Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Trust is an important factor influencing user acceptance of high-tech products. As the artificial intelligence and natural language processing develop, all kinds of conversational agents (chatbot) have appeared around us. These chatbots are able to provide people with convenient services such as ordering food, stock recommendations, fund diagnostics. However, it is still not clear how to make users feel chatbot trustworthy. In this study, we aimed to explore a set of design principles to build trust between users and conversational agents. Based on extensive research on trust, we proposed five design semantics and 10 design principles, and verified their effectiveness through experiments. The result of experiment suggest that our design principles can improve users’ trust towards chatbot, thus provided guidance and suggestions for designing more trustworthy chatbots in the future. PubDate: 2022-05-17
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Artificial Intelligence (AI) dramatically alters traditional healthcare and cognition assessment with its power in ubiquitous perception and smart computation. However, the existing research primarily concerns exploring the application areas and improving the recognition accuracy. How to apply AI with suitable and user-acceptable forms to realize cognitive health assessment is still a significant challenge. In this paper, we conduct a series of field studies to research this challenging problem. Specifically, inspired by clinically validated cognition assessment test—Trail Making Test (TMT), we design two variants of TMT for objective quantitative assessment, including camera-based TMT (cTMT) and touchscreen-based TMT (tTMT). Each form of variants provides three different analysis perspectives. We conduct user studies on 268 subjects to verify their effectiveness. Experimental results show that both variants can achieve satisfactory discrimination accuracy by optimizing the assessment model, but different application forms and analysis perspectives can adapt to different users. PubDate: 2022-05-11
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Analyzing user behaviors is a conventional approach to accomplish personalized recommendation. However, due to the intrinsic complexity of user behaviors, modeling the user behaviors in an accurate and comprehensive way to achieve effective recommendation is still a challenge. In the paper, we focus on user interaction behaviors and propose a model named Sirius, which designs graph neural networks to model the collaborative relations implied in the interactions and capture the dynamics of sequence features (including time and attribute features). Sirius constructs two kinds of graphs from interaction sequences and then builds two kinds of graph neural networks to jointly mine the implied relations in interactions. In particular, the sequence time feature and sequence attribute feature are fused into the generation of item and sequence embeddings. At last, Sirius gives the item recommendation by next item prediction. We conduct extensive experiments on multiple real-world datasets. The experimental results show that Sirius outperforms several state-of-the-art models in terms of recall and mean reciprocal rank (MRR). Moreover, Sirius has been deployed in MX Player, one of India’s largest streaming platforms, and achieves the improvement on online unique click-through rates (CTRs), which demonstrates the effectiveness and feasibility of Sirius in a real-world production environment. PubDate: 2022-05-06
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Mobile crowd sensing (MCS) is an attractive and innovation paradigm in which a crowd of users equipped with smart mobile devices conduct sensing tasks by fully exploiting their carried diverse embedded sensors. With the development of robots and Artificial Intelligence, many MCS studies with pervasive machines (Yu et al. Commun ACM 64:76–80, 2021) (e.g., unmanned vehicles, drones, etc.) as participants have emerged in recent years. Compared to human participants, robot participants have the advantages of being able to perform dangerous and boring tasks, being highly controlled, and not requiring complex incentive mechanisms. However, participants in previous studies usually have only one type of robot, and the use of heterogeneous robots for collaborative sensing was not considered. Second, previous studies have not considered the vulnerability of robots in realistic environments. In this paper, a multi-agent mobile crowdsensing (MA-MCS) system consisting of multiple heterogeneous robots is proposed to address the above two problems, and the task allocation problem of this system is investigated. To enable the robots to overcome the complex real-world environment, this paper proposes the concept of sense area information map (SAIM) and a self-repairing task allocation algorithm based on the information map. The SAIM can reflect the performance of different robots performing sense tasks in different locations and at different times, and provide guidance for task allocation. The self-repairing task assignment algorithm can be used to repair and reassign tasks after the robot has encountered abnormal situations. Through experiments, it is demonstrated that the SAIM and the map-based self-repairing task allocation algorithm can effectively improve task coverage at the expense of certain energy consumption. PubDate: 2022-04-28
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The 5G New Radio Unlicensed (5G-U) technology has enabled manufacturing enterprises to deploy their own private industrial networks, making anomaly event detection more necessary for maintaining wireless communication quality. However, existing statistical analysis algorithms cannot efficiently and accurately detect various kinds of anomaly events caused by the complex industrial environment. These events include electromagnetic interference as well as contention between cross-technology devices for unlicensed spectrum resources. To improve the efficiency, we design a classification algorithm that uses feature extraction in the frequency domain and a convolutional neural network model to detect various kinds of anomaly events (e.g., loose antennas and co-channel interference). We prototyped Slade (Spectrum Learning for Anomaly Detection), an anomaly detection system for industrial 5G networks. To evaluate the system, we collect wireless spectrum data with two industrial 5G-U terminals. Our evaluation on the dataset shows that our methodology can accurately detect different anomaly events, with an accuracy of 97.6% and a recall of 97.1%. PubDate: 2022-04-20
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract With more and more in-depth research on blockchain core technologies such as consensus mechanisms, the overall performance of blockchain systems has been greatly improved. These dedicated blockchains form isolated information islands, and the lack of cross-chain interoperability seriously restricts the large-scale application of blockchain. Digital assets such as fungible and non-fungible tokens can flexibly model blockchain business scenarios, but how to combine digital assets to design cross-chain data model and operation mode remains to be studied. This paper proposes a cross-chain collaboration model and consistency maintenance method for digital assets. This method can capture the causal relationship between cross-chain operations and solve the conflict caused by concurrent cross-chain operations. While improving the responsiveness of cross-chain operation, the eventual consistency of cross-chain data is effectively maintained. Experiments have shown that this method does not depend on the network between chains and the delay of consensus, which has higher responsiveness. PubDate: 2022-04-13
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The growing abundance in complex network data models is constantly increasing the challenges for non-expert users who perform an effective exploratory search in large data collections. In such domains, users search for entities related to a topic of interest and acquire knowledge by investigating the relationships between these entities. Designers, in turn, are challenged by the need to provide tools that enable convenient search and exploration to facilitate productive performance on the task. For this purpose, we introduce HiveRel, a search system that presents search results as tiled hexagons on a map-like surface with center-out relevance ordering and allows on-demand display of relationships between search results. HiveRel’s user interface is based on theoretical principles that reflect how users acquire knowledge through relationships. For the search mechanism, we provide a set of information retrieval definitions leading to the formalization of the Maximal n-Bounded Exploration Subgraph problem and present an implementation of a greedy heuristic algorithm that provides non-optimal solutions to this problem. We develop a proof of concept version of HiveRel. We evaluate it in two user studies that compare users’ performance using HiveRel to standard web search over a range of search knowledge acquisition tasks and two different domains. The results indicate that despite the lack of familiarity with the new system, users were generally more accurate and as fast using HiveRel, and provided positive evaluations for the search experience. PubDate: 2022-04-11
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Indoor localization systems are extensively used to develop positioning in various public buildings, and warehouses, for localization and navigation of users, robots and/or tracking assets. Researchers have developed and worked on variegated technologies such as, Bluetooth Low Energy, motion planning, Received Signal Strength based fingerprinting and mapping for achieving localization. Inertial Measurement Units (IMUs) are widely used in navigation that utilizes accelerometer, magnetometer, and gyroscope to sense acceleration, magnetic field, and angular rate respectively for navigation. IMUs are not only available as wearable sensors but also present in smartphones that are widely carried by users nowadays. Thus, ubiquitous localization systems can be designed with smartphone based IMU sensors. Existing survey articles on indoor localization has mostly focused on the different technologies available, and the different approaches utilized. However, existing works on IMU sensing based user localization methods need special attention as they can be extended toward a ubiquitous localization system that requires minimal fingerprinting effort from the public buildings. Accordingly, the article focuses on providing in-depth knowledge of the working procedure and discusses the challenges smartphone IMU faces. The article also surveys the fusion-based techniques used in indoor positioning and presents a comparative study of the various approaches. PubDate: 2022-03-18
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Natural disasters cause extensive damages to urban cities, demanding authorities to take urgent and effective measures to restore normalcy on transportation. During a disaster, restoring road transportation in a timely manner for rescue, supply and also prevent the risk of road accidents due to obstacles is of vital importance. Traditional post-disaster road obstacle work relies on a manual investigation which is time-consuming and labor-intensive. Predicting road risks can provide decision support for emergency management departments, reducing the damage caused by disasters. In this paper, we propose a three-phase framework for predicting road risks post-disaster leveraging heterogeneous urban data. Firstly, We use a clustering algorithm to extract and classify urban road networks based on the floating car data. Then we extract the spatiotemporal features of the urban roads. Through social network data, we collect historical risk-prone data using the crowdsensing method. To address the challenges of the small amount of labeled data, we train our model based on self-training. We verified the validity of this model by using a real dataset in Xiamen island which proves that our model accurately predicts road risk with precision and recall both more than 85% respectively. PubDate: 2022-03-07 DOI: 10.1007/s42486-022-00095-5
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract In this paper, we examine the psychological antecedents of privacy management strategies in social network sites (SNS) and extend the understanding to collective privacy management. By surveying Facebook users in the US (N = 454), Singapore (N = 467) and South Korea (N = 472), we are able to test our prediction model in these three countries to examine whether the effects of the antecedents are robust from a cross-country perspective. Although some of the effects are significantly different between the three countries, we find that in general users’ privacy attitudes, social norms, and self/collective control beliefs substantially predict the adoption of collective privacy management strategies. These findings contribute to the explanations of users’ adoption of behavioral privacy management strategies. We conclude with global and country-specific recommendations regarding future privacy designs for collective privacy management. PubDate: 2022-03-04 DOI: 10.1007/s42486-022-00092-8
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Emergency vehicle priority (EVP) systems are the need of the hour to reduce the transit time of emergency vehicles in cities. As these cities are major hubs of economic activity they are one of the most densely populated cities in the world. Due to numerous such issues, ambulances are not able to reach patients and hospitals on time. In this paper, we propose a system that detects an ambulance accurately and helps set up a makeshift emergency lane on the routes to be taken by it. The system relies on a neural network-based siren classifier to detect the ambulance using audio processing. The overall accuracy of the siren classifier was 97.2 %. After the ambulance is detected this information is then passed onto a network of Internet of Things (IoT) devices that activate visual indicators on the routes to be taken by the ambulance. On activating the visual indicators the traffic on those roads can start making a temporary emergency lane. The system uses a GPS-based mobile app to get route information of the ambulance. The network of IoT devices consists of a host device and station/node devices in a chain-like connection, where all devices are communicating via local WiFi networks. The host receives information about the ambulance from the neural network and the mobile app. The host then sends this information down the chain to other node devices. Through our proposed system we hope that the transit time of ambulances is reduced and hence accident victims, heart attack patients, etc can get medical attention faster. PubDate: 2022-02-22 DOI: 10.1007/s42486-022-00093-7
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Sound Event Detection (SED) is the task of detecting and demarcating the segments with specific semantics in audio recording. It has a promising application prospect in security monitoring, intelligent medical treatment, industrial production and so on. However, SED is still in the early stage of development and it faces many challenges, including the lack of accurately annotated data and the poor performance on detection due to the overlap of sound events. In view of the above problems, considering the intelligence of human beings and their flexibility and adaptability in the face of complex problems and changing environment, this paper proposes an approach of human–machine collaboration based SED (HMSED). In order to reduce the cost of labeling data, we first employ two CNN models with embedding-level attention pool module for weakly-labeled SED. Second, in order to improve the abilities of these two models alternately, we propose an end-to-end guided learning process for semi-supervised learning. Third, we use a group of median filters with adaptive window size in the post-processing of output probabilities of the model. Fourth, the model is adjusted and optimized by combining the results of machine recognition and manual annotation feedback. Based on HTML and JavaScript, an interactive annotation interface for HMSED is developed. And we do extensive exploratory experiments on the effects of human workload, model structure, hyperparameter and adaptive post-processing. The result shows that the HMSED is superior to some classical SED approaches. PubDate: 2022-02-17 DOI: 10.1007/s42486-022-00091-9
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract With the rise of high-speed networks and remote collaboration platforms, it has become a trend for companies to form group collaboration teams based on open innovation, but there is still a lack of research and mechanism design for group collaboration practices. This research attempts to transfer the open innovation group collaboration model to the design team collaboration environment, and investigates the designer's behaviour model, including originality, profitability and output, based on the task of product colour matching design. The modelling method and the colour-matching design simulation platform proposed in this paper provide a real design platform that can be evaluated, enabling the strategic optimisation of combinations of design subjects with multiple behavioural characteristics, and finding a relatively optimal teamwork model by adjusting the team structure, providing a theoretical and practical basis for the personalised design of design team operation models. PubDate: 2022-01-21 DOI: 10.1007/s42486-022-00088-4
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The trend of IoT brings more and more connected smart devices into our daily lives, which can enable a ubiquitous sensing and interaction experience. However, augmenting many everyday objects with sensing abilities is not easy. BitID is an unobtrusive, low-cost, training-free, and easy-to-use technique that enables users to add sensing abilities to everyday objects in a DIY manner. A BitID sensor can be easily made from a UHF RFID tag and deployed on an object so that the tag’s readability (whether the tag is identified by RFID readers) is mapped to binary states of the object (e.g., whether a door is open or closed). To further validate BitID’s sensing performance, we use a robotic arm to press BitID buttons repetitively and swipe on BitID sliders. The average press recognition F1-score is 98.9% and the swipe recognition F1-score is 96.7%. To evaluate BitID’s usability, we implement a prototype system that supports BitID sensor registration, semantic definition, status display, and real-time state and event detection. Using the system, users configured and deployed a BitID sensor with an average time duration of 4.9 min. 23 of the 24 users deployed BitID sensors worked accurately and robustly. In addition to the previously proposed ’short’ BitID sensor, we propose new ’open’ BitID sensors which show similar performance as ’short’ sensors. PubDate: 2022-01-17 DOI: 10.1007/s42486-022-00087-5
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract In this paper, we focus on an important recommendation problem known as one-class collaborative filtering (OCCF) and propose a novel preference assumption to model users’ implicit one-class feedback such as “examinations” or “likes” in the studied problem. Specifically, we address the limitations of previous pairwise preference learning works by defining the pairwise relations on user-groups and item-sets in the vertical dimension and in the horizontal dimension, respectively. On the basis of the proposed generic dual pairwise preference assumption, we develop a novel recommendation algorithm, i.e., collaborative filtering with implicit feedback via learning pairwise preferences over user-groups and item-sets (CoFi \(^+\) ). The main merit of our CoFi \(^+\) is its capacity for modeling both the horizontal and vertical ranking-oriented preference relations more sufficiently, as well as its generality of absorbing several existing pairwise preference learning algorithms as special cases. We conduct extensive empirical studies on three public datasets and find that our CoFi \(^+\) performs significantly better than the state-of-the-art methods. PubDate: 2022-01-08 DOI: 10.1007/s42486-021-00086-y
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Mobile Internet allows consumers the access to all musics on the mobile platform. Since it is not feasible to manually select music due to the size constraint of mobile devices, music recommendation has become a popular research topic in recent years, and researchers have proposed many effective methods such as collaborative filtering. However, with the tremendous increase of mobile users and music resources, customized music recommendations still face two challenges: (1) how to model complicated relations from user-music interaction data, and (2) how to integrate heterogeneous content information of musics. In this paper, we propose an Attentive Auto-encoder for Content-Aware Music Recommendation ( \(A^2CAMR\) ), which effectively integrates user behaviour records, music content, and similar musics of the target. In particular, we design a hierarchical attention-based encoder layer to learn fine-grained user-user and music-user relationships, thus produce behavior-based hidden representation of musics. We also employ an embedding layer to produce the content-based music representation, and cluster the similar music sets of the target music to predict users’ preferences in the decoder. We conduct extensive experiments on real-world dataset, and the results demonstrate the effectiveness of our model. PubDate: 2022-01-03 DOI: 10.1007/s42486-021-00083-1
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract The paper presents a qualitative study to explore the use of fitness trackers and their social functions in intergenerational settings. The study covered three phases of semi-structured interviews with older and younger adults during individual and intergenerational use of the fitness trackers. The study revealed comparability as common fitness practice for older adults. The findings show that intergenerational fitness tracking practices can increase in-person meetings and daily discourses and thus enhance family social bonds. An unexpected benefit of this practice is its ability to help older adults overcome technology barriers related to the use of fitness trackers. Overall speaking, families whose intergenerational members already enjoy a strong relationship are likely to gain the most from such practices. Many challenges remain especially concerning the motivation and involvement of younger partners and the user experience design aspect of such digital programs. For this purpose, we have developed some recommendations for the future development and deployment of intergenerational fitness tracking systems to stimulate interactions between younger and older family members and thus to promote their physical and emotional well-being. PubDate: 2021-11-16 DOI: 10.1007/s42486-021-00082-2
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Over the past decade, WiFi CSI-based device-free sensing technology has shown great potential in smart homes, assisted living, and many other applications. While model-based device-free sensing approaches analyze and recognize human behaviors by constructing mathematical relationships among WiFi devices, environment, human position/posture, and received channel state information, they have attracted great attention because of the interpretable physical meaning and the ability to guide the WiFi-based sensing system design. In this paper, we retrospect two general-purpose sensing models, i.e., the Fresnel zone model and CSI-ratio model, and demonstrate how these two models are leveraged to extract insightful properties and support a variety of device-free sensing applications. PubDate: 2021-09-28 DOI: 10.1007/s42486-021-00077-z
Please help us test our new pre-print finding feature by giving the pre-print link a rating. A 5 star rating indicates the linked pre-print has the exact same content as the published article.
Abstract: Abstract Image classification on mobile devices can provide convenient and secure services for users when using various social software. The traditional classification method mainly relies on the user’s manual marking, but the accuracy of automatic classification has some defects. With the development of convolutional neural network(CNN), the design of lightweight neural network has become a hot topic. However, the state-of-the-art studies always sacrifice classification accuracy for network lightweight, which greatly frustrates usability. In this paper, a new neural network framework, named MobVi, is proposed to enhance the precision of lightweight neural network by solution space division. MobVi is including image solution space division and judgment class. The former uses clustering method based on deep learning to distinguish which small solution space the image belongs to, while the latter uses lightweight neural network customized for the solution space to judge the class. In order to reduce the amount of model parameters and calculations, we designed a customized CNN module. Finally, we propose an energy prediction model to measure whether the model can be successfully implemented on mobile devices. A series of experiments have proved that MobVi has better performance than most existing models for mobile devices. Our model achieves 83.5% accuracy on CIFAR-10 data set, and the parameter quantity is only 2.0 M. PubDate: 2021-09-10 DOI: 10.1007/s42486-021-00076-0
Hybrid journal (It can contain Open Access articles)
[2469 journals]
