The Systemics
and Informatics World Network
SIWN Abstracts
Index
Integrated Attention for Cognitive Robotics
Raúl Arrabales, Agapito Ledezma and Araceli Sanchis
Communications of SIWN, Vol. 5, August 2008, pp. 1-5
Abstract: An autonomous robot is usually expected to perform its assigned task in complex and unstructured environments, where the high amount of sensory information acquired by sensors cannot be entirely processed in real-time. Like in living organisms, an attention mechanism is required in order to select most relevant stimuli from the environment. The outcome of this selection process comprises all the information that will be available to the control system of the robot. Therefore, an optimal operation of the attention mechanism is necessary for robot mission success. In this work, we present a novel cognitive architecture which permits the implementation and integration of efficient attention mechanisms for autonomous robots. Additionally, the application of this approach to the domain of unknown environment exploration is discussed and results are analyzed. Specifically, the influence of attention in the generation of efficient robot behavior is demonstrated.
Keywords: attention, cognitive modeling, intelligent agents.
sai: cosiwn.2008.08.012
An Agent for Personalized Interaction
Giovanni Cozzolongo and Francesca Muci
Communications of SIWN, Vol. 5, August 2008, pp. 6-10
Abstract: In this paper we describe a personal agent able to employ semantic web tecnologies in combination with user modeling techniques. The agent has the objective to collect information/services useful to different types of customer and to support users in smart environment by personalizing of information and provided services. To achieve this goal the personal agent cooperates with a set of agents able to gather, process and represent all information referring to ontologies. It has been exploited the potential of ontologies and semantic web languages to obtain interoperability and sharing knowledge between communicating parts.
Keywords: ontological mediator, personal assistant, user modeling.
sai: cosiwn.2008.08.013
Integrating Service Composition with Mobile Code Technologies
A. Corradi, A. Landini, E. Lodolo, S. Monti and S. Pasini
Communications of SIWN, Vol. 5, August 2008, pp. 11-17
Abstract: Service composition is an extremely powerful and versatile way to aggregate and reuse distributed services and software components into richer and complex scenarios. Workflow Management Systems (WFMSs) have emerged as one of the leading technologies to describe, manage, and execute service compositions but typically fail to support distributed scenarios, where distributed services should be invoked in a scalable and effective way. Mobile Agent (MA) platforms are a suitable technology to distribute the execution of complex service compositions, and therefore to enable scalability and improve performance. However, current proposals for MA-based WFMSs still target rather static and poorly distributed scenarios and exploit agent migration benefits only in a partial and insufficient way. Our model proposes to overcome these issues with a richer and more effective agent delegation strategy that can also cope with dynamic scenarios where services can move and replicate, so to achieve not only a better integration, but also the advantage of both technologies.
Keywords: distributed system, dynamic binding, mobile agent, service composition, workflow.
sai: cosiwn.2008.08.014
Using Well-Known Techniques for Classifying User Behavior Profiles
Jose Antonio Iglesias and Agapito Ledezma and Araceli Sanchis
Communications of SIWN, Vol. 5, August 2008, pp. 18-22
Abstract: The security of a computer is based on the realization of confidentiality, integrity, and availability. A computer can keep track of computer users to improve the security in the system. However, this does not prevent a user from impersonating another user. If a computer system can model the behavior of the users, it can be very beneficial detecting masqueraders, assisting them or predicting their future actions. In this paper, we present three different methods for classifying the behavior of a computer user. The proposed three methods are: Bayesian Netwoks, Hidden Markov Models and a method based on Term Weighting (TFIDF). These three techniques have been chosen because we want to assess pure statistical techniques (Bayesian Networks) and information-oriented techniques (TFIDF) against a technique that appears to be more adequate (at least in principle) for identifying the behavior of users (HMMs).
Keywords: Agent behavior, user profile models.
sai: cosiwn.2008.08.015
An Evolutionary Approach to Agent-based Pattern Classification
Federico Bergenti
Communications of SIWN, Vol. 5, August 2008, pp. 23-27
Abstract: This paper presents an agent-based approach to pattern classification that intensively uses evolutionary techniques to support supervised learning of classifiers. The paper first provides a formalization of the peculiar sort of agents and multi-agent systems that we consider. Then, a brief description of the adopted evolutionary technique is provided together with a justification for its use. Finally, the proposed approach is embedded into a prototypical, real-world application that offers a solid ground for the concrete assessment of classification performances. The preliminary results that we obtained clearly show that the proposed approach realizes a good trade off between accuracy and speed in a fully distributed and decentralized manner.
Keywords: pattern classification, evolutionary method, multiagent system.
sai: cosiwn.2008.08.016
A Progressive Filtering Approach to Hierarchical Text Categorization
Andrea Addis, Giuliano Armano and Eloisa Vargiu
Communications of SIWN, Vol. 5, August 2008, pp. 28-32
Abstract: The continuous growth of available document collections makes text categorization a challenging task to be investigated. Real-world scenarios are typically characterized by a huge amount of non relevant documents with respect to the documents a user is really looking for. In this paper, we investigate how the ratio between relevant and non relevant documents affects the performances of a classifier system. In particular, to counterbalance the negative impact of imbalanced inputs, we propose a novel progressive filtering technique. Performed on the RCV1-v2 benchmark, experiments confirm the validity of the approach.
Keywords: hierarchical text categorization, information filtering
sai: cosiwn.2008.08.017
Experimenting Combinations of Content-Based and Collaborative Filtering with a Photo Recommender System
Maurizio Agelli, Giuliano Armano, Giancarlo Cherchi, Maria Laura Clemente and Daniela Ghironi
Communications of SIWN, Vol. 5, August 2008, pp. 33-38
Abstract: With the on-going information explosion, it is becoming more and more critical to help users in retrieving items of actual interest. This work describes some experiments carried out in the field of photo recommendation, where a collaborative approach is combined in three different ways with a content-based one, in a scenario composed by few users that are very different among them. The content-based approach is an attempt to improve the performances of the collaborative one, by exploiting the textual descriptions that usually come with photos. Preliminary experiments show that improvements in the precision can be achieved for some users, although - in this particular context - the collaborative approach is on the average preferable.
Keywords: user profiling, content-based filtering, collaborative filtering, recommender systems
sai: cosiwn.2008.08.018
Improving Ranked Keyword Search with SENSE: SEmantic N-levels Search Engine
Pierpaolo Basile, Annalina Caputo, Marco de Gemmis, Anna Lisa Gentile, Pasquale Lops and Giovanni Semeraro
Communications of SIWN, Vol. 5, August 2008, pp. 39-45
Abstract: The traditional strategy performed by Information Retrieval (IR) systems is ranked keyword search: For a given query, a list of documents, ordered by relevance, is returned. Relevance computation is primarily driven by a basic stringmatching operation. To date, several attempts have been made to deviate from the traditional keyword search paradigm, often by introducing some techniques to capture word meanings in documents and queries. The general feeling is that dealing explicitly with only semantic information does not improve significantly the performance of text retrieval systems. This paper presents SENSE (SEmantic N-levels Search Engine), an IR system that tries to overcome the limitations of the ranked keyword approach, by introducing semantic levels which integrate (and not simply replace) the lexical level represented by keywords. Semantic levels provide information about word meanings, as described in a reference dictionary, and named entities. We show how SENSE is able to manage documents indexed at three separate levels, keywords, word meanings, and entities, as well as to combine keyword search with semantic information provided by the two other indexing levels.
Keywords: information retrieval, semantic search, word sense disambiguation, named entity disambiguation.
sai: cosiwn.2008.08.019
Kinetic Mashups: Augmenting physical places with motion-aware services
Vincenzo Pallotta
Communications of SIWN, Vol. 5, August 2008, pp. 46-50
Abstract: In this paper we propose a general framework for enabling a new generation of location-based services through the design motion-aware interaction. In order to validate the proposed solution, we present a mobile collaborative workflow application that allows users to interact with mobile services and to easily customize “just-in-time/place” service provisioning using a mashup-based Web2.0 interface.
Keywords: mobile and ubiquitous services, location-based services, kinetic user interfaces, collaboration.
sai: cosiwn.2008.08.020
Personalization and Collaborative Filtering for Information Retrieval on the Web
Emanuela De Vita, Massimo Deriu, Ivan Marcialis and Gavino Paddeu
Communications of SIWN, Vol. 5, August 2008, pp. 51-56
Abstract: Monitoring user’s behaviour, learning his interests and building his profile, is a widespread method to design personalized applications which meet better user needs. In this paper we present an approach that uses these three steps in the context of a Web search engines: analyzing the relevant actions in browsing session is possible to identify user’s favorite pages and, knowing what these pages deal with, to learn user’s favorite topics. An application, embedded in the Web browser, has been developed. It is able to accomplish all the tasks previously mentioned; moreover it provides a 3D interface to formulate the query and an innovative visualization, centered on user profile, of the results set.
Keywords: personalization, collaborative system, feedback, information retrieval, user modeling, user profiling, search engine, semantic, information retrieval, 3D GUI.
sai: cosiwn.2008.08.021
A Pervasive Framework for Multi-Agent Personalization in Smart Homes
Sylvain Giroux, Matthieu Castebrunet, Charles Gouin-Vallerand, Bessam Abdulrazak and Vincent Rialle
Communications of SIWN, Vol. 5, August 2008, pp. 57-61
Abstract: The growing population of cognitively impaired people combined to the scarcity of care giving resources compel for a new vision of assistance at home. But cognitively impaired people do not live alone all the time and the complexity of the devices and the services prevent to involve them in the process. These issues can be tackled in smart homes thanks to autonomic computing, multi-person localization, context awareness, and personalization. First, OSGi has to be adapted for building autonomic pervasive heterogeneous services that can be deployed and managed at distance and that owns self-management features. Then, a context aware multi-agent system for cognitive assistance is built on top of this layer for multi person localization, and personalization when there are many persons in a smart home.
Keywords: smart homes, multi-agent, OSGI, contextawareness, personalization, localization, cognitive assistance.
sai: cosiwn.2008.08.022
A Semantic Approach for Resource Cataloguing and Query Resolution
Angioni Manuela, Demontis Roberto and Tuveri Franco
Communications of SIWN, Vol. 5, August 2008, pp. 62-66
Abstract: The new generation of search engines is evolving from indexing and classification techniques of web resources to more sophisticated ones focusing on aspects like semantics, geo-referencing of the information, collaboration/distribution. In this paper we describe our work, developed in the DART project, that proposes a semantic approach to address such problems. We developed a module responsible of all semantic activities in the system, able to process and analyze textual documents and user queries by means of a syntactic and a semantic analysis and by means of text categorization techniques. The adoption of a semantic approach for resource cataloguing and query resolution allows filtering the number of results increasing the overall response quality. These techniques allow to process queries formulated in Natural Language ensuring in this manner a better level of relevance of the data provided to the user.
Keywords: semantic text categorization, indexing, query analysis, natural language processing.
sai: cosiwn.2008.08.023
Semantic Matchmaking via Non-Monotonic Reasoning: The MaMaS-tng Matchmaking Engine
Tommaso Di Noia, Eugenio Di Sciascio and Francesco M. Donini
Communications of SIWN, Vol. 5, August 2008, pp. 67-72
Abstract: In a semanticWeb service (SWS) matchmaking process, given a request, it is obvious that not only a list of services should be returned, but also a ranking of compatible SWSs should be provided. Obviously having semantically annotated services, the ranking should be based evaluating semantic similarity between descriptions. Furthermore, the availability of such descriptions makes explanation of rank possible and can provide useful information in order to modify or refine the original request. Here we summarize results obtained on this challenging topic exploiting non-monotonic inferences in Description Logics, with particular reference to their implementation within the MaMaS-tng engine.
Keywords: matchmaking, semantic web, seb services, ranking, abduction, contraction.
sai: cosiwn.2008.08.024
Fractal Browsing of Large Geo-Referenced Picture Sets
Davide Carboni, Valentina Marotto, Francesco Massidda and Pietro Zanarini
Communications of SIWN, Vol. 5, August 2008, pp. 73-78
Abstract: Nowadays millions of pictures are available on the Internet, uploaded and shared by large communities of users. With the mass diffusion of GPS devices, both in standalone navigators and in camera enabled PDAs/mobile phones, more and more of these pictures will be automatically georeferenced. Internet users need more effective tools and new interfaces in order to better find the pictures they are looking for. In this paper we describe some lessons learned about the visualization of large collections of geo-referenced pictures, which led us to the development of a novel User Interface (UI) approach.
Keywords: geo-data, mobile applications, GPS, photo browsing, geo-browsing, maps, web.
sai: cosiwn.2008.08.025
Collaborative Information Retrieval on Multitouch Interactive Walls
Alessandro Soro and Gavino Paddeu and Stefano Sanna
Communications of SIWN, Vol. 5, August 2008, pp. 79-83
Abstract: In this paper a novel framework for collaborative information retrieval is presented that exploits the combined use of cell phones and multi-user interactive walls to allow users to collaboratively explore the results of a search query. The user interface spreads among several displays; shared information is displayed on a large multi-touch interactive screen to be freely explored and manipulated. The cell phone serves for text entry, 3d pointing, user identification and to store/execute personal data/applications. User profiles and document classification are provided by the underlying query system and are exploited to visually enhance the relevance of documents with respect to any given user. Users’ gestures in turn are interpreted and fed back to the query system to improve user profiles and document classification.
Keywords: multi-touch, information retrieval, collaborative, multi-user, mobile, cell, phone, graspable, tangible.
-------------------------------------------------------------------------------------
Copyright © 2008 SIWN