Nikolay Shilov

Anonymous Agent Coordination in Smart Spaces: State-of-the-Art

Development of new technologies brings people new possibilities such as smart spaces. Smart spaces can provide better user experience by allowing a user to connect new devices flexibly and to access all the information in the multi device system from any of the devices. The paper describes work-in-progress performed under a joint project of SPIIRAS and Nokia. The project is aimed at the analysis of existing anonymous agent coordination languages and adaptation of them to coordination of smart space devices.

Patterns for context-based knowledge fusion in decision support systems

The here presented research focuses on the context-based knowledge fusion patterns. Patterns are discovered based on an analysis and investigation of knowledge fusion processes in a context aware decision support system at the operational stage of the system functioning. At this stage the context-based knowledge fusion processes are manifested around the context. The patterns are generalized in regard to the following three aspects: (1) the effects that the knowledge fusion processes produce in the system; (2) the preservation of internal structures for the context and multiple sources the information/knowledge is fused from; and (3) the preservation of multiple sources and the context autonomies. At that, seven knowledge fusion patterns have been discovered: simple fusion, extension, instantiated fusion, configured fusion, adaptation, flat fusion, and historical fusion.

Recommendation system for tourist attraction information service

The paper proposes a description of information decision support system in the tourism domain and a set of methods and algorithms for generating recommendations for a user that allow significant increase of the system usability. The system generates for the user recommendations which attractions at the moment are better to attend based on the user preferences and the current situation in the location area. The system also allows showing the user information about interesting attraction in more detail, which is based on analyzing information evaluations made by other users.

On-the-fly ontology matching in smart spaces: a multi-model approach

Proper functioning of smart spaces demands semantic interoperability between knowledge processors connected to it. As a consequence it is required to develop models that would enable knowledge processors to translate on-the-fly between the internal and smart space ontologies. This paper presents the developed multi-model approach to the above problem, where the major elements of the selected approach are described in detail.

Semantic-oriented support of interoperability between production information systems

The paper proposes an approach to support the interoperability in Build-To-Order (BTO) production networks. The approach covers technological, problem and semantic levels: (i) interoperability between different systems within one company or in several companies, (ii) distributed product data management and (iii) design and development of integrated products based on required functions. In the approach the knowledge is represented via ontologies described using Object-Oriented Constraint Network (OOCN) formalism. This formalism combines the advantages of object-oriented information representation and constraint networks. Since the problems to be solved are presented in the same formalism but may differ in structure and in values of variables, an on-the-fly problem modification and solving via an adaptive service are proposed. The approach is illustrated via an example from automotive industry and possible future work including the development of a methodology for customer-oriented complex product configuration based on its required functionality.

Smart Logistic Service for Dynamic Ridesharing

The paper describes a service-based approach to dynamic ridesharing based on the smart space concept. Presented system allows planning and assisting a tourist for museum attending and finds a driver for reaching the museums. The paper applies an approach presented in FRUCT 11 to assisting tourists in a certain area (e.g., in a given city). Smart-M3 information platform is used as a smart space infrastructure for the presented approach. The service is based on the Smart-M3 ontology which is formed by ontology slices of user’s mobile devices. The paper presents an algorithm for finding appropriate fellow-travelers for drivers as well as definition of acceptable pick-up and drop-off points for them.

Efficient distributed information management in smart spaces

Development of new technologies brings people new possibilities such as smart spaces. Smart spaces can provide better user experience by allowing a user to connect new devices flexibly and to access all the information in the multi device system from any of the devices. However, this also raises new challenges. The paper considers the problem of efficient distributed information sharing in smart spaces. Models for information distribution are presented together with a simulation framework.

Ontology-driven intelligent service for configuration support in networked organizations

Nowadays, organizations must continually adapt to market and organizational changes to achieve their most important goals. Migration to business services and service-oriented architectures provides a valuable opportunity to attain the organization objectives. This migration causes evolution both in organizational structure and in technology-enabling businesses to dynamically change vendors and services. One of the forms of organizational structures is the form of networked organization. Technologies of business intelligence and Web intelligence effectively support business processes within the networked organizations. While business intelligence focuses on development of services for consumer needs recognition, information search, and evaluation of alternatives; Web intelligence addresses advancement of Web-empowered systems, services, and environments. The paper proposes a technological ontology-driven framework for configuration support as applied to networked organization. The framework integrates concepts of business intelligence and Web intelligence into a collaboration environment of a networked organization on the base of attainment of knowledge logistics purposes. This framework referred to as KSNet is based on the integration of software agent technology and Web services. Knowledge logistics functions of KSNet are complemented by technological functions of knowledge-gathering agents. The services of these agents are implemented with CAPNET, a FIPA compliant agent platform. CAPNET allows consuming services of agents in a service-oriented way. Applicability of the approach is illustrated through a “Binni scenario”-based case study of a portable field hospital configuration.

Ontology for cyber-physical-social systems self-organisation

Cyber-Physical-Social Systems (CPSSs) integrate various resources from physical, cyber, and social worlds. Efficient interaction of these resources is essential for CPSSs operation. Ontologies do not only provide for semantic operability between different resources but also provide means to create sharable ontology-based context models specified for actual settings. Usage of the context supports situation-driven behavior of CPSSs resources and thus is an enabler for their self-organisation. The present research inherits the idea of context ontologies usage for modelling context in CPSSs. In this work, an upper level context ontology for CPSSs is proposed. This ontology is applied in the domain of self-organising resource network.

Well-structured model checking of multiagent systems

We address model checking problem for combination of Computation Tree Logic (CTL) and Propositional Logic of Knowledge (PLK) in finite systems with the perfect recall synchronous semantics. We have published already an (update+abstraction)-algorithm for model checking with detailed time upper bound. This algorithm reduces model checking of combined logic to model checking of CTL in a finite abstract space (that consists of some finite trees). Unfortunately, the known upper bound for size of the abstract space (i.e. number of trees) is a non-elementary function of the size of the background system. Thus a straightforward use of a model checker for CTL for model checking the combined logic seems to be infeasible. Hence it makes sense to try to apply techniques, which have been developed for infinite-state model checking. In the present paper we demonstrate that the abstract space provided with some partial order on trees is a well-structured labeled transition system where every property expressible in the propositional µ-Calculus, can be characterized by a finite computable set of maximal elements. We tried feasibility of this approach to model checking of the combined logic in perfect recall synchronous environment by automatic model checking a parameterized example.