Classification using Synergistic Teams

Hyun, Baro; Faied, Mariam; Kabamba, Pierre T.; Girard, Anouck

doi:10.3182/20110828-6-it-1002.03152

Cited by 1 publication

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper extends the work on fusion-rule-based team classification presented in [12]. That paper considered only the cases when two teams were identical, i.e., where each classifier had the same false positive and false negative rates.…”

Section: Relationship To Past Workmentioning

confidence: 87%

“…Here, we formulate the moderated decisions based on the decisions of individual classifier in the team. We extend the work in [12] and provide the results for all the possible combinations of α and β . For more details, see [12].…”

Section: Problem Formulationmentioning

confidence: 95%

“…We extend the work in [12] and provide the results for all the possible combinations of α and β . For more details, see [12]. Table II gives a summary of the optimal fusion rules for various cases.…”

Section: Problem Formulationmentioning

confidence: 95%

“…VI. Note that The performance of three-classifier teams without a moderator is similar to the two-classifier case analyzed in [12]. Thus we omit the mathematical formulation in this paper, and show the simulation results directly in Sec.…”

Section: E Organizationmentioning

confidence: 98%

“…Table I shows the 16 possible fusion rules for a twoclassifier team. The basic logical operators of the 16 possible fusion rules, such as conjunction, disjunction and etc., are proposed to demonstrate the approach in the Section 4.1 in [12].…”

Section: B Fusion Rulesmentioning

confidence: 99%

See 4 more Smart Citations

Optimal fusion rules in team classification under three decision structures

Pan

Hyun

Kabamba

et al. 2013

2013 American Control Conference

Self Cite

View full text Add to dashboard Cite

In this paper, we study the performance of a team of dichotomous classifiers, where the classifiers' decisions are combined by logical fusion rules. Three decision structures are derived using the confusion matrix of a single classifier and a priori information, and the performances of the different decision structures are compared. First, we consider the performance of a team of three classifiers with a total of 256 fusion rules. Then, we propose a decision structure that utilizes a moderator, i.e., an entity that exploits Bayesian inference from individual classifiers' decisions and makes final decisions based on maximum likelihood classification. We show the benefits of using a moderator (compared to a decision structure without a moderator). Finally, we propose a decision structure that exploits pairing, i.e., fusing the classifiers' decisions sequentially two-by-two. Two pairing schemes, i.e., incremental and tournament-like, are proposed and we show that incremental pairing is the most effective decision structure among the proposed ones.

show abstract

Section: Relationship To Past Workmentioning

confidence: 87%

Section: Problem Formulationmentioning

confidence: 95%