TY - JOUR A1 - Becker, Martin A1 - Caminiti, Saverio A1 - Fiorella, Donato A1 - Francis, Louise A1 - Gravino, Pietro A1 - Haklay, Mordechai (Muki) A1 - Hotho, Andreas A1 - Loreto, Virrorio A1 - Mueller, Juergen A1 - Ricchiuti, Ferdinando A1 - Servedio, Vito D. P. A1 - Sirbu, Alina A1 - Tria, Franesca T1 - Awareness and Learning in Participatory Noise Sensing JF - PLOS ONE N2 - The development of ICT infrastructures has facilitated the emergence of new paradigms for looking at society and the environment over the last few years. Participatory environmental sensing, i.e. directly involving citizens in environmental monitoring, is one example, which is hoped to encourage learning and enhance awareness of environmental issues. In this paper, an analysis of the behaviour of individuals involved in noise sensing is presented. Citizens have been involved in noise measuring activities through the WideNoise smartphone application. This application has been designed to record both objective (noise samples) and subjective (opinions, feelings) data. The application has been open to be used freely by anyone and has been widely employed worldwide. In addition, several test cases have been organised in European countries. Based on the information submitted by users, an analysis of emerging awareness and learning is performed. The data show that changes in the way the environment is perceived after repeated usage of the application do appear. Specifically, users learn how to recognise different noise levels they are exposed to. Additionally, the subjective data collected indicate an increased user involvement in time and a categorisation effect between pleasant and less pleasant environments. KW - exposure Y1 - 2013 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-127675 SN - 1932-6203 VL - 8 IS - 12 ER - TY - JOUR A1 - Schlör, Daniel A1 - Ring, Markus A1 - Hotho, Andreas T1 - iNALU: Improved Neural Arithmetic Logic Unit JF - Frontiers in Artificial Intelligence N2 - Neural networks have to capture mathematical relationships in order to learn various tasks. They approximate these relations implicitly and therefore often do not generalize well. The recently proposed Neural Arithmetic Logic Unit (NALU) is a novel neural architecture which is able to explicitly represent the mathematical relationships by the units of the network to learn operations such as summation, subtraction or multiplication. Although NALUs have been shown to perform well on various downstream tasks, an in-depth analysis reveals practical shortcomings by design, such as the inability to multiply or divide negative input values or training stability issues for deeper networks. We address these issues and propose an improved model architecture. We evaluate our model empirically in various settings from learning basic arithmetic operations to more complex functions. Our experiments indicate that our model solves stability issues and outperforms the original NALU model in means of arithmetic precision and convergence. KW - neural networks KW - machine learning KW - arithmetic calculations KW - neural architecture KW - experimental evaluation Y1 - 2020 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-212301 SN - 2624-8212 VL - 3 ER - TY - JOUR A1 - Ring, Markus A1 - Landes, Dieter A1 - Hotho, Andreas T1 - Detection of slow port scans in flow-based network traffic JF - PLoS ONE N2 - Frequently, port scans are early indicators of more serious attacks. Unfortunately, the detection of slow port scans in company networks is challenging due to the massive amount of network data. This paper proposes an innovative approach for preprocessing flow-based data which is specifically tailored to the detection of slow port scans. The preprocessing chain generates new objects based on flow-based data aggregated over time windows while taking domain knowledge as well as additional knowledge about the network structure into account. The computed objects are used as input for the further analysis. Based on these objects, we propose two different approaches for detection of slow port scans. One approach is unsupervised and uses sequential hypothesis testing whereas the other approach is supervised and uses classification algorithms. We compare both approaches with existing port scan detection algorithms on the flow-based CIDDS-001 data set. Experiments indicate that the proposed approaches achieve better detection rates and exhibit less false alarms than similar algorithms. Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-226305 VL - 13 IS - 9 ER - TY - JOUR A1 - Davidson, Padraig A1 - Düking, Peter A1 - Zinner, Christoph A1 - Sperlich, Billy A1 - Hotho, Andreas T1 - Smartwatch-Derived Data and Machine Learning Algorithms Estimate Classes of Ratings of Perceived Exertion in Runners: A Pilot Study JF - Sensors N2 - The rating of perceived exertion (RPE) is a subjective load marker and may assist in individualizing training prescription, particularly by adjusting running intensity. Unfortunately, RPE has shortcomings (e.g., underreporting) and cannot be monitored continuously and automatically throughout a training sessions. In this pilot study, we aimed to predict two classes of RPE (≤15 “Somewhat hard to hard” on Borg’s 6–20 scale vs. RPE >15 in runners by analyzing data recorded by a commercially-available smartwatch with machine learning algorithms. Twelve trained and untrained runners performed long-continuous runs at a constant self-selected pace to volitional exhaustion. Untrained runners reported their RPE each kilometer, whereas trained runners reported every five kilometers. The kinetics of heart rate, step cadence, and running velocity were recorded continuously ( 1 Hz ) with a commercially-available smartwatch (Polar V800). We trained different machine learning algorithms to estimate the two classes of RPE based on the time series sensor data derived from the smartwatch. Predictions were analyzed in different settings: accuracy overall and per runner type; i.e., accuracy for trained and untrained runners independently. We achieved top accuracies of 84.8 % for the whole dataset, 81.8 % for the trained runners, and 86.1 % for the untrained runners. We predict two classes of RPE with high accuracy using machine learning and smartwatch data. This approach might aid in individualizing training prescriptions. KW - artificial intelligence KW - endurance KW - exercise intensity KW - precision training KW - prediction KW - wearable Y1 - 2020 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-205686 SN - 1424-8220 VL - 20 IS - 9 ER - TY - JOUR A1 - Sîrbu, Alina A1 - Becker, Martin A1 - Caminiti, Saverio A1 - De Baets, Bernard A1 - Elen, Bart A1 - Francis, Louise A1 - Gravino, Pietro A1 - Hotho, Andreas A1 - Ingarra, Stefano A1 - Loreto, Vittorio A1 - Molino, Andrea A1 - Mueller, Juergen A1 - Peters, Jan A1 - Ricchiuti, Ferdinando A1 - Saracino, Fabio A1 - Servedio, Vito D.P. A1 - Stumme, Gerd A1 - Theunis, Jan A1 - Tria, Francesca A1 - Van den Bossche, Joris T1 - Participatory Patterns in an International Air Quality Monitoring Initiative JF - PLoS ONE N2 - The issue of sustainability is at the top of the political and societal agenda, being considered of extreme importance and urgency. Human individual action impacts the environment both locally (e.g., local air/water quality, noise disturbance) and globally (e.g., climate change, resource use). Urban environments represent a crucial example, with an increasing realization that the most effective way of producing a change is involving the citizens themselves in monitoring campaigns (a citizen science bottom-up approach). This is possible by developing novel technologies and IT infrastructures enabling large citizen participation. Here, in the wider framework of one of the first such projects, we show results from an international competition where citizens were involved in mobile air pollution monitoring using low cost sensing devices, combined with a web-based game to monitor perceived levels of pollution. Measures of shift in perceptions over the course of the campaign are provided, together with insights into participatory patterns emerging from this study. Interesting effects related to inertia and to direct involvement in measurement activities rather than indirect information exposure are also highlighted, indicating that direct involvement can enhance learning and environmental awareness. In the future, this could result in better adoption of policies towards decreasing pollution. KW - transport microenvironments KW - exposure KW - pollution KW - carbon Y1 - 2015 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-151379 VL - 10 IS - 8 ER - TY - JOUR A1 - Düking, Peter A1 - Hotho, Andreas A1 - Holmberg, Hans-Christer A1 - Fuss, Franz Konstantin A1 - Sperlich, Billy T1 - Comparison of Non-Invasive Individual Monitoring of the Training and Health of Athletes with Commercially Available Wearable Technologies JF - Frontiers in Physiology N2 - Athletes adapt their training daily to optimize performance, as well as avoid fatigue, overtraining and other undesirable effects on their health. To optimize training load, each athlete must take his/her own personal objective and subjective characteristics into consideration and an increasing number of wearable technologies (wearables) provide convenient monitoring of various parameters. Accordingly, it is important to help athletes decide which parameters are of primary interest and which wearables can monitor these parameters most effectively. Here, we discuss the wearable technologies available for non-invasive monitoring of various parameters concerning an athlete's training and health. On the basis of these considerations, we suggest directions for future development. Furthermore, we propose that a combination of several wearables is most effective for accessing all relevant parameters, disturbing the athlete as little as possible, and optimizing performance and promoting health. KW - sports technology KW - wearable technologies KW - performance parameters KW - health monitoring KW - performance monitoring Y1 - 2016 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-165516 VL - 7 IS - 71 ER - TY - JOUR A1 - Sperlich, Billy A1 - Becker, Martin A1 - Hotho, Andreas A1 - Wallmann-Sperlich, Birgit A1 - Sareban, Mahdi A1 - Winkert, Kay A1 - Steinacker, Jürgen M. A1 - Treff, Gunnar T1 - Sedentary behavior among national elite rowers during off-training — a pilot study JF - Frontiers in Physiology N2 - The aim of this pilot study was to analyze the off-training physical activity (PA) profile in national elite German U23 rowers during 31 days of their preparation period. The hours spent in each PA category (i.e., sedentary: <1.5 metabolic equivalents (MET); light physical activity: 1.5–3 MET; moderate physical activity: 3–6 MET and vigorous intense physical activity: >6 MET) were calculated for every valid day (i.e., >480 min of wear time). The off-training PA during 21 weekdays and 10 weekend days of the final 11-week preparation period was assessed by the wrist-worn multisensory device Microsoft Band II (MSBII). A total of 11 rowers provided valid data (i.e., >480 min/day) for 11.6 week days and 4.8 weekend days during the 31 days observation period. The average sedentary time was 11.63 ± 1.25 h per day during the week and 12.49 ± 1.10 h per day on the weekend, with a tendency to be higher on the weekend compared to weekdays (p = 0.06; d = 0.73). The average time in light, moderate and vigorous PA during the weekdays was 1.27 ± 1.15, 0.76 ± 0.37, 0.51 ± 0.44 h per day, and 0.67 ± 0.43, 0.59 ± 0.37, 0.53 ± 0.32 h per weekend day. Light physical activity was higher during weekdays compared to the weekend (p = 0.04; d = 0.69). Based on our pilot study of 11 national elite rowers we conclude that rowers display a considerable sedentary off-training behavior of more than 11.5 h/day. KW - recovery KW - sedentary behavior KW - accelerometer KW - microsoft band 2 KW - multi-sensor KW - wearable Y1 - 2017 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-158753 VL - 8 IS - 655 ER - TY - JOUR A1 - Steininger, Michael A1 - Kobs, Konstantin A1 - Davidson, Padraig A1 - Krause, Anna A1 - Hotho, Andreas T1 - Density-based weighting for imbalanced regression JF - Machine Learning N2 - In many real world settings, imbalanced data impedes model performance of learning algorithms, like neural networks, mostly for rare cases. This is especially problematic for tasks focusing on these rare occurrences. For example, when estimating precipitation, extreme rainfall events are scarce but important considering their potential consequences. While there are numerous well studied solutions for classification settings, most of them cannot be applied to regression easily. Of the few solutions for regression tasks, barely any have explored cost-sensitive learning which is known to have advantages compared to sampling-based methods in classification tasks. In this work, we propose a sample weighting approach for imbalanced regression datasets called DenseWeight and a cost-sensitive learning approach for neural network regression with imbalanced data called DenseLoss based on our weighting scheme. DenseWeight weights data points according to their target value rarities through kernel density estimation (KDE). DenseLoss adjusts each data point’s influence on the loss according to DenseWeight, giving rare data points more influence on model training compared to common data points. We show on multiple differently distributed datasets that DenseLoss significantly improves model performance for rare data points through its density-based weighting scheme. Additionally, we compare DenseLoss to the state-of-the-art method SMOGN, finding that our method mostly yields better performance. Our approach provides more control over model training as it enables us to actively decide on the trade-off between focusing on common or rare cases through a single hyperparameter, allowing the training of better models for rare data points. KW - supervised learning KW - imbalanced regression KW - cost-sensitive learning KW - sample weighting KW - Kerneldensity estimation Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-269177 SN - 1573-0565 VL - 110 IS - 8 ER - TY - JOUR A1 - Hentschel, Simon A1 - Kobs, Konstantin A1 - Hotho, Andreas T1 - CLIP knows image aesthetics JF - Frontiers in Artificial Intelligence N2 - Most Image Aesthetic Assessment (IAA) methods use a pretrained ImageNet classification model as a base to fine-tune. We hypothesize that content classification is not an optimal pretraining task for IAA, since the task discourages the extraction of features that are useful for IAA, e.g., composition, lighting, or style. On the other hand, we argue that the Contrastive Language-Image Pretraining (CLIP) model is a better base for IAA models, since it has been trained using natural language supervision. Due to the rich nature of language, CLIP needs to learn a broad range of image features that correlate with sentences describing the image content, composition, environments, and even subjective feelings about the image. While it has been shown that CLIP extracts features useful for content classification tasks, its suitability for tasks that require the extraction of style-based features like IAA has not yet been shown. We test our hypothesis by conducting a three-step study, investigating the usefulness of features extracted by CLIP compared to features obtained from the last layer of a comparable ImageNet classification model. In each step, we get more computationally expensive. First, we engineer natural language prompts that let CLIP assess an image's aesthetic without adjusting any weights in the model. To overcome the challenge that CLIP's prompting only is applicable to classification tasks, we propose a simple but effective strategy to convert multiple prompts to a continuous scalar as required when predicting an image's mean aesthetic score. Second, we train a linear regression on the AVA dataset using image features obtained by CLIP's image encoder. The resulting model outperforms a linear regression trained on features from an ImageNet classification model. It also shows competitive performance with fully fine-tuned networks based on ImageNet, while only training a single layer. Finally, by fine-tuning CLIP's image encoder on the AVA dataset, we show that CLIP only needs a fraction of training epochs to converge, while also performing better than a fine-tuned ImageNet model. Overall, our experiments suggest that CLIP is better suited as a base model for IAA methods than ImageNet pretrained networks. KW - Image Aesthetic Assessment KW - CLIP KW - language-image pre-training KW - text supervision KW - prompt engineering KW - AVA Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-297150 SN - 2624-8212 VL - 5 ER - TY - JOUR A1 - Wienrich, Carolin A1 - Carolus, Astrid A1 - Roth-Isigkeit, David A1 - Hotho, Andreas T1 - Inhibitors and enablers to explainable AI success: a systematic examination of explanation complexity and individual characteristics JF - Multimodal Technologies and Interaction N2 - With the increasing adaptability and complexity of advisory artificial intelligence (AI)-based agents, the topics of explainable AI and human-centered AI are moving close together. Variations in the explanation itself have been widely studied, with some contradictory results. These could be due to users’ individual differences, which have rarely been systematically studied regarding their inhibiting or enabling effect on the fulfillment of explanation objectives (such as trust, understanding, or workload). This paper aims to shed light on the significance of human dimensions (gender, age, trust disposition, need for cognition, affinity for technology, self-efficacy, attitudes, and mind attribution) as well as their interplay with different explanation modes (no, simple, or complex explanation). Participants played the game Deal or No Deal while interacting with an AI-based agent. The agent gave advice to the participants on whether they should accept or reject the deals offered to them. As expected, giving an explanation had a positive influence on the explanation objectives. However, the users’ individual characteristics particularly reinforced the fulfillment of the objectives. The strongest predictor of objective fulfillment was the degree of attribution of human characteristics. The more human characteristics were attributed, the more trust was placed in the agent, advice was more likely to be accepted and understood, and important needs were satisfied during the interaction. Thus, the current work contributes to a better understanding of the design of explanations of an AI-based agent system that takes into account individual characteristics and meets the demand for both explainable and human-centered agent systems. KW - explainable AI KW - human-centered AI KW - recommender agent KW - explanation complexity KW - individual differences Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-297288 SN - 2414-4088 VL - 6 IS - 12 ER - TY - JOUR A1 - Koopmann, Tobias A1 - Stubbemann, Maximilian A1 - Kapa, Matthias A1 - Paris, Michael A1 - Buenstorf, Guido A1 - Hanika, Tom A1 - Hotho, Andreas A1 - Jäschke, Robert A1 - Stumme, Gerd T1 - Proximity dimensions and the emergence of collaboration: a HypTrails study on German AI research JF - Scientometrics N2 - Creation and exchange of knowledge depends on collaboration. Recent work has suggested that the emergence of collaboration frequently relies on geographic proximity. However, being co-located tends to be associated with other dimensions of proximity, such as social ties or a shared organizational environment. To account for such factors, multiple dimensions of proximity have been proposed, including cognitive, institutional, organizational, social and geographical proximity. Since they strongly interrelate, disentangling these dimensions and their respective impact on collaboration is challenging. To address this issue, we propose various methods for measuring different dimensions of proximity. We then present an approach to compare and rank them with respect to the extent to which they indicate co-publications and co-inventions. We adapt the HypTrails approach, which was originally developed to explain human navigation, to co-author and co-inventor graphs. We evaluate this approach on a subset of the German research community, specifically academic authors and inventors active in research on artificial intelligence (AI). We find that social proximity and cognitive proximity are more important for the emergence of collaboration than geographic proximity. KW - collaboration KW - dimensions of proximity KW - co-authorships KW - co-inventorships KW - embedding techniques Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-269831 SN - 1588-2861 VL - 126 IS - 12 ER - TY - JOUR A1 - Steininger, Michael A1 - Abel, Daniel A1 - Ziegler, Katrin A1 - Krause, Anna A1 - Paeth, Heiko A1 - Hotho, Andreas T1 - ConvMOS: climate model output statistics with deep learning JF - Data Mining and Knowledge Discovery N2 - Climate models are the tool of choice for scientists researching climate change. Like all models they suffer from errors, particularly systematic and location-specific representation errors. One way to reduce these errors is model output statistics (MOS) where the model output is fitted to observational data with machine learning. In this work, we assess the use of convolutional Deep Learning climate MOS approaches and present the ConvMOS architecture which is specifically designed based on the observation that there are systematic and location-specific errors in the precipitation estimates of climate models. We apply ConvMOS models to the simulated precipitation of the regional climate model REMO, showing that a combination of per-location model parameters for reducing location-specific errors and global model parameters for reducing systematic errors is indeed beneficial for MOS performance. We find that ConvMOS models can reduce errors considerably and perform significantly better than three commonly used MOS approaches and plain ResNet and U-Net models in most cases. Our results show that non-linear MOS models underestimate the number of extreme precipitation events, which we alleviate by training models specialized towards extreme precipitation events with the imbalanced regression method DenseLoss. While we consider climate MOS, we argue that aspects of ConvMOS may also be beneficial in other domains with geospatial data, such as air pollution modeling or weather forecasts. KW - Klima KW - Modell KW - Deep learning KW - Neuronales Netz KW - climate KW - neural networks KW - model output statistics Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-324213 SN - 1384-5810 VL - 37 IS - 1 ER - TY - JOUR A1 - Wienrich, Carolin A1 - Carolus, Astrid A1 - Markus, André A1 - Augustin, Yannik A1 - Pfister, Jan A1 - Hotho, Andreas T1 - Long-term effects of perceived friendship with intelligent voice assistants on usage behavior, user experience, and social perceptions JF - Computers N2 - Social patterns and roles can develop when users talk to intelligent voice assistants (IVAs) daily. The current study investigates whether users assign different roles to devices and how this affects their usage behavior, user experience, and social perceptions. Since social roles take time to establish, we equipped 106 participants with Alexa or Google assistants and some smart home devices and observed their interactions for nine months. We analyzed diverse subjective (questionnaire) and objective data (interaction data). By combining social science and data science analyses, we identified two distinct clusters—users who assigned a friendship role to IVAs over time and users who did not. Interestingly, these clusters exhibited significant differences in their usage behavior, user experience, and social perceptions of the devices. For example, participants who assigned a role to IVAs attributed more friendship to them used them more frequently, reported more enjoyment during interactions, and perceived more empathy for IVAs. In addition, these users had distinct personal requirements, for example, they reported more loneliness. This study provides valuable insights into the role-specific effects and consequences of voice assistants. Recent developments in conversational language models such as ChatGPT suggest that the findings of this study could make an important contribution to the design of dialogic human–AI interactions. KW - intelligent voice assistant KW - smart speaker KW - social relationship KW - social role KW - long-term analysis KW - social interaction KW - human–computer interaction KW - anthropomorphism Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-313552 SN - 2073-431X VL - 12 IS - 4 ER -