Search

Customer data acquisition is an important task in data-driven business analytics. Recently, there has been a growing interest in the effective use of an organization's internal customer data, also known as first-party data. This work studies the acquisition of new data for business analytics based on first-party data resource. We address issues related to both acquisition cost and data quality. To reduce acquisition cost, we consider using auction-based methods, such as the generalized second price (GSP) auction, for acquiring data with differential prices for different customers. We find that the GSP-based data acquisition method incurs a lower cost and/or achieves a higher response rate than fixed price methods. To maximize data quality, we propose novel optimization models for different data acquisition methods and data quality measures. The proposed models maximize the quality of the acquired data while satisfying budget constraints. We derive and discuss the solutions to the optimization models analytically and provide managerial insights from the solutions. The proposed approach is effective in increasing customer responses, reducing selection bias, and enabling more accurate estimation and prediction for business analytics. The results of the experimental evaluation demonstrate the advantage of the proposed approach over existing data acquisition methods. History: Accepted by Ram Ramesh, Area Editor for Data Science and Machine Learning. Supplemental Material: The software that supports the findings of this study is available within the paper and its Supplemental Information (https://pubsonline.informs.org/doi/suppl/10.1287/ijoc.2022.0037) as well as from the IJOC GitHub software repository (https://github.com/INFORMSJoC/2022.0037). The complete IJOC Software and Data Repository is available at https://informsjoc.github.io/. [ABSTRACT FROM AUTHOR]

The present study investigates the flow around a uniform porous-coated cylinder under subcritical flow state (Re = 5 × 10 4 ) through numerical analysis to explore the influence of porous media parameters on the wake flow and radiated noise of cylinders. A numerical approach that integrates large eddy simulation and the Ffowcs Williams–Hawkings acoustic analog method is employed and validated through comparison with existing experimental and numerical data. The research shows that the application of porous media not only significantly alters the flow pattern behind the cylinder, stabilizes the boundary layer separation from the cylinder surface, widens the wake width, and reduces the pressure fluctuations on the cylinder surface, but effectively suppresses vortex shedding in the wake region and enhances the base pressure on the leeward side, thereby reducing the drag coefficient and shedding frequency of the cylinder, which achieves the effects of drag reduction and noise mitigation. The present study found that when the covering angle on the leeward side of the cylinder was 300° and the coating thickness was 16 mm, the average drag coefficient decreased by 36.27% and the aerodynamic noise by 74.91%. Furthermore, when the permeability was 1.5 × 10 − 5 m 2 , the average drag coefficient decreased by 45.78% and the aerodynamic noise by 73.79%. Additionally, the comparative analyses of the overall sound pressure level show that though applying porous media does not alter the radiation mode of noise, it can reduce the sound source intensity. This study is expected to provide useful information for the flow and noise control of high-speed train pantograph structures. [ABSTRACT FROM AUTHOR]

When acquiring consumer data for marketing or new business initiatives, it is important to decide what attributes or features of potential customers should be acquired. We study a new feature selection problem in the context of customer data acquisition in which different features have different acquisition costs. This feature selection problem is studied for linear regression and logistic regression. We formulate the feature selection and acquisition problems as nonlinear discrete optimization problems that minimize prediction errors subject to a budget constraint. We derive the analytical properties of the solutions for the problems, develop a computational procedure for solving the problems, provide an intuitive interpretation for the feature selection criteria, and discuss managerial implications of the solution approach. The results of the experimental study demonstrate the effectiveness of our approach. This paper was accepted by Kartik Hosanagar, information systems. Supplemental Material: Data are available at https://doi.org/10.1287/mnsc.2022.4551. [ABSTRACT FROM AUTHOR]

Integration of radar and communication systems based on OFDM signals results in large Peak-to-Average Power Ratio (PAPR). Limited by the code rate, algorithm that use the Golay sequence code to restrain PAPR can only be applied under the condition of a few subcarriers. This study proposes an algorithm to restrain the PAPR of systems with a large number of subcarriers. The algorithm combines the group parallel code with the optimization of weight coefficients. First, bit streams are divided into several groups of parallel bits. Next, every group proceeds with Golay sequence coding, data symbol modulating and inverse Fourier transform. Finally, the parallel result is combined with an OFDM symbol. Before the parallel data are combined, several weight coefficients for every group are introduced; thus, the system has several candidate symbols for transmitting. Then the symbol with minimum PAPR is then selected as the transmitting signal, and the PAPR of the whole system is reduced. PAPR performance, Bit Error Radio (BER) and wideband ambiguity function of three block methods with different coding rate are also simulated. The simulations show that the PAPR of the system decreases and the BER performance improves significantly. The signal exhibits a thumbtack ambiguity function, which suggests good resolution and accuracy for distance and velocity measurements.

Due to growing concerns about the privacy of personal information, organizations that use their customers' records in data mining activities are forced to take actions to protect the privacy of the individuals. A frequently used disclosure protection method is data perturbation. When used for data mining, it is desirable that perturbation preserves statistical relationships between attributes, while providing adequate protection for individual confidential data. To achieve this goal, we propose a kd-tree based perturbation method, which recursively partitions a data set into smaller subsets such that data records within each subset are more homogeneous after each partition. The confidential data in each final subset are then perturbed using the subset average. An experimental study is conducted to show the effectiveness of the proposed method. Index Terms--Privacy, data mining, data perturbation, microaggregation, kd-trees.

Multiple-model approach provides the state-of-the-art solutions to many problems involving estimation, filtering, control, and/or modeling. One of the most important problems in the application of the multiple-model approach is the design of the model set used in a multiple-model algorithm. To our knowledge, however, it has never been addressed systematically in the literature. This paper deals with this challenging topic in a general setting. General problems of model-set design are considered. A concept of a random model is introduced. In other words, modeling of models used in a multiple model (MM) algorithm as well as the true model as random variables is proposed. Three classes of general methods for optimal design of model sets--by minimizing distribution mismatch, minimizing modal distance, and moment matching, respectively--are proposed. Theoretical results that address many of the associated issues are presented. Examples that demonstrate how some of these theoretical results can be used as well as their effectiveness are given. Many of the general results presented in this paper are also useful for performance evaluation of MM algorithms. Index Terms--Adaptive control, adaptive estimation, modeling, model-set design, multiple models, target tracking, variable structure.

A new decision tree method for application in data mining, machine learning, pattern recognition, and other areas is proposed in this paper. The new method incorporates a classical multivariate statistical method, linear discriminant function, into decision trees' recursive partitioning process. The proposed method considers not only the linear combination with all variables, but also combinations with fewer variables. It uses a tabu search technique to find appropriate variable combinations within a reasonable length of time. For problems with more than two classes, the tabu search technique is also used to group the data into two superclasses before each split. The results of our experimental study indicate that the proposed algorithm appears to outperform some of the major classification algorithms in terms of classification accuracy, the proposed algorithm generates decision trees with relatively small sizes, and the proposed algorithm runs faster than most multivariate decision trees and its computing time increases linearly with data size, indicating that the algorithm is scalable to large datasets. Index Terms--Classification, data mining, decision trees, linear discriminant, tabu search.

This paper concerns a decision-tree pruning method, a key issue in the development of decision trees. We propose a new method that applies the classical optimization technique, dynamic programming, to […]