Heterogeneous Users Research Articles

Heterogeneous Information Network Embedding for Mention Recommendation

Mention recommendation is the task of recommending the right candidate users in a message. Many works have been conducted on the problem of whom to mention. However, due to the sparsity and heterogeneous of mention data, none of them well solve the problem. The recent advances in network embedding representation learning provide an effective approach to model the sparsity and heterogeneous simultaneously in heterogeneous information network. To this end, we propose a novel Network Embedding Mention (NEM) recommendation model to recommend the right users in a message. NEM constructs a heterogeneous mention network based on different relationships among different entities. Then NEM learns a unified low dimensional embedding vector using random walk for users and messages by considering network structure and vertex content information. Finally, whom to mention is ranked by calculating the relevance scores from heterogeneous user and message embeddings. To evaluate the proposed method, we construct a large dataset and their corresponding social networks from a real-world social media platform. Through extensive experiments on real-world mention collection, we demonstrate that our proposed model outperforms the previous state-of-the-art methods in term of recommendation task.

Cognitive Data Offloading in Mobile Edge Computing for Internet of Things

Data offloading to Mobile Edge Computing (MEC) servers is an attractive choice for resource-constrained Internet of Things (IoT) devices, towards reducing their computational effort. In this paper, we investigate the potential of partial data offloading to MEC servers, under the perspective of users’ cognitive IoT devices presenting loss averse and gain seeking behavior. Due to the sharing nature of the access environment and the MEC server’s computational characteristics, we treat the MEC server option as a common pool of resources with uncertain payoff returned to the users, while the local computation capability is treated as a safe option for each user. Following the properties of Prospect Theory, users’ prospect-theoretic utilities are formulated exploiting the local computing and offloading overhead options under probabilistic uncertainty. Such a modeling allows for the infusion of human awareness, inherent cognitive biases and behavioral characteristics into the devices’ operation, their data offloading decisions and the edge computing environment that the devices are interacting with. Accordingly, each user’s optimal offloaded data to the MEC server is obtained as the outcome of a non-cooperative game, with users attempting to maximize their own utilities. The existence and uniqueness of a Pure Nash Equilibrium (PNE) are proven under the probabilistic nature of the respective payoff functions, while a distributed algorithm that convergences to the PNE is designed. Numerical results are provided that demonstrate the operation and superiority of the proposed framework under different IoT scenarios and behaviors, considering both homogeneous and heterogeneous users.

Flow pattern and optimal capacity in a bi-modal traffic corridor with heterogeneous users

This paper considers a bi-modal traffic corridor with a highway and a transit line where car-available-travelers tend to be more sensitive to crowding than car-unavailable-travelers in transit vehicles. Modal spilt and departure time choices of heterogeneous travelers (car-available- and car-unavailable-travelers) are considered. Two capacity design problems, namely total travel cost minimization with budget constraint and total system cost minimization with budget and equity constraints, are studied. It was found that when there is a construction/operational technology improvement for one traffic mode, the cost for its unit capacity reduces and its optimal total investment may increase (i.e. Jevons paradox).

Using Webcam Video Feed to Measure and Evaluate National Park Recreation Patterns

Park managers and researchers have increasingly sought new approaches for data collection to understand recreation patterns and experiences in a more sophisticated and holistic manner. Traditional methods of monitoring visitors through on-site observation and surveys can be time consuming, expensive, and prone to error, and park managers recognize the value of new technologies in providing reliable and cost-effective visitor monitoring programs that achieve a balance between precision, accuracy, and efficiency. Live webcam video feeds are becoming increasingly available online and provide an opportunity to monitor aspects of recreation that have previously been difficult to measure. Additionally, webcam monitoring, which can be done remotely, provides a cost-effective alternative to on-site data collection because it eliminates the need for travel costs and on-site personnel. The purpose of this study is to provide a methodological investigation of extracting recreation data from webcams to introduce park managers and leisure researchers to this new data collection technique. We provide a typology for the types of questions this method can address, carefully document an approach for conducting research with webcam video feeds, and provide a proof-of-principle through presenting a case study. Taken together, we demonstrate that an online approach using webcam footage for recreation research is feasible and viable. This research has broad applications in recreation contexts, allowing the study of detailed spatio-temporal recreation patterns, visit volumes, user heterogeneity, and large tour tracking. Webcam information can also be combined with relevant data on weather conditions, air quality, gas prices, or other conditions to learn about factors shaping the timing of recreation experiences. This study has implications for park and recreation managers, who make decisions about facilitating or hindering the use of webcams at their sites through their attention to expanding or contracting the supply of webcams on public lands, positioning cameras, and maintaining live feeds. Decisions made by park staff about webcams should not only consider the impacts of webcams on enhancing user experience, but also consider their use as a tool to monitor recreation patterns by park personnel and by leisure and tourism researchers more broadly.

Bilinear Pairings Signcryption Scheme For User Authentication Keys In Heterogeneous Networks

Heterogeneous network user authentication keys are vulnerable to plain text attacks, resulting in poor encryption performance. Therefore, a heterogeneous bilinear pairing signcryption scheme is proposed. The symbol frequency detection method is used to detect symbols of user authentication data, establish an ORK user protocol in a heterogeneous network, and perform a pairwise signcryption scheme to linearly encode user cognitive information in a heterogeneous network. Use Ryption technology to authenticate and re-sign users in heterogeneous networks, and use a bilinear pairing signature encryption scheme to control cipher text in user processes to encrypt identity authentication information in heterogeneous networks. Establish a statistical analysis model to encrypt user authentication information in a heterogeneous network to achieve intelligence. Simulation results show that the method has good scrambling ability and strong anti-attack ability. It is used for bilinear pairing signature encryption of user authentication keys in heterogeneous networks.

Generating Recommendations From Multiple Data Sources: A Methodological Framework for System Design and Its Application

Recommender systems (RSs) are systems that produce individualized recommendations as output or drive the user in a personalized way to interesting or useful objects in a space of possible options. Recently, RSs emerged as an effective support for decision making. However, when people make decisions, they usually take into account different and often conflicting information such as preferences, long-term goals, context, and their current condition. This complexity is often ignored by RSs. In order to provide an effective decision-making support, a RS should be “holistic”, i.e., it should rely on a complete representation of the user, encoding heterogeneous user features (such as personal interests, psychological traits, health data, social connections) that may come from multiple data sources. However, to obtain such holistic recommendations some steps are necessary: first, we need to identify the goal of the decision-making process; then, we have to exploit common-sense and domain knowledge to provide the user with the most suitable suggestions that best fit the recommendation scenario. In this article, we present a methodological framework that can drive researchers and developers during the design process of this kind of “holistic” RS. We also provide evidence of the framework validity by presenting the design process and the evaluation of a food RS based on holistic principles.

Market Share Contracts in B2B Procurement Settings with Heterogeneous User Preferences

Market share contracts, a form of loyalty discounts, where the discount is contingent on the buyer meeting or exceeding a target share of total procurement, are common in many B2B settings. We study the impact of such contracts on demand allocation, prices, and welfare in a setting where a single central B2B buyer procures multiple units of a product on behalf of a set of users with heterogeneous preferences. We find that regular (linear) pricing creates a demand distortion, which goes away with the use of market share contracts. These contracts serve as strategic tools for vendors with preferred products to shift the locus of competition and extract away rents from weaker rivals, and sometimes from buyers. The impact of such contracts on the welfare of the buyers is therefore ambiguous, but when these contracts are used, the overall surplus goes up as disutility from demand distortion is avoided. While similar in many respects, quantity threshold contracts cannot guarantee the avoidance of demand distortion when buyer demand is uncertain.

Joint Power and Resource Block Allocation for Mixed-Numerology-Based 5G Downlink Under Imperfect CSI

Fifth-generation (5G) of wireless networks are expected to accommodate different services with contrasting quality of service (QoS) requirements within a common physical infrastructure in an efficient way. In this article, we address the radio access network (RAN) slicing problem and focus on the three 5G primary services, namely, enhanced mobile broadband (eMBB), ultra-reliable and low-latency communications (URLLC) and massive machine-type communications (mMTC). In particular, we formulate the joint allocation of power and resource blocks to the heterogeneous users in the downlink targeting the transmit power minimization and by considering mixed numerology-based frame structures. Most importantly, the proposed scheme does not only consider the heterogeneous QoS requirements of each service, but also the queue status of each user during the scheduling of resource blocks. In addition, imperfect Channel State Information (CSI) is considered by including an outage probabilistic constraint into the formulation. The resulting non-convex problem is converted to a more tractable problem by exploiting Big-M formulation, probabilistic to non-probabilistic transformation, binary relaxation and successive convex approximation (SCA). The proposed solution is evaluated for different mixed-numerology resource grids within the context of strict slice-isolation and slice-aware radio resource management schemes via extensive numerical simulations.

Software-Defined Space-Air-Ground Integrated Network Architecture with the Multi-Layer Satellite Backbone Network

Under the background of the rapid development of ground mobile communication, the advantages of high coverage, survivability, and flexibility of satellite communication provide air support to the construction of space information network. According to the requirements of the future space information communication, a software-defined Space-Air-Ground Integrated network architecture was proposed. It consisted of layered structure satellite backbone network, deep space communication network, the stratosphere communication network and the ground network. The SpaceAir-Ground Integrated network was supported by the satellite backbone network. It provided data relay for the missions such as deep space exploration and controlled the deep-space spacecraft when needed. In addition, it safeguarded the anti-destructibility of stratospheric communication and assisted the stratosphere to supplement ground network communication. In this paper, algorithm requirements of the congestion control and routing of satellite backbone protocols for heterogeneous users’ services were proposed. The algorithm requirements of distinguishing different service objects for the deep space communication and stratospheric communication network protocols were described. Considering the realistic demand for the dynamic coverage of the satellite backbone network and node cost, the multi-layer satellite backbone network architecture was constructed. On this basis, the proposed Software-defined Space-Air-Ground Integrated network architecture could be built as a large, scalable and efficient communication network that could be integrated into space, air, and ground.

Cognitive Multi-Point Free Space Optical Communication: Real-Time Users Discovery Using Unsupervised Machine Learning

Multi-user free-space optical communication (FSOC) is beginning to draw a significant attention for its ability to support increased system network capacity while using single receiving photodiode and satisfying size, weight, and power (SWaP) constraints imposed by space- and aerial-based mobile communication. Despite these advantages, support of multi-user capabilities cause increased system complexity due to accommodating heterogenous users communications with varying transmission and data rate requirements. Machine learning (ML) has recently been considered as a promising approach for introducing cognition into the network to mitigate some of the complexity. A cognitive method based on unsupervised ML was derived for estimating the number of users communicating and sharing time and bandwidth resources with a single-node receiver. A weighted clustering approach was introduced and experimentally validated when users received with similar amplitude information that results in underestimation. Obtained results confirmed that the proposed methodology was able to accurately differentiate the number of simultaneously transmitting users with accuracy greater than 92%— even in the presence of moderate atmospheric turbulence. An experimental analysis was conducted to determine data size and receiver sampling rate requirements for accurate estimation. Furthermore, an empirical model was derived to evaluate the effect of preamble signal length given a particular sampling rate on the accuracy of the estimation. The model was validated for up-to four users.

Pareto-Improving Pricing Schemes for Route Assignment of Heterogeneous Users

Traffic congestion constitutes a major problem in commercial areas having negative effects on travel times, fuel consumption and other operational costs. Additionally, the continuously increasing use of GPS technologies has made drivers to make routing decisions in an effort to minimize their own individual travel time which is known to lead to an inefficient road usage. In this paper, we propose a novel pricing scheme to alleviate traffic congestion by controlling the freight routing decisions through a coordination mechanism. The proposed mechanism asks the truck drivers to declare their Origin-Destination (OD) pair and their individual Value Of Time (VOT) and guarantees that every participant truck driver will be better-off compared to the User Equilibrium (UE) by providing them individual incentives to truthfully declare their VOT while leading to a budget balanced on average mechanism. The optimum route assignment and the resulting pricing scheme can be calculated by solving a nonconvex optimization problem. To reduce the dimensionality of the problem, we propose a second pricing scheme and we prove that satisfies the aforementioned characteristics. Finally, the evaluation of our approach using the Sioux Falls network shows that the proposed pricing schemes can make the network approach the System Optimum (SO) solution.

Transportation Mode Detection by Embedded Sensors Based on Ensemble Learning

Context-aware computing has become a certainty due to the widespread use of smartphone devices equipped with sensors. A wide range of services, such as vehicular traffic monitoring and smart parking, can be accomplished with the help of awareness of user mobility. Transportation mode detection (TMD) using machine learning algorithms and the data captured from smartphone embedded sensors have attracted research community attention. In this research, ensemble learning is utilized to differentiate between transportation modes, including walking, standing, riding a train, driving a car, and riding a bus. The ensemble learning consists of three classifiers; each classifier votes independently on the instances, and the majority vote is applied for robust generalization. The proposed method was validated using three datasets; the samples included in these datasets were gathered by smartphone sensors (belonging to heterogeneous users), such as rotation vector sensors, accelerometers, uncalibrated gyroscopes, linear acceleration, orientation, speed, game rotation vector, sound, and gyroscopes. The proposed ensemble learning method achieves an accuracy of 89%, 93%, and 95% on the first, second, and third datasets, respectively, when 10% and 90% of the data are used for testing and training, respectively. On the other set of experiments, in which 30% and 70% of the data are used for testing and training, respectively, the proposed method yields accuracies of 86.8%, 92.1%, and 94.9% on the first, second, and third datasets, respectively. The proposed method shows promising results compared to existing human activity recognition (HAR) methods.

A FabLab as integrative part of a Learning Factory

The maker movement was born in the early 2000s when the first FabLab in the USA was opened at MIT. Since then a dynamic, worldwide growth of FabLabs has been recognized and nowadays they are important for education and research. The FabLab, which was opened at FH JOANNEUM in 2018 as an integrated part of the Learning Factory, offers heterogeneous user groups the opportunity to explore a wide range of issues in the context of Industry 4.0 and digitalization in an open high-tech factory. After twelve months of experience with the FabLab operation as an integrative component in the Learning Factory, first findings on a scientific and organizational level can be derived. This paper answers the questions of how a Learning Factory with integrated FabLab works and what benefits can be generated for the Learning Factory and the FabLab. First, the development of FabLabs and the motivation for implementing a FabLab at FH JOANNEUM will be discussed. This is followed by a description how a FabLab can be integrated as a sub-lab in an existing Learning Factory. In addition to this concept, the FabLab process is described and shows the effects on different user groups and the university. Further on the advantages and potentials of this concept are elaborated.

Proactive Data Center Management Using Predictive Approaches

Data Center (DC) management aims at promptly serving user requests while minimizing the energy consumed. This is achieved by turning off unnecessary servers to save energy and adapting the number of servers that are $on$ to the time-varying and heterogeneous user requests. A great change in the number of servers $on$ leads to a considerable management effort, also called control effort in the literature, which should be reduced as much as possible. Since feedback control can improve the performance of computing systems and networks, we propose to use it to achieve this dynamic capacity provisioning of the DC. In order to design this feedback control, first, we developed a dynamic model of the DC. The purpose of this paper is to design a feedback control strategy based on the DC model, able to optimize i) the Quality of Service, ii) the energy consumed and iii) the management effort. A simple Reactive open-loop Control which provides an amount of energy equal to the amount requested in the previous time interval is considered as a benchmark for comparison. Second, two feedback controls based on the balance equations of the DC are studied, namely i) Reactive Feedback Control providing an amount of energy equal to that provided by the reactive open-loop control but adding the accumulated demand that has not yet been served, and ii) Model Predictive Control optimizing a constrained cost that weights the management effort and the prediction error. Reactive Control, Reactive Feedback Control and Model Predictive Control are compared in terms of energy consumed, energy error and management effort. Quantitative results of the comparative performance evaluation are given, based on a data set collected from a real DC.

Optimal Pricing for Peer-to-Peer Sharing With Network Externalities

In this paper, we analyse how a peer-to-peer sharing platform should price its service to maximize profit, when user participation increases the value of the service to others by causing positive externalities. Modelling the service as an excludable public good, we propose a bounded utility model to capture many infrastructure sharing applications with bounded network value, in which complete coverage generates finite user valuation (e.g., WiFi or hotspot). Unbounded utility models are used to capture the large-scale user interactions in social media, where the network value follows Metcalfe's or Zipf's law. For these utility models, we analyze the optimal pricing schemes in the case of heterogeneous users under complete and incomplete information of users' service valuations. We propose the concept of `price of information' (PoI) to characterize the profit loss due to lack of information, and present asymptotic PoI bounds for different utility models. We also show that the difficult-to-implement differentiated pricing scheme, which is optimal under incomplete user information, can be replaced by a simple uniform price scheme that is asymptotic optimal. Finally, we extend our pricing schemes to a two-sided market by including a new group of `pure' service users who do not contribute to the public good, and show that the platform may charge zero price to the original group of users in order to attract this pure user group.

Collaboratively Replicating Encoded Content on RSUs to Enhance Video Services for Vehicles

With the development of smart cities, Internet services will be pervasively accessible for moving vehicles. It is envisioned that the video content demand of vehicles will explode in the near future. However, the strategy to efficiently distribute video content in large-scale vehicular networks is still absent due to challenges arising from the huge video population, heavy bandwidth consumption, heterogeneous user devices, and vehicles’ mobility. In this work, we propose to collaboratively replicate video content on Roadside Units (RSUs) to enhance video distribution services based on the fact that the contact period between moving vehicles and a single RSU is not long enough to complete video downloading. In our design, a video file is split into multiple chunks. Each RSU replicates a small number of original chunks and chunks encoded by network coding. Replicating encoded chunks can reduce redundancy of chunks on different RSUs so that RSUs can complement each other better, whereas original chunks can be transrated to chunks with lower bitrates flexibly to fit in users’ devices. Therefore, we replicate both original and encoded chunks on RSUs to take advantages of both sides. Stochastic models are employed to analyze chunk download processes and a convex optimization problem is formulated to determine the optimal partition of space allocated to each kind of chunks. Furthermore, we extend our strategy to support video streaming services and empirically prove that the influence caused by limitations of network coding is moderate. In the end, we conduct extensive simulations which not only validate the accuracy of our models but also demonstrate that our strategy can effectively boost video distribution services.

Consensus reaching with heterogeneous user preferences, private input and privacy-preservation output

This paper deals with a generic problem of matching agents with underlying preferences while guaranteeing a certain level of privacy is met. As a general framework, we consider consumers and prosumers who trade energy on a platform. Consumers buy energy to the platform to maximize their usage benefit while minimizing the cost paid to the platform. Prosumers, who have the possibility to generate energy, self-consume part of it to maximize their usage benefit and sell the rest to the platform to maximize their revenue. Inspired by a variant of the Hotelling model, product differentiation is introduced and consumers can specify preferences regarding locality and green origin of their supply. The consumers and prosumers problems being coupled through a matching probability, we provide analytical characterization of the resulting Nash equilibrium, and conditions for existence and uniqueness. Assuming supply shortages occur on the platform, we reformulate the local market clearing problem as a consensus problem that we solve using Consensus Alternating Direction Method of Multipliers (C-ADMM), enabling minimal information exchanges between prosumers and consumers. C-ADMM complexity is recalled and strategyproofness is analysed. The algorithm is then run on a case study made of 300 prosumers from New South Wales in Australia, equipped with solar panels. We consider privacy-preservation output against a centralized benchmark approach, and evaluate C-ADMM computational time under three scenarios with an increasing number of agents. Regarding economic analysis, we observe that it is more profitable for prosumers than for consumers to be flexible within a local energy community, and that belonging to a local energy community incentivizes them to reduce their demands by comparison with their initial targets. Furthermore, the expectation to make a substantial profit is a main driver for prosumers’ engagement within a community; whereas for consumers, the green origin of the supply is determinant.

A joint neural network model for combining heterogeneous user data sources: An example of at‐risk student prediction

AbstractInformation service providers often require evidence from multiple, heterogeneous information sources to better characterize users and offer personalized service. In many cases, statistic information (for example, users' profiles) and sequentially dynamic information (for example, logs of interaction with information systems) are two prominent sources that can be combined to achieve optimized results. Previous attempts in combining these two sources mainly exploited models designed for either static or sequential information, but not both. This study aims to fill the gap by proposing a novel joint neural network model that can naturally fit both static and sequential user data. To evaluate the effectiveness of the proposed method, this study uses the problem of at‐risk student prediction as an example where both static data (personal profiles) and sequential data (event logs) are involved. A thorough evaluation was conducted on an open data set, with comparisons to a range of existing approaches including both static and sequential models. The results reveal superb performances of the proposed method. Implications of the findings on further research and applications of joint models are discussed.

Heterogeneity in desired bus service quality from users’ and potential users’ perspective

Abstract Investigating on transit users’ opinions represents a useful strategy to understand the level of quality of a service, and consequently pursuing the most convenient interventions for satisfying users and improving the service. On the other hand, an important issue for increasing the use of transit systems is linked to the necessity to attract new users, in addition to customize the current users. For this reason, it becomes fundamental to know also the perceptions of people who do not use transit systems with the aim to discover which are the reasons of their travel choices and particularly which are the most relevant transit service aspects for them. In this work, a Stated Preferences (SP) survey was conducted in order to investigate on both transit current and potential users’ preferences. This kind of survey allows potential users opinions to be captured due to its peculiarity of the possibility to propose to interviewees hypothetical services and not only real services, which could not be judged by people who do not use the service. The collected data were adopted to calibrate discrete choice models such as Random Parameters mixed Logit models and Latent Class models. These models allow heterogeneity of users’ opinions to be analysed. Calibrating this kind of models using data collected from both users and potential users permitted to capture the differences among users in their preferences about transit service quality, and the differences among potential users in their desired service quality. Model results showed some interesting findings concerning the differences of categories of users and potential users, and especially the diversity of preferences between people who know well the service and people who could become users even if they have not a real perception of it. We also calculated the Willingness-To-Pay (WTP), which allowed to discover that the WTP values obtained for the potential users are very much higher than the values obtained for the users. The proposed models can be also useful for the practitioners and transit operators to identify the best combination of the quality levels to assign to the various service aspects.

A personalized requirement identifying model for design improvement based on user profiling

AbstractThe personalization of products and services has become an inevitable trend in the manufacturing and service industry, but it is very difficult to identify users' personalized requirements accurately. This paper solves this problem by constructing an identifying model for personalized requirement based on user profiling. Firstly, the framework of the proposed model and the process of identifying the user's personalized requirements with this model are introduced, and then an experimental scheme for obtaining users' profiling data is designed. On this basis, an experiment is performed by investigating users' requirements for the computer to obtain the data, and the data are used for the analysis based on the proposed model. The analysis result shows that the model can reveal the difference among heterogeneous users well, find out the implicit requirements of users, and identify the gap between existing products and users' personalized requirements, which provides support to the subsequent improvement of product design.