Showing papers on "Ward's method published in 2013"

MAPSOM: User Involvement in Ontology Matching

Abstract: This paper presents a semi-automatic similarity aggregating system for ontology matching problem. The system consists of two main parts. The first part is aggregation of similarity measures with the help of self-organizing map. The second part incorporates user feedback for refining self-organizing map outcomes. The system calculates different similarity measures e.g., string-based similarity measure, WordNet-based similarity measure... to cover different causes of semantic heterogeneity. The next step is similarity aggregation by means of the self-organizing map and the ward clustering. The final step is the active learning phase for results tuning. We implemented this idea as MAPSOM framework. Our experimental results show that MAPSOM framework can be used for problems where the highest precision is needed.

Financial performance analysis of European banks using a fuzzified Self-Organizing Map

Abstract: Due to the recent wave of bank failures, stress tests have been conducted on banks within the European Union. The stress tests, however, only consider the adequacy of a bank's capital ratios, whereas the general financial performance of individual banks is disregarded. In this paper, we use the Self-Organizing Map SOM to perform a visual multidimensional and temporal financial performance analysis of European banks. We address several problems concerning financial performance analysis. We deal with the problem of selecting suitable financial ratios by performing dimensionality reduction using Principal Component Analysis. We also deal with difficult data using outlier trimming and normalization techniques, and use the SOM for imputing missing values. We use a decision-framework for choosing the final model, based upon a set of map and clustering quality measures. In addition, we implement a second-level fuzzified Ward clustering for visualization purposes and for assessing the crispness of the solution. The result is a visual SOM model for financial performance analysis of European banks.

A Fuzzy Approach to Ward’s Method of Classification: An Application Case to the Italian University System

Abstract: A great part of statistical techniques has been thought for exact numerical data, although available information is often imprecise, partial, or not expressed in truly numerical terms. In these cases the use of fuzzy numbers can be seen as an appropriate way for a more effective representation of observed data. Diamond introduced a metrics into the space of triangular fuzzy numbers in the context of a simple linear regression model; in this work we suggest a multivariate generalization of such a distance between trapezoidal fuzzy numbers to be used in clustering techniques. As an application case of the proposed measure of dissimilarity, we identify homogeneous groups of Italian universities according to graduates’ opinion (itself fuzzy) on many aspects concerning internship activities, by disciplinary area of teaching. Since such an opinion depends not only on the quality of internships, but also on the local context within which the activity is carried out, the obtained clusters are analyzed paying attention particularly to the membership of each university to Northern, Central, or Southern Italy. [This work is the result of joint reflections by the authors, with the following contributions attributed to Campobasso (Sects. 2.2, 2.3.2 and 2.4), and to Fanizzi (Sects. 2.1, 2.3 and 2.3.1).]

Active Learning based on Hierarchical Clustering

Abstract: Active learning aims to improve the performance of a classification model by repeating the process to select the most helpful unlabeled data and include it to the training set through labelling by expert. In this paper, we propose a method for active learning based on hierarchical agglomerative clustering using Ward`s linkage. The proposed method is able to construct a training set actively so as to include at least one sample from each cluster and also to reflect the total data distribution by expanding the existing training set. While most of existing active learning methods assume that an initial training set is given, the proposed method is applicable in both cases when an initial training data is given or not given. Experimental results show the superiority of the proposed method.

Оценка инновационного потенциала экономических систем

Abstract: In article the algorithm of an assessment of innovative potential is presented, classification of the factors influencing its formation is given. The short review of existing techniques of an assessment of innovative capacity of economic systems is carded out. Merits and demerits of each of techniques are revealed. Were selected for analysis are often used for the evaluation of innovative potential of economic systems methods. During the analysis, the evaluation stages, describes the factors used in the assessment and identified advantages and disadvantages of each of the following methods: indicator method, the expertise and the method of cluster analysis. Choice of method for the innovation potential of the economic system, it is necessary to consider the advantages and disadvantages of each of them and select the most appropriate for a particular case evaluation method.