Abstract Kernel methods have long played a central role in machine learning by enabling nonlinear modeling in high dimensional feature spaces. Among them, the Gaussian Radial Basis Function (RBF) kernel is widely used due to its smoothness, universality, and excellent performance in supervised learning. In contrast, the Jaccard/Tanimoto coefficient, originally defined over sets and binary vectors, has achieved notable success in chemoinformatics and information retrieval. Recent studies have shown that the Jaccard/Tanimoto coefficient can be extended to real-valued vectors and interpreted as a valid kernel. In this paper, we propose a natural generalization of the Tanimoto kernel by introducing a bandwidth-like parameter, leading to the Bandwidth-Adjustable Tanimoto (BAT) kernel. We focus on the shared structure of the BAT and the Gaussian RBF kernels in terms of their infinite polynomial expansions and investigate the universality of the BAT kernels, clarifying their relationship to the universal Taylor kernels. In particular, we show that by incorporating a bias term, the BAT kernel attains the universal approximation property comparable to that of the Gaussian kernel. Through experiments on classification, regression, and clustering tasks using benchmark datasets, we demonstrate that the BAT kernel performs comparably to the Gaussian kernel, while even the original Tanimoto kernel achieves competitive results. In particular, the BAT kernel attains performance on par with the Gaussian kernel in classification tasks, and it slightly surpasses the Gaussian kernel in regression and clustering tasks. A comparison of their functional forms, together with experiments on both regression and classification tasks involving nonlinear structures, further shows that the BAT kernel yields smoother predictive surfaces than the RBF kernel. Overall, our results indicate that the BAT kernel offers universality and practical performance, making it an alternative to the Gaussian kernel.

Traditional survey methods are at a crossroads in Japan owing to decreasing response rates and survey participation. Therefore, this study developed a novel survey method utilizing a Web-based voice recognition survey app as a new “interviewer” that replaces humans to conduct large-scale surveys for probability samples following the same procedure as conventional interview surveys. This study entails the development of the survey process and evaluating the practical feasibility of the new survey methodology. Overall, the response rate of the new survey method was not as high as that of the usual interview surveys. Furthermore, this study shows that the response rate was low compared with the login rate. Consequently, more attractive survey participation features should be developed to retain participants’ responses from the time they log in until they complete the survey.

In a regression model, prediction is typically performed after model selection. The large variability in the model selection makes the prediction unstable. Thus, it is essential to reduce the variability in model selection and improve prediction accuracy. To achieve this goal, a parametric bootstrap smoothing can be applied. In this method, model selection is performed for each resampling from a parametric distribution, and these models are then averaged such that the distribution of the selected models is considered. Here, the prediction accuracy is highly dependent on the choice of a distribution for resampling. In particular, an experimental study shows that the choice of error variance significantly changes the distribution of the selected model and thus plays a key role in improving the prediction accuracy. We also observe that the true error variance does not always provide optimal prediction accuracy. Therefore, it is not always appropriate to use unbiased estimators of the true parameters or standard estimators of the parameters for the resampling distribution. In this study, we propose employing cross-validation to choose a suitable resampling distribution rather than unbiased estimators of parameters. Our proposed method was applied to several data sets. The results indicate that the proposed method provides a better prediction accuracy than the existing method.

This study proposes a method of measuring social value orientation using model-based scoring and a task suitable for such scoring. We evaluated this method by means of parameter recovery simulation (Study 1), and we examined its retest reliability (Study 2) and its predictive validity (Study 3). The results indicate that the proposed method has low bias and sufficient predictive validity. While the improvement in predictive validity of altruism was negligible and comparable to previous scoring methods in terms of confidence intervals, the measurement of equality using the proposed model and task combination produced a moderate correlation that was not observed with other methods. Although SVO is a concept used primarily in psychology, the model assumed in this study is mathematically equivalent to a well-known economics model. We, therefore, suggest that this method may lead to cross-disciplinary research.

Abstract Random forests are currently one of the most popular algorithms for supervised machine learning tasks. By taking into account for many trees instead of a single one the resulting forest model is no longer easy to understand and also often denoted as a black box. The paper is dedicated to the interpretability of random forest models using tree-based explanations. Two different concepts, namely most representative trees and surrogate trees are analyzed regarding both their ability to explain the model and to be understandable by humans. For this purpose explanation trees are further extended to groves, i.e. small forests of few trees. The results of an application to three real world data sets underline the inherent trade of between both requirements. Using groves allows to control for the complexity of an explanation while simultaneously analyzing their explanatory power.