Dissimilarity learning for nominal data

Cheng, Viviana

doi:10.1016/s0031-3203(04)00041-x

Cited by 8 publications

(12 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, this metric implies loss of information because it considers that all attributes values are of equal distance from each other, not taking into account different degrees of similarity. There exist alternative dissimilarity measures such as the Value Difference Metric (VDM) (Wilson and Martinez 1997), and Adaptive Dissimilarity Matrix (ADM) (Cheng et al 2004), among others. For the VDM two feature values are considered to be closer if they have similar classifications (i.e.…”

Section: Rbfns and The Classification Problem Frameworkmentioning

confidence: 99%

“…On the other hand, ADM takes into account the possible correlation between the attributes. In Cheng et al (2004), the RBFN model is used to test the efficiency of the OM, VDM and ADM dissimilarity measures and it is demonstrated that VDM and ADM outperform OM. It is more difficult to conclude which dissimilarity measure is the most efficient when comparing VDM and ADM results.…”

Section: Rbfns and The Classification Problem Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

CO2RBFN: an evolutionary cooperative–competitive RBFN design algorithm for classification problems

et al. 2009

View full text Add to dashboard Cite

This paper presents a new evolutionary cooperative-competitive algorithm for the design of Radial Basis Function Networks (RBFNs) for classification problems. The algorithm, CO 2 RBFN, promotes a cooperativecompetitive environment where each individual represents a radial basis function (RBF) and the entire population is responsible for the final solution. The proposal considers, in order to measure the credit assignment of an individual, three factors: contribution to the output of the complete RBFN, local error and overlapping. In addition, to decide the operators' application probability over an RBF, the algorithm uses a Fuzzy Rule Based System. It must be highlighted that the evolutionary algorithm considers a distance measure which deals, without loss of information, with differences between nominal features which are very usual in classification problems. The precision and complexity of the network obtained by the algorithm are compared with those obtained by different soft computing methods through statistical tests. This study shows that CO 2 RBFN obtains RBFNs with an appropriate balance between accuracy and simplicity, outperforming the other methods considered.

show abstract

Section: Rbfns and The Classification Problem Frameworkmentioning

confidence: 99%

Section: Rbfns and The Classification Problem Frameworkmentioning

confidence: 99%

CO2RBFN: an evolutionary cooperative–competitive RBFN design algorithm for classification problems

et al. 2009

View full text Add to dashboard Cite

show abstract

“…Of course, one could define a degree of quasi-oppositeness and determine the slope of gradual increase/decrease based on problem specifications. This, however, will fall into the scope of similarity/dissimilarity measures, which has been extensively investigated for many other research fields [5,10,19,20].…”

Section: Definition 4 (Type-i Quasi-oppositionmentioning

confidence: 99%

“…For example, the well-known bisection method [4] for solving equations can be considered an I-OBC algorithm since it shrinks the search interval by looking at positive versus negative sign change. As well, using Bayes theorem [2] is an implicit usage of oppositional relationship between the conditional probabilities p(A|B) and p(B|A) 5 . A different, stronger version of implicit incorporation of oppositional concepts is Bayesian Yin-Yang Harmony Learning which is based on the alternative viewpoints of Bayes Rule (Chapter 10).…”

Section: Definition 11 (Implicit Obc Algorithms) Any Algorithm That mentioning

confidence: 99%

Opposition-Based Computing

Tizhoosh

Ventresca

Rahnamayan

Studies in Computational Intelligence

View full text Add to dashboard Cite

Summary. Diverse forms of opposition are already existent virtually everywhere around us but the nature and significance of oppositeness is well understood only in specific contexts within the fields of philosophy, linguistics, psychology, logic and physics. The interplay between entities and opposite entities is apparently fundamental for maintaining universal balance. However, it seems that there is a gap regarding oppositional thinking in engineering, mathematics and computer science. Although many opposition-based techniques exist in these fields, the oppositional properties they employ are not usually directly studied. A better understanding of opposition could potentially establish new search, reasoning, optimization and learning schemes with a wide range of applications. For instance, improving convergence rate for hyperdimensional problems could be improved through the use of oppositional strategies.A large number of problems in engineering and science cannot be approached with conventional schemes and are generally handled with intelligent techniques such as evolutionary, neural, reinforcing and swarm-based techniques. These methodologies, however, suffer from high computational costs. In this work, the outlines of opposition-based computing, a proposed framework for computational intelligence, will be introduced. The underlying idea is simultaneous consideration of guess and opposite guess, estimate and opposite estimate, viewpoint and opposite viewpoint and so on in order to make better decisions in a shorter time, something that all aforementioned techniques could benefit from. The goal is to better understand the role of opposition within a computational intelligence context with the intention of improving existing or developing more powerful and robust approaches to handle complex problems.Due to its diversity, it is difficult, if not impossible, to uniquely and universally define the nature and the scope of what we call opposition. However, there is no doubt that we need a mathematical formalism if we are going to, at least to some degree, exploit the oppositional relationships in real-world systems. Hence, we attempt to establish a generic framework for computing with opposites in this chapter, a framework which may not mathematically capture the very essence of oppositeness in all systems accurately but it will be, as we hope, a point of departure for moving toward opposition-based computing. The ideas put forward in this chapter may not be shared by all those who work on some aspect of opposition (and usually without labeling it opposition), nevertheless, we believe that this framework can be useful for understanding and employing opposition.

show abstract

“…However, for nominal attributes (such as rank information), definitions of (dis)similarity becomes non-trivial [55]. A commonly used approach is overlap metric [202], where } for two possible values v, and v., the distance is assigned as zero when v. and vy are identical and one if they are different.…”

Section: Determining Similaritymentioning

confidence: 99%

A generalized multidimensional index structure for multimedia data to support content-based similarity searches in a collaborative search environment

Chen

Chetterjee

View full text Add to dashboard Cite

Chatterjee, Kasturi, "A generalized multidimensional index structure for multimedia data to support content-based similarity searches in a collaborative search environment" (2010 Since multimedia data, such as images and videos, are way more expressive and informative than ordinary text-based data, people find it more attractive to communicate and express with them. Additionally, with the rising popularity of social networking tools such as Facebook and Twitter, multimedia information retrieval can no longer be considered a solitary task. Rather, people constantly collaborate with one another while searching and retrieving information. But the very cause of the popularity of multimedia data, the huge and different types of information a single data object can carry, makes their management a challenging task. Multimedia data is commonly represented as multidimensional feature vectors and carry high-level semantic information. These two characteristics make them very different from traditional alpha-numeric data. Thus, to try to manage them with frameworks and rationales designed for primitive alpha-numeric dat, will be ineflicient.An index structure is the backbone of any database management system.. It has been seen that index structures present in existing relational database management frameworks cannot handle multimedia data effectively. Thus, ill this dissertation, a generalized multidimensional index structure is proposed which accommodates the atypical multidimensional representation and the semantic information carried by different multimedia data seamlessly from within one single framework. Additionally, the dissertation vi investigates the evolving relationships among multimedia data in a collaborative environment and how such information can help to customize the design of the proposed index structure, when it is used to manage multimedia data in a shared environment.Extensive experiments were conducted to present the usability and better performance of the proposed framework over current state-of-art approaches.

show abstract

Dissimilarity learning for nominal data

Cited by 8 publications

References 10 publications

CO2RBFN: an evolutionary cooperative–competitive RBFN design algorithm for classification problems

CO2RBFN: an evolutionary cooperative–competitive RBFN design algorithm for classification problems

Opposition-Based Computing

A generalized multidimensional index structure for multimedia data to support content-based similarity searches in a collaborative search environment

Contact Info

Product

Resources

About