Sotiris Batsakis

Improving the performance of focused web crawlers

This work addresses issues related to the design and implementation of focused crawlers. Several variants of state-of-the-art crawlers relying on web page content and link information for estimating the relevance of web pages to a given topic are proposed. Particular emphasis is given to crawlers capable of learning not only the content of relevant pages (as classic crawlers do) but also paths leading to relevant pages. A novel learning crawler inspired by a previously proposed Hidden Markov Model (HMM) crawler is described as well. The crawlers have been implemented using the same baseline implementation (only the priority assignment function differs in each crawler) providing an unbiased evaluation framework for a comparative analysis of their performance. All crawlers achieve their maximum performance when a combination of web page content and (link) anchor text is used for assigning download priorities to web pages. Furthermore, the new HMM crawler improved the performance of the original HMM crawler and also outperforms classic focused crawlers in searching for specialized topics.

TOQL: Temporal Ontology Querying Language

We introduce TOQL, a query language for querying time information in ontologies. TOQL is a high level query language that handles ontologies almost like relational databases. Queries are issued as SQL-like statements involving time (i.e., time points or intervals) or high-level ontology concepts that vary in time. Although independent from TOQL, this work suggests a mechanism for representing time evolving concepts in ontologies based on the four-dimensional perdurantist mechanism. However, TOQL prevents users from being familiar with the representation of time in ontologies. To show proof of concept, an application has been developed that supports translation and execution of TOQL queries on temporal ontologies combined with a reasoning mechanism based on event calculus. A real world temporal ontology is also implemented on which several TOQL example queries are processed and discussed.

CHRONOS: A Reasoning Engine for Qualitative Temporal Information in OWL☆

Abstract We propose CHRONOS, a system for reasoning over temporal information in OWL ontologies. Representing both qualitative temporal (i.e., information whose temporal extents are unknown such as “before”, “after” for temporal relations) in addition to quantitative information (i.e., where temporal information is defined precisely e.g., using dates) is a distinctive feature of the proposed approach. Qualitative representations are very common in natural language expressions such as in free text or speech and can be proven to be valuable in the Semantic Web. CHRONOS reasoner applies over temporal relations in order to infer implied relations and to detect inconsistencies while retaining soundness, completeness and tractability over the supported relations set. The experimental results demonstrated that CHRONOS runs up to several times faster compared to its SWRL counterpart and scales-up well for large relation sets.

Representing Temporal Knowledge in the Semantic Web: The Extended 4D Fluents Approach

Representing information that evolves in time in ontologies, as well as reasoning over static and dynamic ontologies are the areas of interest in this work. Building upon well established standards of the semantic Web and the 4D-fluents approach for representing the evolution of temporal information in ontologies, this work demonstrates how qualitative temporal relations that are common in natural language expressions (i.e., relations between time intervals like “before”, “after”, etc.) are represented in ontologies. Existing approaches allow for representations of temporal information, but do not support representation of qualitative relations and reasoning.

Imposing restrictions over temporal properties in OWL: a rule-based approach

Introducing the dimension of time in ontologies turns binary relations into ternary which cannot be handled by OWL. Approaches such as the N-ary relations or the 4D-fluents approach discussed in this work, offer satisfactory solutions to this problem. However, data and property semantics are not preserved in the resulting representations nor can they be handled by ordinary reasoners such as Pellet. We propose a rule-based solution to this problem using SWRL.

The TOQL System

TOQL, is a query language for querying time information in ontologies. An application has been developed that supports translation and execution of TOQL queries on temporal ontologies. A Graphical User Interface (GUI) has been also developed to facilitate user interaction and supports operations such as syntax highlighting, code autosuggestion, loading of the ontology into the main memory, results and error display.