Concept Of Similarity Measure Research Articles

Evaluation of Scalable Database Driven Reverse Dictionary

While a traditional forward dictionary maps words to their definitions, a reverse dictionary takes a user input phrase with a desired concept, and returns a set of candidate words closely related to the input phrase. This application is significant not only for the general public, but mainly to those who work personally with words. It is also important in the general field of conceptual search. Upon receiving a search concept, the Reverse Dictionary consults the forward dictionary and selects those words whose definitions are similar to the given concept. And thus it is reduced to a concept similarity problem. In this paper, different concept similarity measures are compared and the best among them is proposed. The experimental results shows that the approach used here provides significant improvements in performance level without losing the quality of the result.

An Optimization for Hybrid Semantic Similarity Computation

Semantic similarity computation is of great importance in many applications such as natural language processing, knowledge acquisition and information retrieval. In recent years, many concept similarity measures have been developed for ontology and lexical taxonomy. Generally speaking, ontology concepts semantic similarity computation is tedious and time-consuming. This paper puts forward an optimization algorithm to simplify semantic similarity computation. The optimization algorithm utilizes hierarchical relationship between concepts to simplify similarity computation process. Simulation experiments showed the optimization algorithm could make similarity computation simple and convenient, and similarity computation speed was improved by one time. The more complexity an ontology structure, and the bigger the maximum depth of ontology, the more significantly the performance improved.

Concept Similarity Measure with Hierarchy Structure and Information

Calculating semantic similarity between concepts in ontology is an important issue in natural language processing and so on, so how to measure the similarity becomes a hot topic among many scholars. However, most existing methods cannot distinguish the similarity further. Confronting this problem, we propose a new semantic similarity method combining hierarchy structure of ontology and information content of two concepts based on domain ontology, which highlights the semantic information of leaves in the ontology structure. Our experiment demonstrates that, compared with other available methods, our proposal can improve the accuracy between two leaves and between leaf and non-leaf.

A hierarchy-based similarity measure for intuitionistic fuzzy sets

Due to some unreasonable results obtained from most current similarity measures for intuitionistic fuzzy sets (IFSs), we introduce a necessary condition to obtain a stronger definition of similarity measures for IFSs, and present a new similarity measure derived from a general idea of similarity measures for concepts on a lattice. In experiments, we focus our attention on two basic directions of performance evaluation: one is how much the proposed measure is reasonable and the other is how much accuracy the measure produces when it is applied to classification problems. The experimental results show that the proposed measure is reasonable and achieves a satisfactory performance on classification problems.

Semantically Aware Web Service Composition Through AI Planning

Web service composition is a significant problem as the number of available web services increases; however, manual composition is not an efficient option. Automated web service composition can be performed using AI Planning techniques, utilizing descriptions of available atomic web services, enhanced with semantic awareness and relaxation. This paper discusses a unified, semantically aware approach, handling both semantic (OWL-S & SAWSDL) and non-semantic (WSDL) web service descriptions. In the first case, ontology analysis is adopted to semantically enhance the planning domains and problems, in order to deal with cases where exact syntactic input-to-output matching is not feasible. In the non-semantic descriptions case, semantic information is acquired utilizing alternative sources such as lexical thesauri. Concept similarity measures are applied and utilized to achieve the desired degree of semantic relaxation. The solution to a web service composition problem is a plan describing the desired composite service. To support the proposed approach, the PORSCE framework has been implemented. The framework is modular, integrating discrete web service description languages and semantic relaxation techniques. Based on the similarity measures suggested in the paper, performance issues are also explored.

Similarity-based relaxed instance queries

In Description Logics (DL) knowledge bases (KBs), information is typically captured by clear-cut concepts. For many practical applications querying the KB by crisp concepts is too restrictive; a user might be willing to lose some precision in the query, in exchange of a larger selection of answers. Similarity measures can offer a controlled way of gradually relaxing a query concept within a user-specified limit.In this paper we formalize the task of instance query answering for DL KBs using concepts relaxed by concept similarity measures (CSMs). We investigate computation algorithms for this task in the DL EL, their complexity and properties for the CSMs employed regarding whether unfoldable or general TBoxes are used. For the case of general TBoxes we define a family of CSMs that take the full TBox information into account, when assessing the similarity of concepts.

Designing a novel cross ontology measure to extract resources from knowledge base for an adaptive e-learning system

E-Learning deals with the interaction between the teacher and learner on the basis of knowledge possessed by the learner. Ontology is a key concept that helps in providing more relevant and accurate information to the users for a particular domain or from multiple domains. Finding semantic similarity between the concepts of different ontologies will enhance the integration and retrieval of resources in a more meaningful way. In this paper, an adaptive e-learning system is proposed, which generates use r-specific e-learning contents based on their knowledge level and by comparing the similarity of concepts between ontologies. Based on this a new approach is designed to retrieve more relevant and accurate documents by embedding a new Concept Similarity measure ConSim, after evaluating the user's knowledge based on the search query. The experimental results prove that after embedding the ConSim measure the retrieval results have been improved.

A new case-based classification using incremental concept lattice knowledge

This paper proposes a new case-based classification system with an incremental knowledge base. The new system employs a concept lattice with formal concept analysis as a knowledge structure. The paper also proposes a new efficient algorithm for knowledge construction as well as an effective retrieval method for formal concepts. The proposed retrieval method uses a concept similarity measure based on an appearance frequency of formal concepts. In addition, we provide a mathematical proof that the similarity measure satisfies a formal similarity metric definition. Experiment results on standard datasets show that our classifier with the proposed similarity measure gives accuracy better than with other existing similarity measures.

Comparison of fuzzy numbers based on preference

Ranking fuzzy numbers plays an important role in a fuzzy decision-making process. However, fuzzy numbers may not be easily ordered into one sequence due to the overlap between them. A new approach is introduced to detect the overlapped fuzzy numbers based on the concept of similarity measure, incorporating the preference of the decision-maker into the fuzzy ranking process. Numerical examples and comparisons with other methods are presented to evaluate the new method. The computational process of the proposed method is straightforward and is practically capable of comparing similar fuzzy numbers. The proposed method is an absolute ranking and no pairwise comparison of fuzzy numbers is necessary. Furthermore, through some examples discussed in this work, it is proved that the proposed method possesses several good characteristics compared to other methods examined in this work.

FCA based Concept Constructing and Similarity Measurement Algorithms

Traditionally information retrieval consists mainly of determining which documents of a collection contain the keywords in the user query. However, a growing number of tasks, especially those related to Semantic Web technologies and applications rely on accurately measuring the similarity between documents and online texts. Instead of giving the absolute similarity degree of two documents, this paper presents a semantic corpus and Formal Concept Analysis-based procedure to build the concept map for a given set of documents and quantify the semantic relations between the concepts. The proposed approach includes three algorithms - a) the concept lattice constructing algorithm, b) the concept similarity measure, and c) the sub-lattice similarity measure.

A Fuzzy Concept Similarity Measure Based on Lattice Structures

With the rapid development of the semantic web, determining the degree of similarity between concepts from same or different ontologies plays an increasing crucial role. In this paper, a new similarity model based on lattice structural information is proposed to evaluate the similarity degree between fuzzy concepts in the framework of fuzzy formal concept analysis. The proposed method preserves more structural information, which can be viewed as another extension and development of de Souza and Davis’s model in fuzzy context.

A service concept recommendation system for enhancing the dependability of semantic service matchmakers in the service ecosystem environment

A Service Ecosystem is a biological view of the business and software environment, which is comprised of a Service Use Ecosystem and a Service Supply Ecosystem. Service matchmakers play an important role in ensuring the connectivity between the two ecosystems. Current matchmakers attempt to employ ontologies to disambiguate service consumers’ service queries by semantically classifying service entities and providing a series of human computer interactions to service consumers. However, the lack of relevant service domain knowledge and the wrong service queries could prevent the semantic service matchmakers from seeking the service concepts that can be used to correctly represent service requests. To resolve this issue, in this paper, we propose the framework of a service concept recommendation system, which is built upon a semantic similarity model. This system can be employed to seek the concepts used to correctly represent service consumers’ requests, when a semantic service matchmaker finds that the service concepts that are eventually retrieved cannot match the service requests. Whilst many similar semantic similarity models have been developed to date, most of them focus on distance-based measures for the semantic network environment and ignore content-based measures for the ontology environment. For the ontology environment in which concepts are defined with sufficient datatype properties, object properties, and restrictions etc., the content of concepts should be regarded as an important factor in concept similarity measures. Hence, we present a novel semantic similarity model for the service ontology environment. The technical details and evaluation details of the framework are discussed in this paper.

Applying Concept Similarity to the Evaluation of Common Understanding in Multidisciplinary Learning

The success of an architecture, engineering, and construction (AEC) project largely depends on the effective collaboration of project participants. This characteristic of the industry requires that AEC education shall foster multidisciplinary collaboration skills of students. On the other hand, many AEC education programs have started using information and communication technologies (ICTs) to facilitate teaching and learning such as distance learning. One of the challenges in a computer-mediated learning environment arises from the need for computer systems to determine whether a common understanding regarding a certain subject is established among a group of students. To achieve this goal, ICT needs to be able to detect changes in the knowledge structure of students in order to provide better mediation. This paper presents a study on the development of a concept similarity measure that has less computational complexity than graph-based similarity analyses. The similarity measure is intended to determine the similarity of multiple knowledge structures [e.g., concept maps (CMAPs)] of students. The concept similarity measure is developed based on the feature-based method in which propositions associated with a concept are considered as features of the concept. Therefore, the similarity of concepts can be measured by comparing the propositions of the concepts. In addition, since concepts are the key component of a knowledge structure, the common understanding of students can be determined by measuring the similarity of concepts in the knowledge structure of the students. The proposed measure is evaluated by (1) comparing it with the Dice coefficient for analyzing two sets of concepts; (2) analyzing its performance in a generic case of four CMAPs; and (3) applying it to a case using photovoltaic system as an example. In this case, results are intuitively obvious so that calculation results can be corroborated by human judgment. Finally, the concept similarity measure is applied to derive the similarity of CMAPs in the case study. Based on the initial evaluations, this study shows that the proposed measure has demonstrated promising features for determining the similarity of multiple knowledge structures or the common understanding of students. However, when the number of knowledge structures increases, concept similarity analyses become more complicated because uncertain situations arise due to ambiguous human perception to propositions that are shared by multiple concepts. Future research is thus needed to understand and clearly define concept similarity in those situations. In addition, other studies are also identified for future research.

A Recipe for Concept Similarity

Abstract: Sometimes your concept and mine have exactly the same content. When this is so, it is comparatively easy for me to understand what you say when you deploy your concept, for us to disagree, agree, and so on. But what if your concept and mine do not have exactly the same content? This question has occupied a number of philosophers, including Paul Churchland, Jerry Fodor, and Ernie Lepore. This paper develops a novel and rigorous measure of concept similarity, Proportion, such that concepts with different contents but sufficiently high Proportion scores will also conduce to understanding, agreement, and disagreement.

A Parameterized Representation of Uncertain Conceptual Spaces

AbstractMost conceptual modeling in geographic information science to date has used a symbolic approach with little or no recognition of the semantic uncertainty often found in geographic concepts. This work describes a concept model based on parameterized concept descriptions that uses a spatial metaphor, the conceptual space, as an organizing structure (Gärdenfors 2000). This cognitive theory of conceptual spaces is combined with a formal representation of semantic uncertainty based on rough fuzzy sets. The conceptual space then represents each concept as a collection of rough fuzzy property definitions with associated salience weights, where a property itself can be treated as a special case of a concept. Instead of explicitly defining concept hierarchies, we can allow different conceptual structures to emerge through measures of concept inclusion and similarity. A land use/land cover example demonstrates how the model represents concepts, concept similarity, hierarchical structures and the context dependence of concepts. The final section of the paper points to the need for further studies of context effects, concept similarity measures, and uncertainty representation using the proposed model.

The relation of knowledge organization to performance of a complex cognitive task

The relationship between organization of knowledge in memory and level of proficiency in the performance of a complex cognitive task was evaluated. Based on their performance in a scenario, 42 Electronic Warfare (EW) technicians were categorized into 3 performance levels. To measure the organization of knowledge, the EWs rated all possible pairs of 24 concepts on degree of relatedness, and a Pathfinder net was created for each subject and for 2 expert EWs. A correlation of 0.40 between performance scores and a concept similarity measure indicates a significant positive association between level of proficiency and knowledge organization. Both high and intermediate performance groups produced nets that were more similar to the expert net than were those of the low performance group. © 1998 John Wiley & Sons, Ltd.