1,721,006 research outputs found
Different fuzzy cluster validity indexes for the evaluation of the quality of the resulting partitioning
No full textFuzzy clustering is useful to mine complex and multi-dimensional datasets, where the members have partial or fuzzy relations. The procedure of evaluating the results of a fuzzy clustering algorithm is known under the term cluster validity. There are three principal approaches to investigate cluster validity: external, internal and relative criteria (Theodoridis and Koutroubas, 1999). These methods give an indication of the quality of the resulting partitioning and, so, they can be considered as an instrument available to experts in order to assess the results of fuzzy clustering. In this work we apply some indexes to evaluate the quality of the results obtained from a case study
Data integration techniques for the measurement of the reliability of sample variables
No full textThe objective of the present study is to investigate the possibility of developing an integrated database with information pertaining to the income of Italian families arising from two major surveys conducted by ISTAT (EU-SILC) and the Bank of Italy (household income survey). Since neither of the surveys has the scope to allow for the construction of a database of information pertaining to income, an integration has been sought between the data from the two archives, assuming that the surveys are reliable in terms of the accuracy of the sample design and control of the representativeness of the sample. The development of our analysis is primarily aimed at carrying out an in-depth comparison between the two surveys in terms of structure, definition of variables and sample homogeneity and secondly, through the use of an integrated dataset, at a verification measurement of the validity of the information, in particular, of the income component
Localizzazione di aree urbane a rischio di povertà: metodi di zonizzazione di dati territoriali
No full text
Metodi di individuazione di hot spot di disagio abitativo per pianificare la rigenerazione urbana
No full textStatistical Methods for Spatial Planning and Monitoring
No full textThe aim of this paper is to identify territorial areas and/or population subgroups characterized by situations of deprivation or strong social exclusion through a fuzzy approach that allows the definition of a measure of the degree of belonging to the disadvantaged group. Grouping methods for territorial units are employed for areas with high (or low) intensity of the phenomenon by using clustering methods that permit the aggregation of spatial units that are both contiguous and homogeneous with respect to the phenomenon under study. This work aims to compare two different clustering methods: the first based on the technique of SaTScan [Kuldorff: A spatial scan statistics. Commun. Stat.: Theory Methods 26, 1481–1496 (1997)] and the other based on the use of Seg-DBSCAN, a modified version of DBSCAN [Ester et al.: A density-based algorithm for discovering clusters in large spatial databases with noise. In: Proceeding of the 2nd International Conference on Knowledge Discovery and Data Mining, pp. 94–99 (1996)]
Analysis of Positional Aspects in the Variation of Real Estate Values in an Italian Southern Metropolitan Area
No full textThe paper show the use of a fuzzy weighting system to identify the correspondence of real estate value with main socio-physical characters of the urban tissue. The descriptor of the relationship with the real estate value is represented by a set of indicators of the urban decay of housing property and the analysis is tested on a real application of a case study. The study gives support to the development of new approach for localizing cadastral values at a more detailed scale, compared to the current scale used in the Italian Cadastre. The utilized statistical approach has been based on the SaTScan application, as a techniques of fuzzy clustering, and on a test of stability based on the comparison of a “fuzzy semantic distance” among the average real estate values of urban quarters, with the expected crisp distance among the same quarter
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
