Deep learning algorithms have exhibited impressive performance across various computer vision tasks; however, the challenge of overfitting persists, especially when dealing with limited labeled data. This survey explores the mitigation of the overfitting issue through a comprehensive examination of image data augmentation techniques, which aim to enhance dataset size and diversity by introducing varied samples. The survey exclusively focuses on these techniques, presenting an insightful overview and introducing a novel taxonomy. The discussion encompasses the strengths and limitations of these techniques. Additionally, the paper provides extensive results evaluating the impact of these techniques on prevalent computer vision tasks: image classification, object detection, and semantic segmentation. The survey concludes with an examination of challenges, limitations, and potential future research directions.

The vast amount of knowledge about past human societies has not been systematically organized and, therefore, remains inaccessible for empirically testing theories about cultural evolution and historical dynamics. For example, what evolutionary mechanisms were involved in the transition from the small-scale, uncentralized societies, in which humans lived 10,000 years ago, to the large-scale societies with an extensive division of labor, great differentials in wealth and power, and elaborate governance structures of today? Why do modern states sometimes fail to meet the basic needs of their populations? Why do economies decline, or fail to grow? In this article, we describe the structure and uses of a massive databank of historical and archaeological information, Seshat: The Global History Databank . The data that we are currently entering in Seshat will allow us and others to test theories explaining how modern societies evolved from ancestral ones, and why modern societies vary so much in their capacity to satisfy their members’ basic human needs.

Ontologies are at the heart of knowledge management and make use of information that is not only written in English but also in many other natural languages. In order to enable knowledge discovery, sharing and reuse of these multilingual ontologies, it is necessary to support ontology mapping despite natural language barriers. This paper examines the soundness of a generic approach that involves machine translation tools and monolingual ontology matching techniques in cross-lingual ontology mapping scenarios. In particular, experimental results collected from case studies which engage mappings of independent ontologies that are labeled in English and Chinese are presented. Based on findings derived from these studies, limitations of this generic approach are discussed. It is shown with evidence that appropriate translations of conceptual labels in ontologies are of crucial importance when applying monolingual matching techniques in cross-lingual ontology mapping. Finally, to address the identified challenges, a semantic-oriented cross-lingual ontology mapping (SOCOM) framework is proposed and discussed.

Three key challenges to a whole-system approach to process improvement in health systems are the complexity of socio-technical activity, the capacity to change purposefully, and the consequent capacity to proactively manage and govern the system. The literature on healthcare improvement demonstrates the persistence of these problems. In this project, the Access-Risk-Knowledge (ARK) Platform, which supports the implementation of improvement projects, was deployed across three healthcare organisations to address risk management for the prevention and control of healthcare-associated infections (HCAIs). In each organisation, quality and safety experts initiated an ARK project and participated in a follow-up survey and focus group. The platform was then evaluated against a set of fifteen needs related to complex system transformation. While the results highlighted concerns about the platform’s usability, feedback was generally positive regarding its effectiveness and potential value in supporting HCAI risk management. The ARK Platform addresses the majority of identified needs for system transformation; other needs were validated in the trial or are undergoing development. This trial provided a starting point for a knowledge-based solution to enhance organisational governance and develop shared knowledge through a Community of Practice that will contribute to sustaining and generalising that change.

Advanced data augmentation techniques have demonstrated great success in deep learning algorithms. Among these techniques, single-image-based data augmentation (SIBDA), in which a single image’s regions are randomly erased in different ways, has shown promising results. However, randomly erasing image regions in SIBDA can cause a loss of the key discriminating features, consequently misleading neural networks and lowering their performance. To alleviate this issue, in this paper, we propose the random slices mixing data augmentation (RSMDA) technique, in which slices of one image are placed onto another image to create a third image that enriches the diversity of the data. RSMDA also mixes the labels of the original images to create an augmented label for the new image to exploit label smoothing. Furthermore, we propose and investigate three strategies for RSMDA: (i) the vertical slices mixing strategy, (ii) the horizontal slices mixing strategy, and (iii) a random mix of both strategies. Of these strategies, the horizontal slice mixing strategy shows the best performance. To validate the proposed technique, we perform several experiments using different neural networks across four datasets: fashion-MNIST, CIFAR10, CIFAR100, and STL10. The experimental results of the image classification with RSMDA showed better accuracy and robustness than the state-of-the-art (SOTA) single-image-based and multi-image-based methods. Finally, class activation maps are employed to visualize the focus of the neural network and compare maps using the SOTA data augmentation methods.

Data is central to modern decision making and value creation. Society creates, consumes and collects data at an increasing pace. Despite advances in processing power, data is expensive to maintain and curate. So, it is imperative to have methods and tools to distinguish between data based on its value. Yet, there is no consensus on what characterises the value of data or how this data value should be assessed. This results in heterogeneous data value models and inconsistent measurement techniques that are siloed in specific application domains. This limits the formalisation and exploitation of these concepts. We present in this paper a methodical literature analysis that discusses data value models, assessment metrics and current applications. We also highlight challenges hindering the development and exploitation of data value as concept. This leads to the identification of a set of research questions to help researchers contribute to this emerging field. The aim of this article is to stimulate further research and deployment of quantitative data value models and value-driven applications.