Grażyna Brzykcy

Multi-agent blackboard architecture for a mobile robot

In the paper, the architecture of a mobile robot cooperating with other robots and some stationary devices in a task of collective perception and world modelling is considered. We analyze data-driven processing of information performed by an individual robot treated as an agent and we propose to organize it as a set of experts (also treated as agents) exchanging data by means of a blackboard. We analyze functions performed by the blackboard agents and present results of preliminary experiments with real robots.

Extended rules in knowledge-based data access

We present a method for an efficient knowledge-based access to relational data. Knowledge is represented as a set of rules (basic rules) and describes a source data at concept (ontological) level. Forward chaining in the integrated system is performed with extended rules, which are obtained by a goal- and dependency-directed transformation of the basic rules. The novel feature of our method is generality - every rule is generated so that includes all possible binding of the head predicates, and variable dependencies, while in many implementations of the magic method the succession of bindings depends on a query. We demonstrate a query answering algorithm and our prototypical implementation of the system coupled with the Jess engine. The results of performance evaluation are presented and compared to the results described in our previous works.

Semantic Data Integration in P2P Environment Using Schema Mappings and Agent Technology

We discuss the problem of semantic data integration in a highly-dynamic environment consisting of a community of peer-to-peer cooperating agents (partners). Peers decide when to join and when to leave the system, how to describe their local data, when to communicate and share their resources with partners. An agent issues queries to its partners (friends) which are able partly answer the query and then propagate the query to their partners along semantic paths existing in the system. Semantic paths are determined by schema mappings defined between partners. We propose a method for specifying schema mappings and to translate them to XQuery expressions. Mappings are represented by means of logical formulas. We also propose a declarative specification of semantic-driven communication in the system. The specification is made in a peer-oriented extension of Prolog.

Data Access Based on Faceted Queries over Ontologies

We propose a method for generating and evaluating faceted queries over ontology-enhanced distributed graph databases. A user, who only vaguely knows the domain ontology, starts with a set of keywords. Then, an initial faceted query is automatically generated and the user is guided in interactive modification and refinement of successively created faceted queries. We provide the theoretical foundation for this way of faceted query construction and translation into first order monadic positive existential queries.

Faceted Query Answering in a Multiagent System of Ontology-Enhanced Databases

We discuss an architecture and functionality of a multiagent system supporting data integration based on faceted query evaluation in a system of ontology-enhanced graph databases. We use an ontology (a subset of OWL 2 RL profile) to support a user in interactive query formulation. A set of server agents cooperate one with another in query answering using the ontological knowledge for query rewriting and for deciding about query propagations. We propose some algorithms ensuring tractable complexity of query rewriting and query evaluation.

Data integration in a system with agents' models

A problem of data integration in an environment consisting of a community of peer-to-peer cooperative agents is discussed. Data in the system are locally structured by the agents under different schemas. Communication and cooperation in the system are based on asking and answering queries by the agents. We show how to represent data and to merge related information from multiple data sources by means of unification operation and diverse Prolog mechanisms. The declarative approach is suitable to model data integration task and other constraints on data: from ontological to locally induced by an agent.

Information Flow in a Peer-to-Peer Data Integration System

In the paper a problem of semantic integration of XML data in a peer-to-peer agent system is considered. In the system each agent manages its local data, and can communicate and cooperate with other agents by asking and answering queries. Problem of information flow between local data resources is analysed upon a channel theory. The abstract notions of classification and infomorphism are used to clarify the roles of queries and mappings in the system. The query reformulation task is modelled by the composition of an appropriate infomorphisms. In order to specify classifications and infomorphisms ***-terms are used. Logic programming systems are chosen because of the logical variables and partial data representation, effective techniques of data unification and partial evaluation, suitable built-in terms and procedures, and automated reasoning.

Reasoning with the depth-first iterative deepening strategy in the DLog system

We propose an extension to the DLog system, which is a resolution-based ABox reasoner for the SHIQ description logic, particularly intended for instance checking and instance retrieval. We replace the original search strategy of the derivation tree, namely the depth-first search, by the depth-first iterative deepening search. The latter approach is proven to be asymptotically optimal among brute-force strategies in terms of proof length, space and time. The extension leads to shorter proofs and, on average, to better timing results when it is enough to compute one positive answer to the given query. We compare the performance of the original and the extended version of the DLog system on a simple benchmark set.

Information flow based specification of data integration problem

In the paper a problem of semantic integration of XML data in a distributed heterogeneous system is considered. In the system local data sources are organized upon different schemas of labelled graph structures. Agents manage its local data, and communicate and cooperate with each other by asking and answering queries. To establish the appropriate models for the problem the information flow theory is chosen with the abstract notions of classification and infomorphism. Generalized terms from extended logic programming systems are used to implement the notions. They are intended to represent structure and semantics of data and to model partial data. The extended unification operation plays the primary role in query reformulation and data merging tasks. An algorithm of automatic inferring mappings between XML schemas is defined over the operation. Equipped with automated reasoning and partial evaluation technique the constraint logic programming systems are particularly suitable for data integration task.

Some Notes on Declarative Specification of Semantic Data Integration Task in a Peer-to-peer Agent System

In the paper, a problem of semantic integration of XML data in a peer-to-peer agent system is analysed upon a channel theory. In the system, each agent manages its local data, and communication and cooperation actions, executed by the agents, consist of asking and answering queries. The abstract notions of classification and infomorphism are used to clarify the roles of queries and mappings in the system. The query reformulation task is modelled by the composition of an appropriate infomorphisms. In order to specify classifications and infomorphisms executable Prolog-like clauses and generelized terms are used. Logic programming systems are chosen because of the logical variables and partial data representation, effective techniques of data unification and partial evaluation, suitable built-in terms and procedures, and automated reasoning.