Multilabel feature selection involves the selection of relevant features from multilabeled datasets, resulting in improved multilabel learning accuracy. Evolutionary search‐based multilabel feature selection methods have proved useful for identifying a compact feature subset by successfully improving the accuracy of multilabel classification. However, conventional methods frequently violate budget constraints or result in inefficient searches due to ineffective exploration of important features. In this paper, we present an effective evolutionary search‐based feature selection method for multilabel classification with a budget constraint. The proposed method employs a novel exploration operation to enhance the search capabilities of a traditional genetic search, resulting in improved multilabel classification. Empirical studies using 20 real‐world datasets demonstrate that the proposed method outperforms conventional multilabel feature selection methods.

Multi-label feature selection that identifies important features from the original feature set of multi-labeled datasets has been attracting considerable attention owing to its generality compared to conventional single-label feature selection. The unimportant features are filtered by scoring the dependency of features to labels. In conventional multi-label feature filter studies, the score function is obtained by approximating a dependency measure such as joint entropy because direct calculation is often impractical due to the presence of multiple labels with limited training patterns. Although the efficacy of approximation can differ depending on the characteristics of the multi-label dataset, conventional methods presume a certain approximation method, leading to a degenerated feature subset if the presumed approximation is inappropriate for the given dataset. In this study, we propose a strategy for selecting an approximation among a series of approximations depending on the number of involved variables and consequently instantiate a score function based on the chosen approximation. The experimental results demonstrate that the proposed strategy and score function outperform conventional multi-label feature selection methods.

The goal of music genre classification is to identify the genre of given feature vectors representing certain characteristics of music clips. In addition, to improve the accuracy of music genre classification, considerable research has been conducted on extracting spectral features, which contain critical information for genre classification, from music clips and feeding these features into training models. In particular, recent studies argue that classification accuracy can be enhanced by employing multiple spectral features simultaneously. Consequently, fusing information from multiple spectral features is a critical consideration in designing music genre classification models. Hence, this paper provides a short survey of recent studies on music genre classification and compares the performance of the most recent CNN-based models with a newly devised model that employs a late fusion strategy for the multiple spectral features. Our empirical study of 12 public datasets, including Ballroom, ISMIR04, and GTZAN, showed that the late fusion CNN model outperforms other compared methods. Additionally, we performed an in-depth analysis to validate the effectiveness of the late fusion strategy in music genre classification.

Mobile devices play a crucial role in human activity recognition as they enable real-time sensing of user interaction for learning algorithms like neural networks. To facilitate human activity recognition on mobile devices, it is important to deploy efficient neural network architectures due to the limited computational capacity of these devices. However, conventional neural architecture search methods often generate less effective architectures because they neglect the specific requirements of target devices on which the neural network would operate in real-time. Moreover, these methods are impractical in the mobile device environment due to their high computational cost for architecture search. To address these challenges, we propose an efficient neural architecture search method based on a latency lookup table. The experimental results on public datasets provide evidence that the proposed method outperforms conventional methods in terms of speed. We achieved a search time of under 1.5 hours on each dataset, which is more than seven times faster on average compared to conventional methods. Furthermore, our in-depth analysis shows that the optimal architecture can vary depending on the target mobile devices, such as Galaxy A31 and S10. By tailoring the models to each device, optimized models achieved an additional 4-5% improvement in inference time for each respective device.

The goal of missing value imputation is to replace the missing values in the dataset with specific values. In particular, this preprocessing step plays a crucial role in knowledge discovery and data mining, as most data analysis methods assume complete data and cannot be applied directly to datasets with missing entries. Among various approaches, the neural network-based missing value imputation method has recently gained significant attention due to its superior prediction accuracy based on its excellent capability to fit the given training dataset. Specifically, these approaches conventionally begin by applying a naïve missing value imputation to fill all missing entries in the dataset and then train the network on the completed data. Thus, the performance of missing value imputation can be limited because the neural network is trained on an unreliable dataset filled with roughly guessed values. Instead, we may consider an alternative strategy to use only the features without missing values or carefully imputed features obtained during the imputation process, which can be regarded as an asymmetric process because it progressively adds the newly imputed features into the training dataset. In this study, we propose an effective neural network-based imputation method that incrementally constructs a cumulative feature set during training. The experimental results on 25 publicly available datasets showed that the proposed method outperforms conventional methods significantly.