*Toward Strategy Selection for Ontology Alignment

نویسندگان

  • Yi Li
  • Jie Tang
  • Duo Zhang
  • Juanzi Li
چکیده

This paper is concerned with the issue of strategy selection for ontology alignment. By strategy selection we mean selecting strategies to align ontologies for different tasks according to the characteristics of the ontologies. Previously, strategy selection issues were often conducted manually or in an adhoc fashion. This paper first gives a formalization of the entire problem. It then proposes an approach to estimate two factors for each alignment task and perform the selection based on the factors. The proposed approach has been implemented in an alignment tool called RiMOM. Given two ontologies, RiMOM intends to combine multiple strategies, aiming at finding the “optimal” alignments from the source ontology to the target one. RiMOM has six basic strategies: edit-distance based strategy, statistical-learning based strategy, dictionary based strategy, and three similarity-propagation based strategies. Each strategy is defined based on one kind of ontological-information or approach. The proposed strategy selection approach is used to determine which strategies will be used in the alignment task. We conducted experiments on the data sets of OAEI’2006. Experimental results indicate that our proposed approach performs significantly better than most of the participants and is comparable with COMA++ and Falcon-AO on the benchmark data set.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Result of Ontology Alignment with RiMOM at OAEI'07

In this report, we give a brief explanation of how RiMOM obtains the ontology alignment results at OAEI’07 contest. RiMOM integrates different alignment strategies: edit-distance based strategy, vector-similarity based strategy, path-similarity based strategy, background-knowledge based strategy, and three similarity-propagation based strategies. Each strategy is defined based on one specific o...

متن کامل

Result of Ontology Alignment with RiMOM at OAEI'06

In this report, we briefly describe our system RiMOM and its underlying techniques. Given two ontologies, RiMOM intends to combine multiple strategies, aiming at finding the “optimal” alignments from the source ontology to the target one. RiMOM integrates multiple strategies: edit-distance based strategy, statistical-learning based strategy, and three similaritypropagation based strategies. Eac...

متن کامل

A Dynamic Multistrategy Ontology Alignment Framework Based on Semantic Relationship using WordNet

Ontology matching has emerged as a crucial step when information sources are being integrated. Hence, ontology matching has attracted considerable attention in both academia and industry. Clearly, as information sources grow rapidly, manual ontology matching becomes tedious, time-consuming and leads to errors and frustration. Thus the need for automated and semi-automated approaches becomes inc...

متن کامل

Towards a Methodology for Selection of Suitable Matching Approaches - A Case Study

The natural solution to fully explore the potentials hidden in existing matching approaches is a strategy based on the reuse of available matchers. However, building semantic applications based on the reuse of existing approaches is still a tedious process, this is due to fact that reusing available alignment methodologies within new application contexts is currently not being discussed deeply....

متن کامل

RiMOM Results for OAEI 2008

In this report, we give a brief explanation of how RiMOM obtains the ontology alignment results at OAEI 2008 contest. We introduce the alignment process of RiMOM and more than 8 different alignment strategies integrated in RiMOM. Since every strategy is defined based on one specific ontological-information, we, in particular, study how the different strategies perform for different alignment ta...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2007