfelt B, Steinhauer HJ, Falkman G, Olson N, Nelhans G, Nolin J (2015) Modeling uncertainty in bibliometrics and information retrieval: an information fusion approach. Scientometrics 102 (3):2255–2274
Klir GJ, Lewis IIIHW (2008) Remarks on measuring ambiguity in the evidence theory. IEEE Trans Syst Man Cybern Syst Hum 38(4):995–999
Klir GJ, Parviz B (1992) A note on the measure of discord. In: Proceedings of the Eighth international conference on Uncertainty in artificial intelligence, Morgan Kaufmann Publishers Inc., pp 138–141
Klir GJ, Ramer A (1990) Uncertainty in the dempster-shafer theory: a critical re-examination. Int J Gen Syst 18(2):155–166
Klir GJ, Smith RM (1999) Recent developments in generalized information theory. Int J Fuzzy Syst 1 (1):1–13
Klir GJ, Wierman MJ (1999) Uncertainty-based information: elements of generalized information theory, vol 15. Springer Science & Business Media
Liu HC, You JX, Fan XJ, Lin QL (2014a) Failure mode and effects analysis using D numbers and grey relational projection method. Expert Syst Appl 41(10):4670–4679
Liu W, Islamaj Doğan R, Kim S, Comeau DC, Kim W, Yeganova L, Lu Z, Wilbur WJ (2014b) Author name disambiguation for PubMed. J Assoc Inf Sci Technol 65(4):765–781
Liu Zg, Pan Q, Dezert J, Mercier G (2015) Credal c-means clustering method based on belief functions. Knowl.-Based Syst 74:119–132
Zg Liu, Pan Q, Dezert J, Martin A (2016) Adaptive imputation of missing values for incomplete pattern classification. Pattern Recogn 52:85–95
Ma J, Liu W, Miller P, Zhou H (2016) An evidential fusion approach for gender profiling. Inf Sci 333:10–20
Maluf DA (1997) Monotonicity of entropy computations in belief functions. Intell Data Anal 1(3):207–213
Mardani A, Jusoh A, Zavadskas EK (2015) Fuzzy multiple criteria decision-making techniques and applications - two decades review from 1994 to 2014. Expert Syst Appl 42(8):4126–4148
Masson MH, Denoeux T (2008) ECM: An evidential version of the fuzzy c-means algorithm. Pattern Recogn 41(4):1384–1397
Merigó J M, Casanovas M (2009) Induced aggregation operators in decision making with the Dempster-Shafer belief structure. Int J Intell Syst 24(8):934–954
Ning X, Yuan J, Yue X, Ramirez-Serrano A (2014) Induced generalized Choquet aggregating operators with linguistic information and their application to multiple attribute decision making based on the intelligent computing. J Intell Fuzzy Syst 27(3):1077– 1085
Ning X, Yuan J, Yue X (2016a) Uncertainty-based optimization algorithms in designing fractionated spacecraft. Sci Rep 6:22,979
Ning X, Zhang T, Wu Y, Zhang P, Zhang J, Li S, Yue X, Yuan J (2016b) Coordinated parameter identification technique for the inertial parameters of non-cooperative target. PloS one 11(4):e0153,604
Pawlak Z (1982) Rough sets. Int J Comput Inform Sci 11(5):341–356
Pearl J (1990) Reasoning with belief functions: an analysis of compatibility. Int J Approx Reason 4(5-6):363–389
Sabahi F, Akbarzadeh-T MR (2013) A qualified description of extended fuzzy logic. Inf Sci 244:60–74
Shafer G (1976) A mathematical theory of evidence. Princeton University Press, Princeton
Shannon CE (2001) A mathematical theory of communication. ACM SIGMOBILE Mobile Comput Commun Rev 5(1):3–55
Smalheiser NR, Torvik VI (2009) Author name disambiguation. Annual Rev Inf Sci Technol 43(1):1–43
Steinhauer HJ, Karlsson A, Andler SF (2013) Traceable uncertainty. In: The 16th international conference on information fusion, (FUSION), IEEE, pp 1582–1589
Yager RR (1983) Entropy and specificity in a mathematical theory of evidence. Int J Gen Syst 9(4):249–260
Yang Y, Han D (2016) A new distance-based total uncertainty measure in the theory of belief functions. Knowl-Based Syst 94:114–123
Zadeh LA (1965) Fuzzy sets. Inf Control 8(3):338–353
Acknowledgments
The authors are grateful to anonymous reviewers for their useful comments and suggestions on improving this paper. The work is supported by National Natural Science Foundation of China (Grant Nos. 61174022, 61573290, 61503237).
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
Proof of Property 1 Given a BPA m over the FOD Θ, let the belief interval of element 𝜃 i in Θ be [Bel(𝜃 i ),Pl(𝜃 i )]. Since \(0 \le {d_{E}^{I}} \left ({[Bel(\theta _{i}), Pl(\theta _{i})],[0 ,1 ]} \right ) \le 1\), in (21) the upper bound of iTU I(m), which is n, is obtained when \({d_{E}^{I}} \left ({[Bel(\theta _{i}), Pl(\theta _{i})],[0 ,1 ]} \right ) = 0\), ∀𝜃 i ∈Θ. The condition \({d_{E}^{I}} \left ({[Bel(\theta _{i}), Pl(\theta _{i})],[0 ,1 ]} \right ) = 0\), ∀𝜃 i ∈Θ, can uniquely be satisfied in the vacuous BPA m(Θ)=1.
Similarly, the lower bound of iTU I(m), which is 0, is obtained when \({d_{E}^{I}} \left ({[Bel(\theta _{i}), Pl(\theta _{i})],[0 ,1 ]} \right ) = 1\), ∀𝜃 i ∈Θ, which corresponds to [Bel(𝜃 i ),Pl(𝜃 i )]=[0,0] or [Bel(𝜃 i ),Pl(𝜃 i )]=[1,1]. It is easily known that the condition of iTU I getting its lower bound is associated with a BPA m(𝜃 i )=1 and m(A)=0 ∀A≠𝜃 i ,A⊆Θ.
For each 𝜃 i ∈Ω, assume the belief interval is [Bel(𝜃 i ),Pl(𝜃 i )]. Since BPA m is initially defined on Θ, we have [Bel(𝜃 i ),Pl(𝜃 i )]=[0,0] if 𝜃 i ∈Φ and 𝜃 i ∉Θ. Therefore,
Since \(\forall A \subseteq {\Theta } :[Bel_{m_{1} } (A),Pl_{m_{1} } (A)] \subseteq [Bel_{m_{2} } (A),Pl_{m_{2} } (A)]\) , there exists \(\forall \theta _{i} \in {\Theta } :[Bel_{m_{1} } (\theta _{i}),Pl_{m_{1} } (\theta _{i})] \subseteq [Bel_{m_{2} } (\theta _{i}),Pl_{m_{2} } (\theta _{i})]\) . Then, we have \({d_{E}^{I}} \left ({[Bel_{m_{1} } (\theta _{i} ),Pl_{m_{1} } (\theta _{i} )],[0,1]} \right ) \ge {d_{E}^{I}} \left ({[Bel_{m_{2} } (\theta _{i} ),Pl_{m_{2} } (\theta _{i} )],[0,1]} \right )\), ∀𝜃 i ∈Θ. Therefore, according to (21), iTU I(m 1)≤iTU I(m 2).
Let us first prove property 4.1.
In a bayesian BPA, since m(A)=0, ∀|A|≠1, we have m(𝜃 i ) = Bel(𝜃 i ) = Pl(𝜃 i ), ∀𝜃 i ∈Θ. Assume m(𝜃 i ) = Bel(𝜃 i ) = Pl(𝜃 i ) = x i for 𝜃 i , so \(\sum \limits _{i = 1}^{n} {x_{i} } = 1\), and
Then
Therefore, we have the following Lagrangian function
where λ is the Lagrangian constant. By calculating the derivatives, we obtain
Then, by equaling these two functions to 0, we have
At the same time, since each maximum value satisfies
we have
Hence
namely,
When n≥2, we have λ≤0. Therefore,
which implies \(x_{i} \le \frac {1}{2}\), ∀i∈{1,⋯,n}. So, we obtain that
Namely, L(m) gets a maximum value at
Moreover, we calculate the second-order derivative of the Lagrangian function L(m)
which means that iTU I is strictly concave. Therefore, for the bayesian BPA m, iTU I(m) attains the global maximum value when m(𝜃 i )=1/n, ∀𝜃 i ∈Θ.
Then, let us first prove property 4.2.
Assume that there is a bayesian BPA m defined on FOD Θ with |Θ| = n. In terms of property 3.1, iTU I(m) attains the maximum value when m(𝜃 i )=1/n, ∀𝜃 i ∈Θ. Then we have
Here,
which means f is monotonically increasing with respect to n, \(n \in \mathbb {N}^ +\). Furthermore, \(\underset {{n \to + \infty }}{\lim } f = 1\). Hence
for bayesian BPAs m 1 defined on Θ1 with |Θ1| = n 1 and m 2 defined on Θ2 with |Θ2| = n 2, where n 1<n 2.
Rights and permissions
About this article
Cite this article
Deng, X., Xiao, F. & Deng, Y. An improved distance-based total uncertainty measure in belief function theory. Appl Intell 46, 898–915 (2017). https://doi.org/10.1007/s10489-016-0870-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-016-0870-3