Agent-controlled intelligent subnetworks are envisioned as an integral part of Beyond 5G (B5G) communications for automatically selecting distinct policies based on dynamic performance constraints. In the B5G era, energy efficiency will be a prominent feature of Radio Access Network (RAN) slicing for achieving sustainable networks and lowering operational costs while multiplexing users with heterogeneous Quality-of-Service (QoS) requirements. Considering the importance of Intelligent Transportation Systems (ITS) within the realm of B5G applications, in this paper, we propose the LazyRAN framework as an energy-efficient Radio Resource Block (RRB) allocation approach for multi-policy RAN slicing in B5G ITS edge networks. Initially, we focus on resource utilization efficiency and define Lazy Skip Markov Decision Process (LS-MDP) formulation for spectrum agents to individually perform fine and coarse stochastic control depending on performance requirements incorporating varying levels of laziness. Followingly, we propose LazyRAN framework that uses Bayesian Optimization (BO) based Offline Policy Selection (OPS) for optimality calculations in case of multiple slicing policies. Our framework enables efficient multi-policy evaluation employing both exploration and exploitation in the agent policy space. The OPS method utilizes a Gaussian Process (GP) surrogate function combining logged data with online agent interactions before searching for the best slicing policy with BO. Using an energy-aware approach, hybrid QoS reward per energy consumption (HQEC), we compare the performance of LazyRAN framework in centralized and decentralized settings considering diverse agent policies. Our results show that the proposed scheme can significantly improve energy utilization with greater HQEC and higher throughput measurements while satisfying hybrid QoS demands.

This paper introduces an innovative wayfinding system featuring addressable LED strips mounted on the ground of an environment, controlled via conventional button interfaces and advanced voice interaction integrated with GPT-4. These LED strips provide visual cues to guide users to various rooms, enhancing navigation. We detail the hardware setup, including LEDs, controllers, and network infrastructure, along with the design of the user interfaces. The system’s performance was evaluated through a user survey, which indicated a preference for the touch interface while also highlighting the potential of voice interaction. Our findings demonstrate the system’s effectiveness in providing intuitive and flexible navigation assistance without requiring additional user hardware. The proposed approach, the Ground Guiding Assistant System (GGAS), is implemented and utilized daily in an $813 \mathrm{~m}^{2}$ office addressing 13 rooms at the Center for Tangible AI (ZEKI) in Berlin.

The ongoing shift towards electric vehicles and the simultaneous integration of sensor, computing and communication systems for automated driving pose challenges for the interplay between batteries and onboard autonomous mobility functions and services. Research on autonomous test vehicle (ATV) power consumption does not match the current realities of industry solutions brought to early markets worldwide. Current research predicts high power demands of commercial AVs, reflected in extrapolating data points of underlying hardware data sheets, simulations, or the piloting of ATVs. Although we can relay to significant power demands in ATVs, we identified different realities and higher efficiency in emerging industry vehicles. We show a counteracting reality in how AVs power demand is perceived due to a mismatch between ATVs and early market-ready AVs. To bridge the gap between research and practice we pose a power-demand model and core indicators to be considered in estimating power demands.

Object detectors in real-world applications often fail to detect objects due to varying factors such as weather conditions and noisy input. Therefore, a process that mitigates false detections is crucial for both safety and accuracy. While uncertainty-based thresholding shows promise, previous works demonstrate an imperfect correlation between uncertainty and detection errors. This hinders ideal thresholding, prompting us to further investigate the correlation and associated cost with different types of uncertainty. We therefore propose a cost-sensitive framework for object detection tailored to user-defined budgets on the two types of errors, missing and false detections. We derive minimum thresholding requirements to prevent performance degradation and define metrics to assess the applicability of uncertainty for failure recognition. Furthermore, we automate and optimize the thresholding process to maximize the failure recognition rate w.r.t. the specified budget. Evaluation on three autonomous driving datasets demonstrates that our approach significantly enhances safety, particularly in challenging scenarios. Leveraging localization aleatoric uncertainty and softmax-based entropy only, our method boosts the failure recognition rate by 36-60% compared to conventional approaches. Code is available at https://mos-ks.github.io/publications.

A limitation for collaborative robots (cobots) is their lack of ability to adapt to human partners, who typically exhibit an immense diversity of behaviors. We present an autonomous framework as a cobot’s real-time decision-making mechanism to anticipate a variety of human characteristics and behaviors, including human errors, toward a personalized collaboration. Our framework handles such behaviors in two levels: (1) short-term human behaviors are adapted through our novel Anticipatory Partially Observable Markov Decision Process (A-POMDP) models, covering a human’s changing intent (motivation), availability, and capability; (2) long-term changing human characteristics are adapted by our novel Adaptive Bayesian Policy Selection (ABPS) mechanism that selects a short-term decision model, e.g., an A-POMDP, according to an estimate of a human’s workplace characteristics, such as her expertise and collaboration preferences. To design and evaluate our framework over a diversity of human behaviors, we propose a pipeline where we first train and rigorously test the framework in simulation over novel human models. Then, we deploy and evaluate it on our novel physical experiment setup that induces cognitive load on humans to observe their dynamic behaviors, including their mistakes, and their changing characteristics such as their expertise. We conduct user studies and show that our framework effectively collaborates non-stop for hours and adapts to various changing human behaviors and characteristics in real-time. That increases the efficiency and naturalness of the collaboration with a higher perceived collaboration, positive teammate traits, and human trust. We believe that such an extended human-adaptation is a key to the long-term use of cobots.

Intelligent Transportation Systems (ITSs) are expected to have a profound impact on the quality of experience in future smart cities. Anomaly detection is an imperative for urban ITS applications to alleviate vulnerabilities that may cause accidents and fatal causalities. Previously proposed anomaly detection methods mostly require prior knowledge and domain specific training and/or optimization procedures. Therefore, in this work, we propose Competitive Learning based Anomaly Detection (CLAD) as a generic end-to-end approach for unsupervised anomaly detection using Auto Regressive Integrated Moving Average (ARIMA) forecasting model, data imaging and Centroid Neural Networks (CentNNs). Utilizing multi-dimensional time-series data obtained from diverse sensory measurements in the DIGINET-PS smart city infrastructure of TU Berlin, we compare performance of CLAD with unsupervised competitive learning as well as deep learning based anomaly detection techniques. Experimental results show that proposed approach results in higher detection accuracy and precision compared to other methods when multiple degrees of anomalies are considered.

Abstract This paper proposes a new family of algorithms for the online optimisation of composite objectives. The algorithms can be interpreted as the combination of the exponentiated gradient and p-norm algorithm. Combined with algorithmic ideas of adaptivity and optimism, the proposed algorithms achieve a sequence-dependent regret upper bound, matching the best-known bounds for sparse target decision variables. Furthermore, the algorithms have efficient implementations for popular composite objectives and constraints and can be converted to stochastic optimisation algorithms with the optimal accelerated rate for smooth objectives.

Radio Access Network (RAN) slicing is getting increasing attention as a resource allocation technique for satisfying diverse Quality-of-Service (QoS) requirements in 5G vehicular networks. Hierarchical Reinforcement Learning (HRL), such as hierarchical-DQN (h-DQN), is a promising slice management approach that decomposes performance constraints into a subroutine hierarchy and uses Deep Reinforcement Learning (DRL) at different temporal scales for online-learning an optimal policy of bandwidth allocation. In this paper, we tackle RAN slicing problem in 5G vehicle-to-everything (V2X) communications and present h-DQN based Soft Slicing (HSS) method for model-free opportunistic slice management. HSS consists of a multi-controller learning framework where a high-level meta-controller takes state input for determining a subgoal and a low-level controller decides on the action based on the given subgoal and the state. We compare performance of HSS with model-free and model-based Reinforcement Learning (RL) methods in terms of Age of Information (AoI), service delay and network throughput. Our results show that proposed scheme improves sample-efficiency and outperforms traditional and RL-based V2X RAN slice management methods in terms of network utility maximization.

We present a ring-shaped interaction device, called PeriSense, utilizing capacitive sensing in order to enable finger tracking. The finger angles and its adjacent fingers are sensed by measuring capacitive proximity between electrodes and human skin. To map the capacitive measurements to the finger angles, we use long short-term memory (LSTM). By wearing the ring on the middle finger, the angles of the index, middle and ring finger can be determined from the capacitive measurements. We collected sample data from 17 users using a Leap Motion camera, which provided the reference data. In a leave-one-user-out cross-validation test, we revealed a mean absolute error of 13.02 degrees over all finger angles. The motion of the little finger and the thumb are out of the capacitive measurement range or covered by the index and ring finger, respectively. With natural finger movements, the LSTM estimates the angles of the little finger and thumb based on the movement pattern of the other fingers.

Recommender systems have strongly attracted the attention of the machine learning research community with prosperous real-life deployments in the last few decades. The performance and success of most applications developed in this domain highly depend on an elaborate selection of models and configuration of their hyperparameters. The international MoST-Rec 2019 workshop addresses the issues of algorithm selection and parameter tuning for recommender systems. The workshop aims to bring together researchers from the model selection and hyperparameter tuning community in the general scope of machine learning with researchers from the recommender systems community for discussing and exchanging recent advances and open challenges in the field.

Task-level decision-making and AI planning are used to control autonomous robots from a high-level, mission-oriented perspective. The dynamic selection of most suitable actions allows the system to adapt to changes in the environment as well as its own state. Nevertheless, decision-making and AI planning often require a priori definitions of capabilities, rules, decision models, or world knowledge. Due to the challenge of handling the uncertainty of robot applications in dynamic and uncontrolled environments such definitions or descriptions are always incomplete, hence the possible adaptation capabilities are limited. In this paper, we present how the self-adaptation of a robot planning and decision-making system can be improved by incorporating reinforcement learning. Particularly, we show our approach of integrating deep reinforcement learning into the ROS Hybrid Behaviour Planner (RHBP).

We propose an architecture as a robot»s decision-making mechanism to anticipate a human»s state of mind, and so plan accordingly during a human-robot collaboration task. At the core of the architecture lies a novel stochastic decision-making mechanism that implements a partially observable Markov decision process anticipating a human»s state of mind in two-stages. In the first stage it anticipates the human»s task related availability, intent (motivation), and capability during the collaboration. In the second, it further reasons about these states to anticipate the human»s true need for help. Our contribution lies in the ability of our model to handle these unexpected conditions: 1) when the human»s intention is estimated to be irrelevant to the assigned task and may be unknown to the robot, e.g., motivation is lost, another assignment is received, onset of tiredness, and 2) when the human»s intention is relevant but the human doesn»t want the robot»s assistance in the given context, e.g., because of the human»s changing emotional states or the human»s task-relevant distrust for the robot. Our results show that integrating this model into a robot»s decision-making process increases the efficiency and naturalness of the collaboration.

News organizations employ personalized recommenders to target news articles to specific readers and thus foster engagement. Existing approaches rely on extensive user profiles. However frequently possible, readers rarely authenticate themselves on news publishers' websites. This paper proposes an approach for such cases. It provides a basic degree of personalization while complying with the key characteristics of news recommendation including news popularity, recency, and the dynamics of reading behavior. We extend existing research on the dynamics of news reading behavior by focusing both on the progress of reading interests over time and their relations. Reading interests are considered in three levels: short-, medium-, and long-term. Combinations of these are evaluated in terms of added value to the recommendation's performance and ensured news variety. Experiments with 17-month worth of logs from a German news publisher show that most frequent relations between news reading interests are constant in time but their probabilities change. Recommendations based on combined short-term and long-term interests result in increased accuracy while recommendations based on combined short-term and medium-term interests yield higher news variety.

Online news readers exhibit a very dynamic behavior. News publishers have been investigating ways to predict such changes in order to adjust their recommendation strategies and better engage the readers. Existing research focuses on analyzing the evolution of reading interests associated with news categories. Compared to these, we study also how relations among news interests change in time. Observations over a 10-month period on a German news publisher indicate that overall, the relations amid news categories change, but stable periods spanning months are also found. The reasons of these changes and how news publishers could integrate this knowledge in their solutions are subject to further investigation.

In our fast changing world, data streams move into the focus. In this paper, we study recommender systems for news portals. Compared with traditional recommender scenarios based on static data sets, the short life cycle of news items and the dynamics in users' preferences are major challenges when developing news recommender systems. This motivates us to research methods facilitating the inclusion of context and trends into news recommender systems. We explain specific requirements for news recommender system and discuss approaches incorporating trends and temporal user habits in order to improve news recommender system. A detailed data analysis motivates our approach. In addition, we discuss experiences of applying news recommendation algorithms online. The evaluation shows that approaches come with specific strengths and weaknesses. Consequently, publishers should select the recommendation strategy with the specific requirements in mind.

Information and communication technologies (ICTs) have a profound impact on the current state and envisioned future of automobiles. This paper presents an overview of research on ICT-based support and assistance services for the safety of future connected vehicles. A general classification and a brief description of the focus areas for research and development in this direction are given under the titles of vehicle detection, road detection, lane detection, pedestrian detection, drowsiness detection, and collision avoidance. Following an overview and taxonomy of the reviewed research articles, a categorized literature survey of safety critical applications is presented in detail. Future research directions are also highlighted.

In times of smart city and internet of things and services a lot of data is produced. However, there is no benefit in collecting the data without processing it. Smart services are one possibility to enable data access for data processing. Smart services have attracted research along their domain and requirements, benefits for the common as well as possible business models are developed. This work addresses the way how service consumers and service operators conduct business in an open B2B service marketplace. The paper presents and discusses phases of a business relationship in an digital service environment as well as discusses the business actions’ sequence. A role-action framework for service provisioning systems is developed. It contributes to a better understanding of service provision systems and demonstrates of what processes it constitutes. It furthermore presents what needs to be done by the systems’ participants to offer or consume services.

The paper presents an approach for individual and outsourced service clearing within Business-to-Business marketplaces for mobility services in a Software as a Service fashion. The lack of service clearing possibilities within today’s marketplaces solutions for mobility services have been confirmed by interviews experts who highlight the need for clearing. To enable service clearing, appropriate interfaces which enable access upon data are required. Current solutions lack those interfaces and thus service clearing to charge service transactions is not possible. The paper discusses interfaces for outsourced service clearing according to their design, data and role dependencies. A reference solution is implemented to demonstrate the feasibility of the interfaces and the overall clearing approach. A clearing algorithm has been developed to validate the interfaces’ reliability and correctness. Our presented clearing approach enables marketplace participants to outsource the transaction clearing to a provider which offers clearing capabilities. The work on hand contributes to interface and protocol standardization in respect to service clearing and marketplace interconnectivity.

With the move to ubiquitous computing, user interfaces (UI) are no longer bound to specific devices. While this problem can be tackled using the model-based UI development (MBUID) process, the usability of the device-specific interfaces is still an open question. We are presenting a combined system that integrates MBUID with a cognitive modeling framework in order to provide usability predictions at development time. Because of their potential impact, our focus within usability problems lies on user errors. These are captured in a cognitive model that capitalizes on meta-information provided by the MBUID system such as the abstract role of a UI element within a task sequence (e.g., input, output, command). The free parameters of the cognitive model were constrained using data from two previous studies. A validation experiment featuring a new application and UI yielded an unexpected error pattern that was nonetheless consistent with the model predictions.

It is observed that, due to the raising number of cars on the road, more and more cities suffer gridlocks. Mobility services, alike parking services, support drivers with real time mobility information to avoid those gridlocks. However, due to the the fragmented and isolated service operation, they are often not universally discoverable and usable. In this paper we elaborate an architecture for an ecosystem of interconnected service platforms to achieve that parking services can be universally discovered and used to mitigate parking related traffic problems. We present step by step how closed service platforms mature towards open service platforms by joining a centrally managed ecosystem. Respective enterprise integration patterns along with necessary data model are presented and discussed to justify the architecture. The achieved cross-system information exchange enables mobility providers to develop more sophisticated parking solutions in which they incorporate parking data and services from distributed parking operators. We demonstrate how our ecosystem increases access on parking information and how this mitigates the frustrating situation for drivers who are looking for a free parking lot.

The paper presents a Software as a Service approach for individual and outsourced service clearing within Business-to-Business marketplaces for mobility services. Interviewed experts confirm the lack of clearing possibilities within today's marketplaces solutions for mobility services and emphasize the need for service clearing. However, current solutions have no interfaces in place to enable access upon data which is required to perform service clearing and to charge service transactions. The paper discusses interfaces for outsourced service clearing according to their design, data and role dependencies. A reference implementation is developed which enable individual and outsourced service clearing. An algorithm for service clearing has been developed to check the interfaces' reliability and data correctness. Individual service clearing enables marketplace participants to appoint a clearing provider of whom they think fits their clearing requirements best. The introduced and implemented interfaces contribute to protocol standardization within the mobility service domain.

We present a browser-based visualization tool that allows users to explore movies and online videos based on the violence level of these videos. The system offers visualizations of annotations and results of the MediaEval 2012 Affect Task and can interactively download and analyze content from video hosting sites like YouTube.

Collaborative filtering recommender systems often use nearest neighbor methods to identify candidate items. In this paper we present an inverted neighborhood model, k-Furthest Neighbors, to identify less ordinary neighborhoods for the purpose of creating more diverse recommendations. The approach is evaluated two-fold, once in a traditional information retrieval evaluation setting where the model is trained and validated on a split train/test set, and once through an online user study (N=132) to identify users' perceived quality of the recommender. A standard k-nearest neighbor recommender is used as a baseline in both evaluation settings. Our evaluation shows that even though the proposed furthest neighbor model is outperformed in the traditional evaluation setting, the perceived usefulness of the algorithm shows no significant difference in the results of the user study.

Detecting violent scenes in movies is an important video content understanding functionality e.g., for providing automated youth protection services. One key issue in designing algorithms for violence detection is the choice of discriminative features. In this paper, we employ mid-level audio features and compare their discriminative power against low-level audio and visual features. We fuse these mid-level audio cues with low-level visual ones at the decision level in order to further improve the performance of violence detection. We use Mel-Frequency Cepstral Coefficients (MFCC) as audio and average motion as visual features. In order to learn a violence model, we choose two-class support vector machines (SVMs). Our experimental results on detecting violent video shots in Hollywood movies show that mid-level audio features are more discriminative and provide more precise results than low-level ones. The detection performance is further enhanced by fusing the mid-level audio cues with low-level visual ones using an SVM-based decision fusion.

Studying the text messages of a user such as his posts in Facebook or his tweets in Twitter can help in detecting his topics of interests. User in Social Network Systems (SNS) posts text messages about a wide diverse of topics. Posts usually written in a non-standard language, which make it not applicable to the standard Natural Language Processing (NLP) techniques used to catch the relations between words in text. In many cases there are semantic relations between the contained entities of posts that can infer the interest of the user. Bag-Of-Words (BOW) based text classification techniques classify this kind of messages to a wide diverse of topics, but they fail in catching the implicit semantic relation between the contained entities. In this paper we propose a technique to discover the implicit semantic relations between entities in text messages, which can infer the interests of a user. The proposed technique based on a semantically enriched graph representation of entities contained in text messages generated by a user, a new algorithm (Root-Path-Degree) is invented and used to find the most representative sub-graph that reflects the semantic implicit interests of the user. An evaluation was done using manually annotated posts of 687 Facebook users. Precision and Recall results showed our technique performs better than the standard BOW technique.

User Centric Networks (UCN) empowers the endusers as providers rather than just consumers of the network. Cooperation incentives based on reputation and crediting mechanisms are certain core enabling factors of UCN among which ensuring a certain level of QoS plays a great role for the maintenance of the UCN environment. In this paper we propose a QoS-based crediting mechanism, providing incentives for access points (APs) to share the load, i.e. the clients, in a fair way, thereby ensuring a homogenous level of QoS throughout the UCN community. We discuss a utility-based decision making framework for the incentivized cooperation based load balancing mechanism in UCN environment.

Collaborative Filtering Recommender Systems come in a wide variety of variants. In this paper we present a system for visualizing and comparing recommendations provided by different collaborative recommendation algorithms. The system utilizes a set of context-aware, hybrid, and other collaborative filtering solutions in order to generate various recommendations from which its users can pick those corresponding best to their current situation (i.e. context). All user interaction is fed back to the system in order to additionally improve the quality of the recommendations. Additionally, users can explicitly ask the system to treat certain recommenders as more important than others, or disregard them completely if the list of recommended movies is not to their liking.

Human action recognition requires the description of complex motion patterns in image sequences. In general, these patterns span varying temporal scales. In this context, Lagrangian methods have proven to be valuable for crowd analysis tasks such as crowd segmentation. In this paper, we show that, besides their potential in describing large scale motion patterns, Lagrangian methods are also well suited to model complex individual human activities over variable time intervals. We use Finite Time Lyapunov Exponents and time-normalized arc length measures in a linear SVM classification scheme. We evaluated our method on the Weizmann and KTH datasets. The results demonstrate that our approach is promising and that human action recognition performance is improved by fusing Lagrangian measures.

Recommender systems are commonly evaluated by trying to predict known, withheld, ratings for a set of users. Measures such as the Root-Mean-Square Error are used to estimate the quality of the recommender algorithms. This process does however not acknowledge the inherent rating inconsistencies of users. In this paper we present the first results from a noise measurement user study for estimating the magic barrier of recommender systems conducted on a commercial movie recommendation community. The magic barrier is the expected squared error of the optimal recommendation algorithm, or, the lowest error we can expect from any recommendation algorithm. Our results show that the barrier can be estimated by collecting the opinions of users on already rated items.

Internet services are important part of daily activities for most of us. These services come with sophisticated authentication requirements which may not be handled by average Internet users. The management of secure passwords for example creates an extra overhead which is often neglected due to usability reasons. Furthermore, password-based approaches are applicable only for initial logins and do not protect against unlocked workstation attacks. In this paper, we provide a non-intrusive identity verification scheme based on behavior biometrics where keystroke dynamics based-on free-text is used continuously for verifying the identity of a user in real-time. We improved existing keystroke dynamics based verification schemes in four aspects. First, we improve the scalability where we use a constant number of users instead of whole user space to verify the identity of target user. Second, we provide an adaptive user model which enables our solution to take the change of user behavior into consideration in verification decision. Next, we identify a new distance measure which enables us to verify identity of a user with shorter text. Fourth, we decrease the number of false results. Our solution is evaluated on a data set which we have collected from users while they were interacting with their mail-boxes during their daily activities.

Intelligent Environments have the vision of enhancing our everyday environment and interaction with its objects by sensing, computing, and communication capabilities. The major characteristics of such environments are the increasing number of embedded intelligent devices (ubiquity) into the background (transparency). These devices are expected to disappear or blend into the background and will be invisible to the user. However, because of this transparency, users fail to develop an adequate mental model for interaction with such environments. The Ubiquitous Meta User Interfaces (Ubi-MUI) ACM workshop provides a venue for the development of highly intuitive, multimedia supported meta user interfaces that bring transparency, predictability, and control into intelligent environments.

Future homes as smart and interactive environments will include a large and constantly growing number of networked devices that communicate with each other to coordinate activities. In this case the main challenge for developers and designers will be to offer possibilities to simply control this growing network of independent devices for different purposes, embodying complex technology, as well as managing a large number of information flows. In this paper we describe an approach of how to control such environments through the multimodal user interface of a home operating system.

The definition of different variants of a user interface at design time does not seem to be flexible enough to address future needs. Context sensitivity, user preferences and unforeseen situations require new means for automatic adaptation at runtime and mechanisms for user interventions within these adaptations. In our work we aim at the utilization of models at runtime to create user interfaces that are aware of their state and the meaning behind. An important part of this approach is a layout model that constrains and eventually defines the presentation of the user interface. The model contains statements about the orientation and size of the user interface elements that generate a constraint system at runtime to calculate the layout of the interface. Based on this model we provide mechanisms for the automatic calculation of user interface layouts as well as (meta) user interface that allows the user to intervene and adjust the layout to personal preferences.

Increasing use of computerized systems in our daily lives creates new adversarial opportunities for which complex mechanisms are exploited to mend the rapid development of new attacks. Behavioral Biometrics appear as one of the promising response to these attacks. But it is a relatively new research area, specific frameworks for evaluation and development of behavioral biometrics solutions could not be found yet. In this paper we present a conception of a generic framework and runtime environment which will enable researchers to develop, evaluate and compare their behavioral biometrics solutions with repeatable experiments under the same conditions with the same data.

In this paper, we introduce a new scheme for performing IT Risk Management within organizational domains. It adopts a business process-oriented view which integrates risk assessment, vulnerability assessment and risk mitigation into a quantitative framework. Taking the asset dependencies into account, we map business process values to IT hardware components in a hierarchical fashion and combine it with IT system vulnerability and threat analysis to derive risk scores on the IT hardware system level. We then apply discrete-time algorithms for computing cost-optimal quantitative mitigation strategies given a set of available mitigation actions. We illustrate the entire integrated process by means of a case study and show that considerable risk reduction can be achieved.

We study the application of spectral clustering, prediction and visualization methods to graphs with negatively weighted edges. We show that several characteristic matrices of graphs can be extended to graphs with positively and negatively weighted edges, giving signed spectral clustering methods, signed graph kernels and network visualization methods that apply to signed graphs. In particular, we review a signed variant of the graph Laplacian. We derive our results by considering random walks, graph clustering, graph drawing and electrical networks, showing that they all result in the same formalism for handling negatively weighted edges. We illustrate our methods using examples from social networks with negative edges and bipartite rating graphs.

Collaborative tagging services (folksonomies) have been among the stars of theWeb 2.0 era. They allow their users to label diverse resources with freely chosen keywords (tags). Our studies of two real-world folksonomies unveil that individual users develop highly personalized vocabularies of tags. While these meet individual needs and preferences, the considerable differences between personal tag vocabularies (personomies) impede services such as social search or customized tag recommendation. In this paper, we introduce a novel user-centric tag model that allows us to derive mappings between personal tag vocabularies and the corresponding folksonomies. Using these mappings, we can infer the meaning of user-assigned tags and can predict choices of tags a user may want to assign to new items. Furthermore, our translational approach helps in reducing common problems related to tag ambiguity, synonymous tags, or multilingualism. We evaluate the applicability of our method in tag recommendation and tag-based social search. Extensive experiments show that our translational model improves the prediction accuracy in both scenarios.

Smartphones are getting increasingly popular and several malwares appeared targeting these devices. General countermeasures to smartphone malwares are currently limited to signature-based antivirus scanners which efficiently detect known malwares, but they have serious shortcomings with new and unknown malwares creating a window of opportunity for attackers. As smartphones become host for sensitive data and applications, extended malware detection mechanisms are necessary complying with the corresponding resource constraints. The contribution of this paper is twofold. First, we perform static analysis on the executables to extract their function calls in Android environment using the command readelf. Function call lists are compared with malware executables for classifying them with PART, Prism and Nearest Neighbor Algorithms. Second, we present a collaborative malware detection approach to extend these results. Corresponding simulation results are presented.

In document image understanding, public datasets with ground-truth are an important part of scientific work. They are not only helpful for developing new methods, but also provide a way of comparing performance. Generating these datasets, however, is time consuming and cost-intensive work, requiring a lot of manual effort. In this paper we both propose a way to semi-automatically generate ground-truthed datasets for newspapers and provide a comprehensive dataset. The focus of this paper is layout analysis ground truth. The proposed two step approach consists of a module which automatically creates layouts and an image matching module which allows to map the ground truth information from the synthetic layout to the scanned version. In the first step, layouts are generated automatically from a news corpus. The output consists of a digital newspaper (PDF file) and an XML file containing geometric and logical layout information. In the second step, the PDF files are printed, scanned and aligned with the synthetic image obtained by rendering the PDF. Finally, the geometric and logical layout ground truth is mapped onto the scanned image.

Smart environments bring together multiple users, (interaction) resources and services. This creates complex and unpredictable interactive computing environments that are hard to understand. Users thus have difficulties to build up their mental model of such interactive systems. To address this issue users need possibilities to evaluate the state of these systems and to adapt them according to their needs. In this work we present our implementation of the functionalities to evaluate and control multimodal interaction in smart environments, which is accessible for users through a meta user interface.

In this paper we demonstrate how to monitor a smartphone running Symbian operating system and Windows Mobile in order to extract features for anomaly detection. These features are sent to a remote server because running a complex intrusion detection system on this kind of mobile device still is not feasible due to capability and hardware limitations. We give examples on how to compute relevant features and introduce the top ten applications used by mobile phone users based on a study in 2005. The usage of these applications is recorded by a monitoring client and visualized. Additionally, monitoring results of public and self-written malwares are shown. For improving monitoring client performance, Principal Component Analysis was applied which lead to a decrease of about 80% of the amount of monitored features.

User interface modeling is a well accepted approach to handle increasing user interface complexity. The approach presented in this paper utilizes user interface models at runtime to provide a basis for user interface distribution and synchronization. Task and domain model synchronize workflow and dynamic content across devices and modalities. A cooking assistant serves as example application to demonstrate multimodality and distribution. Additionally a debugger allows the inspection of the underlying user interface models at runtime.

Offering user interfaces for interactive applications that are flexible enough to be adapted to various context-of-use scenarios such as supporting different display sizes or addressing various input styles requires an adaptive layout. We describe an approach for layout derivation that is embedded in a model-based user interface generation process. By an interactive and tool-supported process we can efficiently create a layout model that is composed of interpretations of the other design models and is consistent to the application design. By shifting the decision about which interpretations are relevant to support a specific context-of-use scenario from design-time to run-time, we can flexibly adapt the layout to consider new device capabilities, user demands and user interface distributions. We present our run-time environment that is able to evaluate the relevant model layout information to constraints as they are required and to reassemble the user interface parts regarding the updated containment, order, orientation and sizes information of the layout-model. Finally we present results of an evaluation we performed to test the design and run-time efficiency of our model-based layouting approach.

We extend the resistance distance kernel to the domain of signed dissimilarity values, and show how it can be applied to collaborative rating prediction. The resistance distance is a graph kernel inspired by electrical network models where edges of a graph are interpreted as electrical resistances. We model the similarity between users of a large collaborative rating database using this signed resistance distance, generalizing the previously known regular resistance distance kernel which is limited to nonnegative values. We show that the signed resistance distance kernel can be computed effectively using the Moore-Penrose pseudoinverse of the Laplacian matrix of the bipartite rating graph, leading to fast computation based on the eigenvalue decomposition of the Laplacian matrix. We apply this technique to collaborative rating prediction on the Netflix Prize corpus, and show how our new kernel can replace the traditional Pearson correlation for rating prediction.

With rapid developments in the community mesh networks and wireless sensor networks research, the need for more efficient channel access techniques for multi-hop wireless networks has become eminent. In this work, we propose a novel hybrid channel access scheme that spatially limits the contention in the network such that 2-hop neighbors access the channel contention-free among each other whereas only the immediate neighbors may contend among each other. The contention among neighbors can be handled much more efficiently by a basic CSMA protocol as if operating in a single-hop network. We provide a general framework for the SLICON scheme and compare its performance to the conventional RTS/CTS-based collision avoidance scheme. By a case study using the IEEE 802.11 protocol as the underlying CSMA protocol, the proposed scheme pleads itself as a more efficient alternative to the RTS/CTS based collision avoidance scheme for large and dense multi-hop ad hoc networks with stationary nodes, such as wireless mesh and sensor networks.

The proliferation of wireless access technologies and the growth of demand for mobile services pose the need for seamless mobility across heterogeneous networks. Moreover, the widespread use of the TCP/IP protocol stack clearly reinforces the need to enhance its performance in such scenarios. In this paper we present an experimental evaluation of the influence of the so called heterogeneous (vertical) mobility on the behavior TCP protocol, and we propose a cross-layer solution to reduce this impact in future mobile environments. The described solution is based on the dynamic adaptation of the window size for a TCP connection, based on the related mobility conditions. We have evaluated this solution using the BIB3R testbed, which emulates the inter-working between commercial wireless technologies (e.g., UMTS, Wi-Fi, WiMAX, etc). The results showed that by adjusting the window size before the vertical handover process, TCP connections can reduce the adaptation time between 30 and 40 percent depending on the mobility scenario (heterogeneous links).

An open access network (OAN) is characterized by the fact that private WLANs and broadband access lines are made available for public use, enabling bypassing mobile users to profit from continuous coverage. Our insight is that in wireless OANs, the overhead required for the launching or reconfiguration of a real-time session can be drastically reduced. We propose an algorithm which performs parameter injection into a multimedia data stream instead of following the classical negotiation path typically used. A solution to change the session characteristics of an application that adapts to varying network conditions was developed, including a QoS solution with SIP and MIP to optimize the session characteristics by taking into account the changed conditions in the network. With our approach, we achieve a reduction of a factor of 2.5 for session signaling delay, and excellent jitter values in the range of 10-15 milliseconds as our implementation results show.

The widespread deployment of the IEEE 802.11 protocol has made it the de facto standard for wireless network access and allows Internet users to move freely (at least within a hotspot coverage). The convenience brought about by the IEEE 802.11 protocol is also accompanied by new technical challenges such as poor performance of the widely used transport protocol TCP due to the limited and varying bandwidth resources of the wireless medium. In this work, we seek to enhance efficiency of bandwidth usage for TCP over wireless links and improve TCP goodput. Our insight is that TCP acknowledgments can be spared on wireless links to save precious bandwidth resource of these links. We propose a simple technique that leverages the layer coordination between TCP and MAC to suppress TCP acknowledgments. Our technique is transparent to both TCP and MAC and does not require their modification or replacement.

We present a cooperative intrusion detection approach inspired by biological immune system principles and P2P communication techniques to develop a distributed anomaly detection scheme. We utilize dynamic collaboration between individual artificial immune system (AIS) agents to address the well-known false positive problem in anomaly detection. The AIS agents use a set of detectors obtained through negative selection during a training phase and exchange status information and detectors on a periodical and event-driven basis, respectively. This cooperation scheme follows peer-to-peer communication principles in order to avoid a single point of failure and increase the robustness of the system. We illustrate our approach by means of two specific example scenarios in a novel network security simulator.

The Internet, mobile telephony and portable computers let today's user meet novel applications provided by the service developers, who are motivated by the popular customer trends in technology. However, the supply-demand chain observed in telecommunications causes the complexity of both the user devices as well as the network to increase. The complexity is caused by the need to support quality of service, security and mobility for the new services throughout the heterogeneous communication network. These problems slow down the application development process significantly and cause the user demand not to be met in time. We believe that a new look at the network is necessary in order to address these issues. This article presents this new look, which we term network economy. It is a service-centric approach to next generation mobile networks, whereby the network is redefined as a collection of diverse resources. The aim of network economy is to optimise the allocation of these resources such that network efficiency is maximised. In order to achieve this objective, the intelligence level of the network is increased by decision engines, which continuously collect information from the network, classify and store it in a repository to be queried by their peers. The approach is accompanied by context-aware, self-organising user devices. Network economy provides a network-wide infrastructure to benefit users, service developers as well as network operators.

This paper presents an expert peering system for information exchange in the knowledge society. Our system realizes an intelligent, real-time search engine for enterprise Intranets or online communities that automatically relays user queries to knowledgable specialists. According to its very nature, the system requires a sound integration of concepts drawn from various areas of Computer Science. In addition to our solutions to problems in scalable processing, data transfer, and networking, we also address issues of interface design and usability, as well as aspects of machine intelligence. Results obtained from extensive experiments demonstrate the efficiency and robustness of our system and convey its potential for next generation web services.

Some of the established approaches to evaluating text clustering algorithms for information retrieval show theoretical flaws. In this paper, we analyze these flaws and introduce a new evaluation measure to overcome them. Based on a simple yet rigorous mathematical analysis of the effect of certain parameters in cluster based retrieval, we show that certain conclusions drawn in the recent literature must be taken with a grain of salt. Our new measure, in contrast, accounts for statistical biases that have to be expected according to our analysis. A series of experiments and a comparison with results reported recently underlines that this measure is a more suitable performance indicator that allows for more meaningful interpretations.

The Internet is a vast resource of information. Unfortunately, finding and accessing this information is often a very cumbersome task even with existing information platforms. Searching on the WWW suffers from the fact that almost every word is ambiguous to a certain degree in the information-rich environment of the Internet. Clustering search results is a way to solve this problem. This paper demonstrates how to employ novel Information Retrieval measures to derive optimal parametrizations for a cluster algorithm.

Feedback on past queries is a valuable resource for improving retrieval performance on new queries. We introduce a modular approach to incorporating feedback information into given retrieval architectures. We propose to fusion the original ranking with those returned by rerankers, each of which trained on feedback given for a distinct, single query. Here, we examine the basic case of improving a query's original ranking qtest by only using one reranker: the one trained on feedback on the "closest" query qtrain. We examine the use of various distance measures between queries to first identify qtrain and then determine the best linear combination of the original and the reranker's ratings, that is: to find out which feedback to learn from, and how strongly to use it. We show the cosine distance between the term vectors of the two queries, each enriched by representations of the top N originally returned documents, to reliably answer both questions. The fusion performs equally well or better than a) always using only the original ranker or the reranker, b) selecting a hard distance threshold to decide between the two, or c) fusioning results with a ratio that is globally optimized, but fixed across all tested queries.

We propose a hierarchical approach to document categorization that requires no pre-configuration and maps the semantic document space to a predefined taxonomy. The utilization of search engines to train a hierarchical classifier makes our approach more flexible than existing solutions which rely on (human) labeled data and are bound to a specific domain. We show that the structural information given by the taxonomy allows for a context aware construction of search queries and leads to higher tagging accuracy. We test our approach on different benchmark datasets and evaluate its performance on the single- and multi-tag assignment tasks. The experimental results show that our solution is as accurate as supervised classifiers for web page classification and still performs well when categorizing domain specific documents.

Factory of the future is emerging with the existence of new modeling and application tools that can both simulate and manage the whole production process in an autonomous, intelligent and interactive manner. Holonic modeling and its software correspondence agent oriented technology provides us with these tools. Especially the use of learning algorithms trying to optimize the behaviors of software agents within a dynamic environment is the key factor in reaching the required properties. In this paper, we use the well known Q learning algorithm of reinforcement learning (RL) in evaluating production orders within a supply chain management (SCM) framework and making decisions with respect to these evaluations. We introduce our SCM model and show that RL performs better than traditional tools for dynamic problem solving in daily business. We also show cases where RL fails to perform efficiently.