Paper Review Form

 

Section I. Overview

 

A.  Reader Interest

           

            1.  Which category describes this manuscript?

                        _x_Practice/Application/Case Study/Experience Report

                        ___Research/Technology

                        ___Survey/Tutorial/How-To

 

B.  Content

           

1.       Please explain how this manuscript advances this field of research and/or contributes something new to the literature.

 

This paper presents a new method to use RDF algebra for optimization and to query RDF metadata. 

 

           

1.       Is the manuscript technically sound?

            ___Yes

                        ___Appears to be - but didn't check completely

                        _x_Partially

                        ___No

 

C. Presentation

 

            1.  Are the title, abstract, and keywords appropriate?

                        _x_Yes

                        ___No

 

1.       Does the manuscript contain sufficient and appropriate references?  

___References are sufficient and appropriate

                        _x_Important references are missing; more references are needed

                        ___Number of references are excessive

 

2.       Does the introduction state the objectives of the manuscript in terms that encourage the reader to read on?

            ___Yes

                        _x_Could be improved

                        ___No

 

            4.  How would you rate the organization of the manuscript? Is it focused?  Is the length appropriate for the topic?

                        ___Satisfactory

                        _x_Could be improved

                        ___Poor

 

            5.  Please rate and comment on the readability of this manuscript.

                        ___Easy to read

                        _x_Readable - but requires some effort to understand

                        ___Difficult to read and understand

                        ___Unreadable

 

Section II. Evaluation

 

            Please rate the manuscript. Explain your choice.

                        ___Award Quality

                        ___Excellent

                        ___Good

                        _x_Fair

                        ___Poor

 

 

Section III. Detailed Comments

The paper shows the necessity of RDF algebra for optimization and querying RDF metadata. the RDF and RDFS proposed by W3C, can be used for describing metadata, but not for querying metadata.

 

Three main approaches for querying RDF (meta)data have been proposed.

 

The first approach (supported in the W3C working group by Stanford) is to view RDF data as a knowledge base of triples. Triple, the successor of SiLRI (Simple Logicbased RDF Interpreter), maps RDF metadata to a knowledge base in Horn Logic (replacing Frame Logic). A similar approach is taken in Metalog, which matches triples to predicates in Datalog, a subset of Horn Logic. In this way one can query RDF descriptions at a high level of abstraction: the querying takes place at a logical layer that supports inference.

 

The second approach (proposed by IBM) builds upon the XML serialization of RDF. In the “RDF for XML” project (recently removed), an RDF API is proposed on top of the IBM AlphaWork’s XML 4 Java parser. In the context of the same project a declarative query language for RDF (RDF Query) [27] was created for which both input and output are resource containers. One of the nice features of this query language is that it proposes operators similar to the relational algebra, leaving the possibility to reuse some of the 25 years experience with relational databases. Unfortunately, the language fails to include the inference rules specific to RDF Schema, loosing description semantics.

 

The third approach (coming from ICS-FORTH in Greece) uses the RDF Graph Model for defining the RDF query language RQL. It extends previous work on semistructured query languages (e.g., path expressions, filtering capabilities etc.) with RDF peculiarities. Its strength lies in the ability to uniformly query both RDF descriptions and schemas. Compared to the previous approach it exploits the inference given in the RDF Schema (e.g., multiple classification of resources, taxonomies of classes and properties etc.) making it the most advanced RDF query language proposed so far.