Amelia Bădică

L-wrappers: concepts, properties and construction: A declarative approach to data extraction from web sources

In this paper, we propose a novel class of wrappers (logic wrappers) inspired by the logic prog- ramming paradigm. The developed Logic wrappers (L-wrapper) have declarative semantics, and therefore: (i) their specification is decoupled from their implementation and (ii) they can be generated using inductive logic programming. We also define a convenient way for mapping L-wrappers to XSLT for efficient processing using available XSLT processing engines.

FSP and FLTL framework for specification and verification of middle-agents

FSP and FLTL framework for specification and verification of middle-agents Agents are a useful abstraction frequently employed as a basic building block in modeling service, information and resource sharing in global environments. The connecting of requester with provider agents requires the use of specialized agents known as middle-agents. In this paper, we propose a formal framework intended to precisely characterize types of middle-agents with a special focus on matchmakers, brokers and front-agents by formally modeling their interactions with requesters and providers. Our approach is based on capturing interaction protocols between requesters, providers and middle-agents as finite state processes represented using FSP process algebra. The resulting specifications are formally verifiable using FLTL temporal logic. The main results of this work include (i) precise specification of interaction protocols depending on the type of middle-agent (this can also be a basis for characterizing types of middle-agents), (ii) improvement of communication between designers and developers and facilitation of formal verification of agent systems, (iii) guided design and implementation of agent-based software systems that incorporate middle-agents.

Agent-Based System for Brokering of Logistics Services – Initial Report

The competition between logistics companies leads to their continuous concern of improving their competitiveness in the market by increasing the quality of provided logistics services and by reducing logistics costs. The key to solving the above problems is the efficient management of logistics information. Therefore, building a smart logistics services supply system is imperative to each logistics company. In this paper we propose a multi-agent system for smart brokering of logistics services and we provide a preliminary analysis and design sketch of the system.

Rule Learning for Feature Values Extraction from HTML Product Information Sheets

The Web is now a huge information repository with a rich semantic structure that, however, is primarily addressed to human understanding rather than automated processing by a computer. The problem of collecting product information from the Web and organizing it in an appropriate way for automated machine processing is a primary task of software shopping agents and has received a lot of attention during the last years. In this paper we assume that product information is represented as a set of feature-value pairs contained in an HTML product information sheet that is usually formatted using HTML tables. The paper presents a technique for learning extraction rules of product information from such product information sheets. The technique exploits the fact that the Web pages that represent product information of a certain producer are generated on the fly from the producer database and therefore they exhibit uniform structures. Consequently, while the extraction task is executed manually for a few information items by a human user, a general-purpose inductive learner (we have used FOIL in our experiments) can learn extraction rules that will be further applied to the current and other product information sheets to automatically extract other items. The input to the learning algorithm is a relational description of the HTML document tree that defines the HTML tree nodes types and the relationships between them. The approach is demonstrated with appropriate examples, experimental results, and software tools.

Distributed agent-based ant colony optimization for solving traveling salesman problem on a partitioned map

In this paper we discuss the experimental evaluation of an improved configuration of our recent framework ACODA (Ant Colony Optimization on a Distributed Architecture) for solving the Traveling Salesman Problem (TSP). ACODA is a novel multi-agent system architecture for distributed Ant Colony Optimization in a decentralized environment. This new configuration improves the execution time by allowing each software agent of ACODA to manage a part of the TSP map rather than a single map node. Experimental results support this claim of improved efficiency while showing that the scalability of the system is preserved.

Formal Modeling of Agent-Based English Auctions Using Finite State Process Algebra

The vision of global agent-based e-commerce environments that enable dynamic trading between business partners requires the study and development of suitable formal modeling frameworks. In particular, negotiation is a necessary and important activity to allow engagement of business parties in non-trivial business relationships. In this note we propose a formal framework using finite state process algebrafor modeling and analysis of agent-based negotiations, with a focus on a particular price negotiation --- English auction.

A New Path Generalization Algorithm for HTML Wrapper Induction

Recently it was shown that Inductive Logic Programming can be successfully applied to data extraction from HTML. However, the approach suffers from two problems: high computational complexity with respect to the number of nodes of the target document and to the arity of the extracted tuples. In this note we address the first problem by proposing an efficient path generalization algorithm for learning rules to extract single information items. The presentation is supplemented with a description of a sample experiment.

Declarative Representation and Solution of Vehicle Routing with Pickup and Delivery Problem

Abstract Recently we have proposed a multi-agent system that provides an intelligent logistics brokerage service focusing on the transport activity for the efficient allocation of transport resources (vehicles or trucks) to the transport applications. The freight broker agent has a major role to coordinate transportation arrangements of transport customers (usually shippers and consignees) with transport resource providers or carriers, following the freight broker business model. We focus on the fundamental function of this business that aims to find available trucks and to define their feasible routes for transporting requested customer loads. The main contribution of this paper is on formulating our scheduling problem as a special type of vehicle routing with pickup and delivery problem. We propose a new set partitioning model of our specific problem. Vehicle routes are defined on the graph of cities, rather than on the graph of customer orders, as typically proposed by set partitioning formulations. This approach is particularly useful when a large number of customer orders sharing a significantly lower number of pickup and delivery points must be scheduled. Our achievement is the declarative representation and solution of the model using ECLiPSe state-of-the-art constraint logic programming system.

Information flow in a distributed agent-based online auction system

Recently we were involved in the development of a distributed agent-based English auction server. The focus of the work was on emphasizing the advantages brought by the multi-agent systems technology to the high-level of abstraction, modularity and performance of the server architecture, and its implementation. Less effort was spent on the analysis of its external functionalities and usability, as well as on approaching the implementation of a realistic system for online auctions. Therefore, in this paper we present our solution, in terms of information flow management, and its relation to the functionalities of the system for online auctions that incorporates our server. The main outcome of this work is a clean specification of the information exchanges between the agent and non-agent software components of the system. Special attention is also provided for the interoperability of the different data communication, and agent and non-agent software technologies that are employed for the implementation of the system.

Logic wrappers and XSLT transformations for tuples extraction from HTML

Recently it was shown that existing general-purpose inductive logic programming systems are useful for learning wrappers (known as L-wrappers) to extract data from HTML documents. Here we propose a formalization of L-wrappers and their patterns, including their syntax and semantics and related properties and operations. A mapping of the patterns to a subset of XSLT that has a formal semantics is outlined and demonstrated by an example. The mapping actually shows how the theory can be applied to obtain efficient wrappers for information extraction from HTML.