Skip to main content
main-content

Über dieses Buch

This book constitutes the refereed proceedings of the 17th International Conference on Innovations for Community Services, I4CS 2017, held in Darmstadt, Germany, in June 2017.

The 12 revised full papers presented together with two short papers were carefully reviewed and selected from 31 submissions. The papers are organized in topical sections on Social Networks; Cooperative Networks; Optimization Algorithms; Infrastructure Planning; Energy Management; Short Papers.

Inhaltsverzeichnis

Frontmatter

Data Analytics

Frontmatter

Extracting Wikipedia Data to Enrich Spatial Information

Abstract
Freely available geo data allow a developer to create new types of remarkable services related to the user’s location. Even though current geo data sources have a high coverage and quality, they do not contain all information required by new services. This is because geo data sources usually focus on object geometries and object types. Important information is often missing. As an example: city entries mainly contain the city name and border, but not the name of mayor, amount of taxes, year of foundation, number of districts etc. These data are available in online encyclopediae such as Wikipedia, but there is no obvious approach to relate both sources. Our objective was thus to create an automatic import from Wikipedia articles that describe geo objects and extract all relevant data. To extract processible values we are able to identify property types such dates, money values, powers, heights, sizes etc. This makes it possible to use these data for further computation, e.g. to search for maxima, build averages and sums or to create comparative conditions in queries.
Jörg Roth

Towards the Automatic Sentiment Analysis of German News and Forum Documents

Abstract
The fully automated sentiment analysis on large text collections is an important task in many applications scenarios. The sentiment analysis is a challenging task due to the domain-specific language style and the variety of sentiment indicators. The basis for learning powerful sentiment classifiers are annotated datasets, but for many domains and especially with non-English texts hardly any datasets exist. In order to support the development of sentiment classifiers, we have created two corpora: The first corpus is build based on German news articles. Although news articles should be objective, they often excite subjective emotions. The second corpus consists of annotated messages from a German telecommunication forum. In this paper we describe the process of creating the corpora and discuss our approach for tracing sentiment values, defining clear rules for assigning sentiments scores. Given the corpora we train classifiers that yields good classification results and establish valuable baselines for sentiment analysis. We compare the learned classification strategies and discuss how the approaches can be transferred to new scenarios.
Andreas Lommatzsch, Florian Bütow, Danuta Ploch, Sahin Albayrak

Concept of Observer to Detect Special Cases in a Multidimensional Dataset

Abstract
In a dataset, the special cases are data that appears to be inconsistent with the neighborhoods of data. The special cases are warnings that usually require specific processing in social networks, medical applications or complex processes. This paper proposes to use the observer’s paradigm to detect special cases in a multidimensional data space. Thus the observations allow to define the neighborhoods of data. Then we propose a rareness index for each data. The special cases have the highest values of rareness index. Experimental results show the ability of the method to detect these special cases. We conclude this paper with a brief discussion.
Michel Herbin, Didier Gillard, Laurent Hussenet

Cooperative Networks

Frontmatter

An Extended Tester for Cooperative Intelligent Transport Systems

Abstract
C-ITS (Cooperative Intelligent Transport System) equipments are usually validated using testing procedures. The ETSI organism provides a set of test suites to check the conformance of communication protocoles implemented on these equipments. An adapted interface (denoted Upper Tester) has to be developed on any equipment which requires to be tested.
In order to provide validated C-ITS equipments, the need to check the correctness of all functional parts becomes of high interest. One of the main functional part of a C-ITS equipment is to trigger automatically specific events when some conditions are satisfied. In order to test these event triggering conditions, we may emulate the environment (with its sensors) with an appropriate interface (denoted extended upper-tester).
Moreover, a C-ITS equipment is expected to achieve manual events, such as opening a salt container in case of recovering slippery roads. Its execution could trigger conditions to send some appropriate messages as slippery roads ahead. For this purpose, we designed an extended Upper Tester from the one introduced by ETSI. Its aim is to manage the manual and the automatic triggering conditions and to test the results of these events.
In this paper, we present this extended set of tests and their implementation in a real deployment project (SCOOP@F) of a complete C-ITS architecture.
Hacène Fouchal, Geoffrey Wilhelm, Emilien Bourdy, Marwane Ayaida

dSDiVN: A Distributed Software-Defined Networking Architecture for Infrastructure-Less Vehicular Networks

Abstract
In the last few years, the emerging network architecture paradigm of Software-Defined Networking (SDN), has become one of the most important technology to manage large scale networks such as Vehicular Ad-hoc Networks (VANETs). Recently, several works have shown interest in the use of SDN paradigm in VANETs. SDN brings flexibility, scalability and management facility to current VANETs. However, almost all of proposed Software-Defined VANET (SDVN) architectures are infrastructure-based. This paper will focus on how to enable SDN in infrastructure-less vehicular environments. For this aim, we propose a novel distributed SDN-based architecture for uncovered infrastructure-less vehicular scenarios. It is a scalable cluster-based architecture with distributed mobile controllers and a reliable fallback recovery mechanism based on self-organized clustering and failure anticipation.
Ahmed Alioua, Sidi-Mohammed Senouci, Samira Moussaoui

Optimization Algorithms

Frontmatter

On a Fog Computing Platform Built on ARM Architectures by Docker Container Technology

Abstract
Fog computing constitutes currently a challenging effort to establish the concepts and services of cloud computing at the edge of converging wireless networks and wired high-speed backbones. We discuss the concepts of our fog computing platform HCL-BaFog. It is built on top of Hypriot Cluster Lab (HCL) which has been developed by the Hypriot Pirate Crew in recent years based on single board computers with an ARM architecture. It uses LINUX container technology as underlying open source platform that has been established by means of the rapidly evolving framework Docker. We present the design principles of our fog computing platform and discuss its different software components. To create clusters of fog cells subject to high-availability requirements and to provide failsafe data processing, we further summarize some performance results on the integration of the orchestration tools Docker Swarm Mode and Kubernetes on HCL and draw some conclusions regarding their suitability for fog computing.
Andreas Eiermann, Mathias Renner, Marcel Großmann, Udo R. Krieger

Markov Random Walk vs. Higher-Order Factorization Machines: A Comparison of State-of-the-Art Recommender Algorithms

Abstract
World-wide research on recommender systems has resulted in great, highly effective algorithms based on a large variety of different concepts. Two of these promising recommender approaches are the Markov Random Walk and (higher-order) Factorization Machines. Unfortunately, due to the substantial effort for optimizing hyperparameters, most articles that describe new recommender approaches do not compare the obtained results with other state-of-the-art approaches in the recommender domain.
This paper demonstrates how different state-of-the-art recommender algorithms can be compared in a consistent manner. Furthermore, we investigate under which circumstances Factorization Machines should be preferred and in which situations Markov Random Walk is the most striking algorithm. In addition, we include the restart concept into a Markov Random Walk with an optimized walk length and show how the number of factors of each order in a higher-order Factorization Machine can be optimized.
Julian Knoll, David Köckritz, Rainer Groß

Finding the Intersection Points of Networks

Abstract
Two algorithms have been constructed that find the intersections between two sets of line segments in practical networks used for planning and routing, solving the implementation issues of existing algorithms. One of the algorithms is a generalisation of the Bentley-Ottmann-algorithm, relaxing the assumptions in the original algorithm, the other is a smart brute force algorithm. In this article the algorithms are elaborated and will be tested, using real data sets constructed from street networks and trench networks. Both algorithms find all the intersections but with a difference in calculation time.
Niels Neumann, Frank Phillipson

Infrastructure Planning

Frontmatter

Distributed Network Infrastructure for Community Services in the Framework of Next Generation Mobile Networks

Abstract
The concept of next generation networks as is currently consolidating within research and standards defining organizations foresees beside provision of higher flexibility and adaptability to different services’ requirements also increased resource efficiency to enable affordable access in a sustainable way. To this end, a truly universal access has to be provided integrating multiple wireline, wireless, and cellular technologies to support residential and mobile entities of different size/shape/capability sets as is reflected by the variety of typical 5G use cases. Multiple logically separate networks (slices) shall be operated across the same infrastructure offering a performance and user experience meeting the diverse demand as exact as possible. This could include, beside traditional commercial operator services, a type of best effort connectivity to access e.g. urban community information and support daily life within a future smart city environment. This paper addresses a framework concept to integrate such a usage scenario within a future converged 5G system. Main focus of the reported exemplary results is on the issues of flexible service support (including varying mobility requirements) and efficient use of resources which seem to be most relevant for 5G success from a commercial point of view.
Dirk von Hugo, Gerald Eichler

Constrained Wireless Network Planning

Abstract
In this paper we define the Constrained Wireless Network Planning problem. Given is an orientation of access points which, if supplied with network connectivity, is able to provide a required level of coverage to clients. The goal is to find a division of these access points in source locations and repeater locations such that each of the access points is provided with network connectivity, while not all need to be directly connected to an existing network. The origin of the constraints in this problem are threefold. First, there is a restriction on the allowed distance between a source and a repeater location. Second, there is a restriction on the number of repeaters which may be provided with network connectivity by a source. Third, a repeater location may not provide another location with network connectivity. In this paper we propose an Iterated Local Search procedure to solve this problem. We apply this procedure to a problem arising in the field of multi-service planning in Smart Cities.
Timotheus J. C. Vos, Frank Phillipson

Energy Management

Frontmatter

Multi Objective Approach for Tactical Capacity Management of Distributed Generation

Abstract
Stakeholders in an electricity system can have different objectives. For this reason, in this paper a model is presented that can handle uncertainty in demand and supply and can do multi-objective analysis to show the sensitivity of the capacity management to the different objectives. Four possible objectives are presented to be considered by the model: Self-sufficiency rate, Maximum Import, Overcapacity and Return on investment. A case study is presented to show the capabilities of the model and give some results and insight into a particular case study.
Frank Phillipson

DP and RL Approach Optimization for Embedded System Communications with Energy Harvesting

Abstract
In this paper, we consider a point-to-point wireless communication in embedded system. This system is supposed to be battery powered and equipped with an energy harvester. According to the battery level and the harvested energy, the transmitter has to make decision following an optimal policy in order to maximize its reward over the operating period. We first consider a prior stochastic knowledge of the transition matrix probabilities to find out the optimal policy using algorithm originated from DP methods. With no such stochastic knowledge, we will adopt algorithms from RL methods to find out the optimal policies. The resulting performances are then compared.
Mohammed Assaouy, Ouadoudi Zytoune, Driss Aboutajdine

Spatial Guidance [Short Papers]

Frontmatter

Smart Screen System for Smart Buildings Made of Tablets

Abstract
Smart screen system is a system that is used for displaying information about individual rooms, which are located in the building, in which the system is implemented. The system is built on the basis of using tablet as information and recording unit of each room.
For each room, respectively for each group of rooms, the Windows tablet is used as an information unit, which can be accessible remotely through the web server by any privileged user. The user can then dynamically enter various information, which are then displayed to all those interested. According to the type of room, different data can be displayed - e.g. current room schedule (meetings, lectures, telco, …), presence of persons (business trip, lunch, consultations, …), etc.
Michal Hodoň, Martin Húdik, Štefan Tóth, Michal Kochláň

A Realistic Location Service for VANETs

Abstract
Position-based routing also called geographic routing is considered a more promising routing approach for highly dynamic and mobile networks like Vehicular Ad-hoc Networks (VANETs). In this kind of networks, the high-speed mobility of vehicles causes rapid changes in vehicles density and limited-time communication links. Hence, the need of location service has become extremely important to be able to find the position of a target node in a very short time.
This paper proposes a realistic location service for unicast routing over VANETs in an urban environment. The proposed approach is able to find the path with higher connectivity in urban environment by exploiting information of each vehicle in the network to reach the destination. For this reason, we used a new metric called Link Connectivity (LC) in order to find the path with higher connectivity between the source vehicle and the destination vehicle.
Tawfiq Nebbou, Hacène Fouchal, Mohamed Lehsaini, Marwane Ayaida

Backmatter

Weitere Informationen

Premium Partner

    Bildnachweise