Stathes Hadjiefthymiades

Facing the cold start problem in recommender systems

A recommender system (RS) aims to provide personalized recommendations to users for specific items (e.g., music, books). Popular techniques involve content-based (CB) models and collaborative filtering (CF) approaches. In this paper, we deal with a very important problem in RSs: The cold start problem. This problem is related to recommendations for novel users or new items. In case of new users, the system does not have information about their preferences in order to make recommendations. We propose a model where widely known classification algorithms in combination with similarity techniques and prediction mechanisms provide the necessary means for retrieving recommendations. The proposed approach incorporates classification methods in a pure CF system while the use of demographic data help for the identification of other users with similar behavior. Our experiments show the performance of the proposed system through a large number of experiments. We adopt the widely known dataset provided by the GroupLens research group. We reveal the advantages of the proposed solution by providing satisfactory numerical results in different experimental scenarios.

ESW4: enhanced scheme for WWW computing in wireless communication environments

Mobile computing is considered of major importance to the computing industry for the forthcoming years due to the progress in the wireless communications domain. In this paper, we present a proxy-based architecture, called ESW4, which manages to accelerate Web browsing in wireless CPNs. Proxy caches, maintained in base stations, are constantly relocated to accompany the roaming user. We discuss a cache management scheme involving the relocation of full caches to the most candidate cells but also percentages of the cache to less likely neighbors. Relocation is performed according to the output of a movement prediction algorithm based on a learning automaton. The simulation of ESW4 shows substantial benefits for the end user.

Context Awareness in Mobile Computing Environments

In this article, we report software architectures for context awareness in mobile computing environments, sensor centric systems and discuss context modeling issues. Defining an architecture for supporting context-aware applications for mobile devices explicitly implies a scalable description of how to represent contextual information and which are the abstraction models capable of handling such information. Using sensors to retrieve contextual information (e.g., user location) leads to a sensor network scheme that provides services to the applications level. Operations for capturing, collating, storing, and disseminating contextual information at the lowest level and aggregating it into increasingly more abstract models qualify the context-aware systems. In this article, we introduce context aware systems in mobile computing environments, review the basic mechanisms underlying the operation of such systems, and discuss notable work and important architectures in the area.

Scenarios of using web services in M-commerce

The web service paradigm is a promising technology for developing applications in open, distributed and heterogeneous environments. The proliferation of this new technology has coincided with significant advances in the hardware and software capabilities of mobile devices. Due to the great benefits that come with the web service technology, such as interoperability, dynamic service discovery and reusability, there is a strong interest in making mobile devices capable of providing and consuming web services over wireless networks. This paper describes several scenarios of using web services in mobile devices and identifies their advantages, issues and challenges.

Using proxy cache relocation to accelerate Web browsing in wireless/mobile communications

Mobile computing is considered of major importance to the computing industry for the forthcoming years due to the progress in the wireless communications area. A proxy-based architecture for accelerating Web browsing in cellular customer premises networks (CPN) is presented. Proxy caches, maintained in base stations, are constantly relocated to accompany the roaming user. A cache management scheme is proposed, which involves the relocation of full caches to the most probable cells but also percentages of the caches to less likely neighbors. Relocation is performed according to a movement prediction algorithm based on a learning automaton. The simulation of the scheme demonstrates substantial benefits for the end user.

Situational computing: An innovative architecture with imprecise reasoning

Situation awareness is introduced as a more holistic variant of context awareness where situations are regarded as logically aggregated pieces of context. Situational computing can be viewed as the pervasive computing paradigm that deals with situational context representation and reasoning. One important problem that arises in such paradigm is the imperfect observations (e.g., sensor readings) that lead to the estimation of the current user situation. Hence, the knowledge upon which the context/situation aware paradigm is built is rather vague. To deal with this shortcoming, we propose the use of Fuzzy Logic theory with the purpose of determining (inferring) and reasoning about the current situation of the involved user. We elaborate on the architectural model that enables the system to assume actions autonomously according to previous user reactions and current situation. The captured, imperfect contextual information is matched against pre-developed situation ontologies in order to approximately infer the current user context. Finally, we present a series of experimental results that provide evidence on the flexible, efficient nature of the proposed situational computing.

EmerLoc: Location-based services for emergency medical incidents

BACKGROUND Recent developments in positioning systems and telecommunications have provided the technology needed for the development of location aware medical applications. We developed a system, named EmerLoc, which is based upon this technology and uses a set of sensors that are attached to the patients body, a micro-computing unit which is responsible for processing the sensor readings and a central monitoring unit, which coordinates the data flow. OBJECTIVE To demonstrate that the proposed system is technically feasible and acceptable for the potential users. METHOD Transmission speed is assessed mostly by means of transmission of DICOM compliant images in various operational scenarios. The positioning functionality was established both outdoor using GPS and indoor using the UCLA Nibble system. User acceptability was assessed in a hospital setting by 15 physicians who filled in a questionnaire after having used the system in an experimental setting. RESULTS Transmission speeds ranged from 88kB/s for a IEEE 802.11 infrastructure to 2.5kB/s for a GSM/GPRS scenario. Positioning accuracy based on GPS was 5-10m. The physicians rated the technical aspects on average above 3 on a 5-point scale. Only the data presentation was assessed to be not satisfactory (2.81 on a 5-point scale). CONCLUSION The reported results prove the feasibility of the proposed architecture and its alignment with widely established practices and standards, while the reaction of potential users who evaluated the system is quite positive.

Semantically enriched navigation for indoor environments

Location-based mobile services have been in use, and studied, for a long time. With the proliferation of wireless networking technologies, users are mostly interested in advanced services that render the surrounding environment (i.e., the building) highly intelligent and significantly facilitate their activities. In this paper our focus is on indoor navigation, one of the most important location services. Existing approaches for indoor navigation are driven by geometric information and neglect important aspects, such as the semantics of space and user capabilities and context. The derived applications are not intelligent enough to catalytically contribute to the pervasive computing vision. In this paper, a novel navigation mechanism is introduced. Such navigation scheme is enriched with user profiles and the adoption of an ontological framework. These enhancements introduce a series of technical challenges that are extensively discussed throughout the paper.

Enabling Location Privacy and Medical Data Encryption in Patient Telemonitoring Systems

Patient telemonitoring systems (PTS) deal with the acquisition, processing, and secure transmission of a patients physiological and physical parameters to a remote location, where expert medical knowledge is available. In emergency situations, when the patients life is threatened, the trend in modern PTS is to transmit the current location of the patient. Although research in communications security has led to mechanisms that sufficiently protect medical data, research related to location privacy area is still in its early stages. This paper proposes an architecture that enhances PTS through location privacy and data encryption. We study the most popular PTS technologies in conjunction with location privacy architectures and propose an innovative scheme that exploits a point-to-point protocol called Mist. We describe a prototype implementation, developed for validating the proposed framework along with the corresponding evaluation results.

Enhancing Situation-Aware Systems through Imprecise Reasoning

Context awareness is viewed as one of the most important aspects in the emerging pervasive computing paradigm. We focus our work on situation awareness; a more holistic variant of context awareness where situations are regarded as logically aggregated contexts. One important problem that arises in such systems is the imperfect observations (e.g., sensor readings) that lead to the estimation of the current context of the user. Hence, the knowledge upon which the context / situation aware paradigm is built is rather vague. To deal with this shortcoming, we propose the use of Fuzzy Logic theory with the purpose of determining (inferring) and reasoning about the current situation of the involved user. We elaborate on the architectural model that enables the system to assume actions autonomously according to previous user reactions and current situation. The captured, imperfect contextual information is matched against pre-developed ontologies in order to approximately infer the current situation of the user. Finally, we present a series of experimental results that provide evidence of the flexible, efficient nature of the proposed situation awareness architecture.