Interactive algorithms for a broad underlying family of preference functions

Karakaya, Gülşah; Köksalan, Murat

doi:10.1016/j.ejor.2017.07.028

Cited by 13 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Ayrıca büyük α değerlerinden kaynaklanabilecek yuvarlama hatalarından ve uygulamada yaşanabilecek olası sorunlardan kaçınmak için Karakaya vd. 'nin [29] çalışmasında önerildiği gibi tahmin edilen α değerlerinin 4'ten büyük olması durumunda bu değerler sonsuz kabul edilmiştir. Ardışık iki iterasyonda mevcut en iyi çözümün ve tahmin edilen α değerinin aynı olduğu durumda tahmin edilen w vektörleri içindeki ağırlıkların mutlak farklarının en büyüğü 10 -3 'ten küçükse algoritma durmaktadır.…”

Section: Uygulamalar (Applications)unclassified

See 1 more Smart Citation

Karar vericinin tercih ettiği çözümü bulmak için filtreleme yardımlı etkileşimli bir yöntem

Şakar¹,

Karakaya²

2019

Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi

View full text Add to dashboard Cite

An interactive algorithm to find the best solution of a decision maker  Designed for multiple criteria problems in discrete solution space  Good convergence performance and speed  Application to university selection problem with five criteria Many decision making problems involve multiple criteria that need to be considered simultaneously, and these criteria generally conflict with each other. It is a difficult process for a decision maker (DM) to reach a final solution in such problems. In this study, a new interactive approach is proposed to converge towards the most preferred solution of the DM among discrete alternatives that are evaluated with multiple criteria. Figure A. Main steps of the proposed approach Purpose: For multiple criteria choice problems, this study proposes an approach that distinguishes a highly preferred solution of the DM with low cognitive burden. Theory and Methods: At each iteration of the proposed interactive approach, the DM evaluates a small number of solutions and selects the most preferred one. The preferences of the DM are modeled by a weighted Lα function and parameters of this function (weight values and α) are estimated iteratively. Weighted Lα function is flexible and capable of representing a wide variety of preference structures with the use of different parameters. The solutions presented to the DM are determined with a filtering procedure to enhance convergence performance. Results: The proposed approach is tested with different sets of university data from the rankings of Times Higher Education. The results show that in various problem settings, the approach successfully converges to the most preferred solution of the DM with low number of iterations. The approach is also shown to outperform benchmark algorithms. Conclusion: Various forms of DM preferences can be represented through a weighted Lα preference function, and the proposed approach is able to find a highly preferred solution of the DM for different forms. DM is required to perform a simple task in the approach and the number of iterations is low.

show abstract

Section: Uygulamalar (Applications)unclassified

“…Literatürde L α w tercih fonksiyonu Karakaya ve Köksalan'ın [28] ve Karakaya vd. 'nin [29] çalışmalarındaki gibi temsili tercih fonksiyonu olarak ya da Lokman ve Köksalan'ın [30] ve Lokman vd. 'nin [31] çalışmalarındaki gibi test fonksiyonu olarak kullanılmaktadır.…”

Section: Li̇teratür Taramasi (Literature Review)unclassified

Karar vericinin tercih ettiği çözümü bulmak için filtreleme yardımlı etkileşimli bir yöntem

Şakar¹,

Karakaya²

2019

Gazi Üniversitesi Mühendislik-Mimarlık Fakültesi Dergisi

View full text Add to dashboard Cite

show abstract

“…To avoid a poor estimation of the weights, Karakaya et al (2016b) developed a mixed integer mathematical model, Model (Mid ∞ ), to evaluate the DM's preferences. We utilize this model (given below) in iWREA-fs.…”

Section: Improvement Issues Of Itdea-fsmentioning

confidence: 99%

“…whereŵ j represents the estimated weight of the jth objective and y t (z m , z l ) is a binary decision variable (for a detailed explanation of the model, see Karakaya et al, 2016b). It is important to note that the constraint set of Model (Mid ∞ ) guarantees to contain the actual weights of the DM, in contrast with the reduced weight region obtained from the favorable weights of iTDEA-fs.…”

Section: Improvement Issues Of Itdea-fsmentioning

confidence: 99%

Interactive evolutionary approaches to multiobjective feature selection

Özmen

Karakaya

Köksalan

2017

Int Trans Operational Res

View full text Add to dashboard Cite

In feature selection problems, the aim is to select a subset of features to characterize an output of interest. In characterizing an output, we may want to consider multiple objectives such as maximizing classification performance, minimizing number of selected features or cost, etc. We develop a preference‐based approach for multiobjective feature selection problems. Finding all Pareto‐optimal subsets may turn out to be a computationally demanding problem and we still would need to select a solution. Therefore, we develop interactive evolutionary approaches that aim to converge to a subset that is highly preferred by the decision maker (DM). We test our approaches on several instances simulating DM preferences by underlying preference functions and demonstrate that they work well.

show abstract

“…In particular, opportunity to have access to all the details of calculation that enable researchers to have a systematic understanding of modeling mechanism is hardly to get unless modeling is in custom-built procedure. A few literature items on this topic can be found in the econometrics field [17][18][19][20][21][22][23], but these studies have addressed a different application environment and, therefore, it is difficult to apply these to modeling transportation related decisions 2 Modelling and Simulation in Engineering [24,25]. To tackle this kind of challenges that can occur in model developing, researchers must have a capability of doing modeling with custom-built computer codes that are tailored for each unique modeling case.…”

Section: Introductionmentioning

confidence: 99%

Performance Comparison of Mode Choice Optimization Algorithm with Simulated Discrete Choice Modeling

Roh

Sharma

Sahu

et al. 2018

Modelling and Simulation in Engineering

View full text Add to dashboard Cite

Until recently, a majority of modeling tasks of transportation planning, especially in discrete choice modeling, is conducted with the help of commercial software and only concerned about the result of parameter estimates to get a policy-sensitive interpretation. This common practice prevents researchers from gaining a systematic knowledge involved in estimation mechanism. In this research, to shed a light on these limited modeling practices, a standard discrete choice model’s parameter is estimated using Quasi-Newton method, DFP, and BFGS. Two extended algorithms, called DFP-GSM and BFGS-GSM, are proposed for the first time to overcome the weakness of the Quasi-Newton method. The golden section method (GSM) incorporates a nonlinear programming technique to choose an optimal step size automatically. Partial derivatives of log-likelihood function are derived and coded using Visual Basic Application (VBA). Through extensive numerical evaluation, estimation capability of each proposed estimation algorithms is compared in terms of performance measures. The proposed algorithms show a stable estimation performance and the reasons were studied and discussed. Furthermore, useful insights educated in custom-built modeling are present.

show abstract

Interactive algorithms for a broad underlying family of preference functions

Cited by 13 publications

References 16 publications

Karar vericinin tercih ettiği çözümü bulmak için filtreleme yardımlı etkileşimli bir yöntem

Karar vericinin tercih ettiği çözümü bulmak için filtreleme yardımlı etkileşimli bir yöntem

Interactive evolutionary approaches to multiobjective feature selection

Performance Comparison of Mode Choice Optimization Algorithm with Simulated Discrete Choice Modeling

Contact Info

Product

Resources

About