Label Information Research Articles

The taxonomic composition and chronology of a museum collection of Coleoptera revealed through large-scale digitisation

IntroductionHistoric museum collections hold a wealth of biodiversity data that are essential to our understanding of the rapidly changing natural world. Novel curatorial practices are needed to extract and digitise these data, especially for the innumerable pinned insects whose collecting information is held on small labels.MethodsWe piloted semi-automated specimen imaging and digitisation of specimen labels for a collection of ~29,000 pinned insects of ground beetles (Carabidae: Lebiinae) held at the Natural History Museum, London. Raw transcription data were curated against literature sources and non-digital collection records. The primary data were subjected to statistical analyses to infer trends in collection activities and descriptive taxonomy over the past two centuries.ResultsThis work produced research-ready digitised records for 2,546 species (40% of known species of Lebiinae). Label information was available on geography in 91% of identified specimens, and the time of collection in 39.8% of specimens and could be approximated for nearly all specimens. Label data revealed the great age of this collection (average age 91.4 years) and the peak period of specimen acquisition between 1880 and 1930, with little differences among continents. Specimen acquisition declined greatly after about 1950. Early detected species generally were present in numerous specimens but were missing records from recent decades, while more recently acquired species (after 1950) were represented mostly by singleton specimens only. The slowing collection growth was mirrored by the decreasing rate of species description, which was affected by huge time lags of several decades to formal description after the initial specimen acquisition.DiscussionHistoric label information provides a unique resource for assessing the state of biodiversity backwards to pre-industrial times. Many species held in historical collections especially from tropical super-diverse areas may not be discovered ever again, and if they do, their recognition requires access to digital resources and more complete levels of species description. A final challenge is to link the historical specimens to contemporary collections that are mostly conducted with mechanical trapping of specimens and DNA-based species recognition.

Securing Anomaly Detection for Process-Based Time Series

Widespread innovation from artificial intelligence and machine learning (AI/ML) tools presents a lucrative opportunity for the nuclear industry to improve state-of-the-art analyses (e.g. condition monitoring, remote operation, etc.) due to increased data visibility. In recent years, the risk posed by collaborative data exchange has received increased attention due, in part, to a potential adversary’s ability to reverse-engineer intercepted data using domain knowledge and AI/ML tools. While the efficacy of typical encryption has been proven during passive communication and data storage, collaborative exchange typically requires decryption for extended analyses by a third-party, which poses an intrinsic risk due to these trustworthiness concerns. The directed infusion of data (DIOD) 1 1 A.Al Rashdan and H.Abdel-Khalik, Deceptive Infusion of Data, Non-Provisional Patent, Application No. 63/227,389, September 2022. paradigm presented in this paper discusses a novel data masking technique that relies on preserving the usable information of proprietary data while concealing its identity via reduced-order modeling. In contrast to existing state-of-the-art data masking methods, DIOD does not impose limiting assumptions, computational overheads, or induced uncertainties, thereby allowing for secure and flexible data-level security that does not alter the inferential content of the data. This paper focuses on the application of DIOD to a process-based simulation wherein a leaking reservoir with a controlled inlet pump is simulated under various experimental conditions with the goal of producing masked data that preserve the information given by anomalies. These experiments included the injection of statistically significant anomalies, subtle anomalies that occurred over an extended period, and the addition of several independent anomalous states. Each experiment showed that a classifier will identify the same anomalies whether it analyzes the original or masked data. An additional experiment also tested the case of corrupted labeling information wherein labels were arbitrarily randomized, and the loss in labeling accuracy was about the same for both datasets. Each of these experiments show that data obfuscated by DIOD may be utilized in the place of real data for a variety of condition monitoring scenarios with no loss in performance.

Innovation of Processed Baronang Fish as One of The Local Food Products in Gebang Village, Pesawaran

Gebang Pesawaran Village is an area that occupies a coastal area and has quite abundant marine products, but the potential for utilization is less than optimal and there is no food made from any of the existing marine products. This research aims to explore the potential of rabbitfish itself and to create local food products that are losng-lasting but still nutritious in Gebang Pesawaran Village. It is also an effort to create a variety of food dishes that do not only depend on one type of ingredient, but also utilize a variety of food ingredients. The product made in this research was serundeng with coconut and rabbitfish as the basic ingredients, which was named Seronang. The method used in this research is qualitative with a descriptive approach. The results of the research show that the baronang fish serundeng product is a new food innovation in Gebang Village. This product has passed food chemical test analysis and has a nutritional value information label to make it easier for consumers to consume the product. From the results of chemical food tests, it is known that the nutritional content contained in baronang fish serundeng products includes air content of 5.4481%, glucose 6.1474%, fat 32.8965%, carbohydrates 31.8109%, protein 23.6970%, and vitamins 8.8597%.

Few-shot relational triple extraction with hierarchical prototype optimization

Relational Triple Extraction (RTE) aims to extract relations and entities from unstructured text. Current RTE models using supervised learning require a large amount of labeled data, which presents a challenge for real-world applications. Therefore, the research work on Few-Shot Relational Triple Extraction (FS-RTE) has been proposed. However, the existing work cannot effectively construct accurate prototypes from a small number of samples, and it is difficult to model the dependencies between entities and relations, resulting in poor performance in relational triple extraction. In this paper, we propose a Hierarchical Prototype Optimized FS-RTE method (HPO). In particular, to mitigate prototype bias built on a small number of samples, HPO uses prompt learning to merge the information of relational labels into the text. Then, the entity-level prototypes are constructed using a span encoder to avoid label dependency between entity tokens. Finally, the hierarchical contrastive learning (HCL) method is introduced to improve the metric space between the prototypes of entities and relations, respectively. Experiments conducted on two public datasets show that HPO can significantly outperform previous state-of-the-art methods.

Bug numbers matter: An empirical study of effort‐aware defect prediction using class labels versus bug numbers

AbstractPrevious research have utilized public software defect datasets such as NASA, RELINK, and SOFTLAB, which only contain class label information. Most effort‐aware defect prediction (EADP) studies are carried out around these datasets. However, EADP studies typically relying on predicted bug number (i.e., considering modules as effort) or density (i.e., considering lines of code as effort) for ranking software modules. To explore the impact of bug number information in constructing EADP models, we access the performance degradation of the best‐performing learning‐to‐rank methods when using class labels instead of bug numbers for training. The experimental results show that using class labels instead of bug numbers in building EADP models results in an decrease in the detected bugs when module is considering as effort. When effort is LOC, using class labels to construct EADP models can lead to a significant increase in the initial false alarms and a significant increase in the modules that need to be inspected. Therefore, we recommend not only the class labels but also the bug number information should be disclosed when publishing software defect datasets, in order to construct more accurate EADP models.

A new Ranunculus (Ranunculaceae) from Nevada, U.S.A.

A new species of Ranunculus from Elko County, Nevada, R. legerae, is described and illustrated. It is related to R. triternatus, R. glaberimmus, and R. adoneus. The earliest known collection was made in 1937. Poor label information prevented rediscovery, which did not happen until 2017. The second author discovered a large population in 2022 which led to this study and article.

Partial label feature selection via label disambiguation and neighborhood mutual information

Partial label learning aims to learn from training instances, each of which is associated with a set of candidate labels but only one is a ground-truth label. Feature selection is an effective method to improve the generalization capability of the learning model; however, partial label feature selection work is exceptionally challenging due to the limitation and ambiguity of label information. Therefore, this paper proposes a partial label feature selection algorithm based on label disambiguation and neighborhood mutual information. Firstly, neighborhood granularity is utilized to determine the neighborhoods of instances to disambiguate the candidate labels. Secondly, based on label confidence induced by disambiguation, feature relevance and redundancy are measured by neighborhood mutual information, which avoids the negative impact of data discretization on feature selection and directly handles continuous features. Concurrently, the kappa coefficient is employed to estimate the label consistency for describing the influences of feature changes on the label space. Then, the significance of each feature is evaluated by fusing feature relevance, feature redundancy, and label consistency. Finally, the effectiveness of the proposed algorithm is verified by comparing the proposed algorithm with four base classifiers and other feature selection methods. Furthermore, the feasibility of the proposed disambiguation method is demonstrated through comparison with four state-of-the-art disambiguation methods.

CentroIDA: Cross-domain class discrepancy minimization based on accumulative class-centroids for Imbalanced Domain Adaptation

Unsupervised Domain Adaptation (UDA) is a critical task in transfer learning, which seeks to apply the knowledge acquired in one domain (the source domain) to a different but related domain (the target domain). In this context, the two domains exhibit Out-Of-Distribution(OOD), and the label information of the target domain is unavailable. Although numerous UDA methods have made significant strides in mitigating the discrepancies in feature distributions across domains, they often neglect a more prevalent real-world scenario. This scenario, referred to as Imbalanced Domain Adaptation (IDA) task, is characterized by the coexistence of differences in both feature and label distributions. Our empirical analysis suggests that overlooking the disparities in label distributions can severely compromise the performance of the model, thereby limiting the generalizability of domain adaptation models. Consequently, our objective is to introduce a model capable of concurrently addressing differences in both feature and label distributions, thereby fostering a broader application of domain adaptation methodologies. In this paper, we propose a cross-domain class discrepancy minimization method based on accumulative class-centroids for IDA, termed as centroIDA. The strategies of accumulative class-centroids alignment and class-wise feature alignment pay heed to the relationship between cross-domain intra-class features and inter-class features, thereby facilitating conditional feature alignment to overcome both feature and label distribution differences simultaneously. Our experimental evaluation on OfficeHome and DomainNet datasets reveals that, in comparison to TIToK (a novel method also designed to tackle the IDA problem), our method enhances the average accuracy by 5.6% and 2.7%, respectively. Furthermore, as the label shift intensifies, our method consistently outperforms other methods, demonstrating a robust resistance to label shift.

An Adaptive Noisy Label-Correction Method Based on Selective Loss for Hyperspectral Image-Classification Problem

Due to the intricate terrain and restricted resources, hyperspectral image (HSI) datasets captured in real-world scenarios typically contain noisy labels, which may seriously affect the classification results. To address this issue, we work on a universal method that rectifies the labels first and then trains the classifier with corrected labels. In this study, we relax the common assumption that all training data are potentially corrupted and instead posit the presence of a small set of reliable data points within the training set. Under this framework, we propose a novel label-correction method named adaptive selective loss propagation algorithm (ASLPA). Firstly, the spectral–spatial information is extracted from the hyperspectral image and used to construct the inter-pixel transition probability matrix. Secondly, we construct the trusted set with the known clean data and estimate the proportion of accurate labels within the untrusted set. Then, we enlarge the trusted set according to the estimated proportion and identify an adaptive number of samples with lower loss values from the untrusted set to supplement the trusted set. Finally, we conduct label propagation based on the enlarged trusted set. This approach takes full advantage of label information from the trusted and untrusted sets, and moreover the exploitation on the untrusted set can adjust adaptively according to the estimated noise level. Experimental results on three widely used HSI datasets show that our proposed ASLPA method performs better than the state-of-the-art label-cleaning methods.

Study on the labels of selected cat and dog foods from Türkiye

At first look, the label contains details about the contents and properties of the cat and dog foods. Label information must be included in accordance with national and international requirements. The sufficiency of cat and dog food labelling in Türkiye has not before been evaluated. In this study, the labelling information for 215 chosen cat and dog foods (164 cats and 51 dogs) available in Türkiye was analysed using Turkish national legislation and the Federation of the European Pet Food Industry standards. It was determined that all food labelling information was compliant with Turkish feed legislation. In comparison, only 67.1% of cat and 78% of dog diets included the metabolised energy value on the package. Cat foods have a website at 60.8%, whereas dog foods have 81.7%. It is found that the cat and dog food labels totally comply with Turkish legislation rules. On the other hand, more information may be valuable in enlightening customers. The recommended amount of food per day was unclear in numerous products, which is an area of improvement. In future studies, competencies of the daily nutrition advice information on the labels can be evaluated.

Multi-stage ensemble with refinement for noisy labeled data classification

Deep neural networks (DNNs) have made remarkable progress in image classification. However, since DNNs can memorize all the label information in the training dataset due to their excellent feature learning ability, the generalization performance deteriorates when they are trained on the noisy labeled dataset that can be easily found in real-world problems. In this paper, we propose a multi-stage ensemble method with label refinement to build an effective classification model under noisy labels. The proposed method iteratively refines the dataset by re-labeling the samples at the end of each stage, which enables the models trained at each stage to learn different features. By integrating these models, the proposed multi-stage ensemble method exerts powerful generalization performance. Also, we suggest a novel dataset refinement method, demonstrating the effectiveness of a robust function in distinguishing corrupted samples. Experimental results on the benchmark and real-world datasets show that the proposed method outperforms the existing methods on the noisy labeled dataset classification.

Hypergraph-based convex semi-supervised unconstraint symmetric matrix factorization for image clustering

Semi-supervised symmetric nonnegative matrix factorization (SNMF) has been extensively utilized in both linear and nonlinear data clustering tasks. However, the current SNMF model's non-convex objective function faces challenges in global optimization and time efficiency. In this study, we leverage label information to propose a convex and unconstrained symmetric matrix factorization (SMF) model that is thoroughly analyzed for its convexity properties. In order to capture high-order relationships among data, a hypergraph is utilized in the model, which is computationally simple, translation invariant, and naturally normalized. Moreover, based on the analysis and the corresponding experiments in the paper, the model exhibits robustness towards outliers to some extent. Due to the convexity of our proposed model without constraint, it can be efficiently optimized using the Conjugate Gradient (CG) method, one of the most efficient methods available. Therefore, we propose a novel Convex Combination-based Sufficient Descent CG (CSDCG) method, which outperforms other methods across 284 optimization problems within the CUTEst library. In order to evaluate the effectiveness of the proposed method, the semi-supervised clustering experiments are conducted on the eight datasets by comparison with ten state-of-the-art matrix factorization (MF) methods. The experiment results demonstrate its superiority over the other compared methods to handle the clustering problem with better performance and less computational time. The code is available at https://github.com/Pokemer/HCSSMF.

Quality Function Deployment: Identification of Key Parameters in Grape Growing and Wine Making

Background: The study conducted benchmarking research, assessed and identified suitable factors for the efficient benchmarking of Karnataka wines. As the Grape Processing Industry Board is being formed by the Wine Board of Karnataka (Wine Policy ‘2007’), which may have control of grape quality, standards formulation, label approval and market promotion, the study focused on recommending strategies for quality grape-growing, wine-making for label approvals. Methods: The study adopted the Quality Function Deployment (QFD) tool to build the House of Quality to benchmark Karnataka wines with the New World wines. Result: The findings demonstrate that the informative label was given high importance by customers during wine purchases. The wine industry to meet this requirement should focus more, on the factor-crushing method of grapes. Karnataka wineries are inferior to wine from other New World regions when it comes to other aspects such as flavour, colour, brand name, grape variety, meal pairings, alcohol content, wine label designand label information.

Digestible indispensable amino acid score (DIAAS): 10 years on.

The objective of the review is to revisit the findings of the 2011 Food and Agriculture Organization of the United Nations (FAO) Expert Consultation on Dietary Protein Quality Evaluation in Human Nutrition, and to report on progress on uptake of the findings. It is evident that since 2011 there has been a concerted research effort to enhance an understanding of the protein quality of foods. The validity of the growing pig ileal protein digestibility assay has been confirmed and numerous studies reported using the growing pig as a model to give true ileal amino acid digestibility values for foods as consumed by humans. This has allowed for the determination of digestible indispensable amino acid scores (DIAAS) for a range of foods. A new non-invasive true ileal amino acid digestibility assay in humans which can be applied in different physiological states, called the dual-isotope assay, has been developed and applied to determine the DIAAS values of foods. It is concluded that DIAAS is currently the most accurate score for routinely assessing the protein quality rating of single source proteins. In the future, the accuracy of DIAAS can be enhanced by improved information on: the ideal dietary amino acid balance including the ideal dispensable to indispensable amino acid ratio; dietary indispensable amino acid requirements; effects of processing on ileal amino acid digestibility and lysine bioavailability. There is a need to develop rapid, inexpensive in vitro digestibility assays. Conceptual issues relating DIAAS to food regulatory claims, and to holistic indices of food nutritional and health status are discussed. The first recommendation of the 2011 Consultation regarding treating each indispensable amino acid as an individual nutrient has received little attention. Consideration should be given to providing food label information on the digestible contents of specific indispensable amino acids.

Semantic Visual SLAM Algorithm Based on Improved DeepLabV3+ Model and LK Optical Flow

Aiming at the problem that dynamic targets in indoor environments lead to low accuracy and large errors in the localization and position estimation of visual SLAM systems and the inability to build maps containing semantic information, a semantic visual SLAM algorithm based on the semantic segmentation network DeepLabV3+ and LK optical flow is proposed based on the ORB-SLAM2 system. First, the dynamic target feature points are detected and rejected based on the lightweight semantic segmentation network DeepLabV3+ and LK optical flow method. Second, the static environment occluded by the dynamic target is repaired using the time-weighted multi-frame fusion background repair technique. Lastly, the filtered static feature points are used for feature matching and position calculation. Meanwhile, the semantic labeling information of static objects obtained based on the lightweight semantic segmentation network DeepLabV3+ is fused with the static environment information after background repair to generate dense point cloud maps containing semantic information, and the semantic dense point cloud maps are transformed into semantic octree maps using the octree spatial segmentation data structure. The localization accuracy of the visual SLAM system and the construction of the semantic maps are verified using the widely used TUM RGB-D dataset and real scene data, respectively. The experimental results show that the proposed semantic visual SLAM algorithm can effectively reduce the influence of dynamic targets on the system, and compared with other advanced algorithms, such as DynaSLAM, it has the highest performance in indoor dynamic environments in terms of localization accuracy and time consumption. In addition, semantic maps can be constructed so that the robot can better understand and adapt to the indoor dynamic environment.

A cross-scenario and cross-subject domain adaptation method for driving fatigue detection

Objective. The scarcity of electroencephalogram (EEG) data, coupled with individual and scenario variations, leads to considerable challenges in real-world EEG-based driver fatigue detection. We propose a domain adaptation method that utilizes EEG data collected from a laboratory to supplement real-world EEG data and constructs a cross-scenario and cross-subject driver fatigue detection model for real-world scenarios. Approach. First, we collect EEG data from subjects participating in a driving experiment conducted in both laboratory and real-world scenarios. To address the issue of data scarcity, we build a real-world fatigued driving detection model by integrating the real-world data with the laboratory data. Then, we propose a method named cross-scenario and cross-subject domain adaptation (CS2DA), which aims to eliminate the domain shift problem caused by individual variances and scenario differences. Adversarial learning is adopted to extract the common features observed across different subjects within the same scenario. The multikernel maximum mean discrepancy (MK-MMD) method is applied to further minimize scenario differences. Additionally, we propose a conditional MK-MMD constraint to better utilize label information. Finally, we use seven rules to fuse the predicted labels. Main results. We evaluate the CS2DA method through extensive experiments conducted on the two EEG datasets created in this work: the SEED-VLA and the SEED-VRW datasets. Different domain adaptation methods are used to construct a real-world fatigued driving detection model using data from laboratory and real-world scenarios, as well as a combination of both. Our findings show that the proposed CS2DA method outperforms the existing traditional and adversarial learning-based domain adaptation approaches. We also find that combining data from both laboratory and real-world scenarios improves the performance of the model. Significance. This study contributes two EEG-based fatigue driving datasets and demonstrates that the proposed CS2DA method can effectively enhance the performance of a real-world fatigued driving detection model.

Toward Graph Self-Supervised Learning With Contrastive Adjusted Zooming.

Graph representation learning (GRL) is critical for graph-structured data analysis. However, most of the existing graph neural networks (GNNs) heavily rely on labeling information, which is normally expensive to obtain in the real world. Although some existing works aim to effectively learn graph representations in an unsupervised manner, they suffer from certain limitations, such as the heavy reliance on monotone contrastiveness and limited scalability. To overcome the aforementioned problems, in light of the recent advancements in graph contrastive learning, we introduce a novel self-supervised GRL algorithm via graph contrastive adjusted zooming, namely, G-Zoom, to learn node representations by leveraging the proposed adjusted zooming scheme. Specifically, this mechanism enables G-Zoom to explore and extract self-supervision signals from a graph from multiple scales: micro (i.e., node level), meso (i.e., neighborhood level), and macro (i.e., subgraph level). First, we generate two augmented views of the input graph via two different graph augmentations. Then, we establish three different contrastiveness on the above three scales progressively, from node, neighboring, to subgraph level, where we maximize the agreement between graph representations across scales. While we can extract valuable clues from a given graph on the micro and macro perspectives, the neighboring-level contrastiveness offers G-Zoom the capability of a customizable option based on our adjusted zooming scheme to manually choose an optimal viewpoint that lies between the micro and macro perspectives to better understand the graph data. In addition, to make our model scalable to large graphs, we use a parallel graph diffusion approach to decouple model training from the graph size. We have conducted extensive experiments on real-world datasets, and the results demonstrate that our proposed model outperforms the state-of-the-art methods consistently.

Automated vehicle damage classification using the three-quarter view car damage dataset and deep learning approaches

Automated procedures for classifying vehicle damage are critical in industries requiring extensive vehicle management. Despite substantial research demands, challenges in the field of vehicle damage classification persist due to the scarcity of public datasets and the complexity of constructing datasets. In response to these challenges, we introduce a Three-Quarter View Car Damage Dataset (TQVCD dataset), emphasizing simplicity in labeling, data accessibility, and rich information inherent in three-quarter views. The TQVCD dataset distinguishes class by vehicle orientation (front or rear) and type of damage while maintaining a three-quarter view. We evaluate performance using five prevalent pre-trained deep learning architectures—ResNet-50, DenseNet-160, EfficientNet-B0, MobileNet-V2, and ViT—employing a suite of binary classification models. To enhance classification robustness, we implement a model ensemble method to effectively mitigate individual model dependencies' deviations. Additionally, we interview three experts from the used-car platform to validate the necessity of a vehicle damage classification model using the corresponding dataset from an industrial perspective. Empirical findings underscore the dataset's comprehensive coverage of vehicle perspectives, facilitating efficient data collection and damage classification while minimizing labor-intensive labeling efforts.

TGIN: Translation-Based Graph Inference Network for Few-Shot Relational Triplet Extraction.

Extracting relational triplets aims at detecting entity pairs and their semantic relations. Compared with pipeline models, joint models can reduce error propagation and achieve better performance. However, all of these models require large amounts of training data, therefore performing poorly on many long-tail relations in reality with insufficient data. In this article, we propose a novel end-to-end model, called TGIN, for few-shot triplet extraction. The core of TGIN is a multilayer heterogeneous graph with two types of nodes (entity node and relation node) and three types of edges (relation-entity edge, entity-entity edge, and relation-relation edge). On the one hand, this heterogeneous graph with entities and relations as nodes can intuitively extract relational triplets jointly, thereby reducing error propagation. On the other hand, it enables the triplet information of limited labeled data to interact better, thus maximizing the advantage of this information for few-shot triplet extraction. Moreover, we devise a graph aggregation and update method that utilizes translation algebraic operations to mine semantic features while retaining structure features between entities and relations, thereby improving the robustness of the TGIN in a few-shot setting. After updating the node and edge features through layers, TGIN propagates the label information from a few labeled examples to unlabeled examples, thus inferring triplets from these unlabeled examples. Extensive experiments on three reconstructed datasets demonstrate that TGIN can significantly improve the accuracy of triplet extraction by 2.34%~10.74% compared with the state-of-the-art baselines. To the best of our knowledge, we are the first to introduce a heterogeneous graph for few-shot relational triplet extraction.

Effectiveness of labels in digital art experience: psychophysiological and behavioral evidence.

Nowadays museums make large use of digital materials (e.g., virtual tours) to attract visitors. Therefore, it is worthwhile investigating which variables affect the engagement with art outside the museum, and whether digital reproductions of artworks are as effective as museum originals in producing a satisfying aesthetic experience. Here we tested the effectiveness of introducing additional informative materials on the artistic enjoyment of contemporary paintings presented on a computer screen. Naïve observers were exposed to essential and descriptive labels before viewing artworks. We flanked traditional measurement methods - viewing times and questionnaires, with biometric parameters - pupil responses, eye movements, heart rate, and electrodermal activity. The results were then compared to our previous museum study that adopted the same experimental paradigm. Our behavioral and psychophysiological data lead to a complex pattern of results. As found in the museum setting, providing detailed descriptions decreases complexity, evokes more positive sensations, and induces pupil dilation but does not enhance aesthetic appreciation. These results suggested that informative labels improve understanding and emotions but have a limited impact on the hedonic evaluation of artworks in both contexts. However, other results do not mirror those found in the museum; in the laboratory setting, participants spend a similar amount of time, have a comparable gaze behavior, and their electrodermal activity and heart rate do not change when viewing artworks with different types of labels. The main difference between the lab and museum settings is the shorter time spent viewing digital reproductions vs. real paintings, although subjective ratings (e.g., liking, interest) are comparable. Overall, this study indicates that the environmental context does impact the aesthetic experience; although, some beneficial effects of introducing additional relevant content in labels accompanying artworks can also be acquainted through digital media outside of the museum.