Browsing by Author "Funk, Matthias"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
- A feature selection approach in the study of azorean proverbsPublication . Cavique, Luís; Mendes, Armando B.; Funk, Matthias; Santos, Jorge M. A.A paremiologic (study of proverbs) case is presented as part of a wider project based on data collected among the Azorean population. Given the considerable distance between the Azores islands, we present the hypothesis that there are significant differences in the proverbs from each island, thus permitting the identification of the native island of the interviewee, based on his or her knowledge of proverbs. In this chapter, a feature selection algorithm that combines Rough Sets and the Logical Analysis of Data (LAD) is presented. The algorithm named LAID (Logical Analysis of Inconsistent Data) deals with noisy data, and we believe that an important link was established between the two different schools with similar approaches. The algorithm was applied to a real world dataset based on data collected using thousands of interviews of Azoreans, involving an initial set of twenty-two thousand Portuguese proverbs.
- Logical Analysis of Inconsistent Data (LAID) for a paremiologic studyPublication . Cavique, Luís; Mendes, Armando B.; Funk, MatthiasA paremiologic (study of proverbs) case is presented as a part of a wider project, based on data collected by thousands of interviews made to people from Azores, and involving a set of twenty-two thousand Portuguese proverbs, where we searched for the minimum information needed to identify the birthplace island of an interviewee. The concept of birthplace was extended for all respondents that have lived in any locations more than 5 years,unintentionally introducing inconsistencies in the data classification task. The rough sets differ from classical sets by their ability to deal with inconsistent data. A parallel approach to data reduction is given by the logical analysis of data (LAD). LAD handicaps, like the inability to cope with the contradiction and the limited number of classification classes, will be overcome in this version of Logical Analysis of Inconsistent Data (LAID).
- Proverbs knowledge discovery in the virtual social network due to common knowledge of proverbsPublication . Mendes, Armando B.; Funk, Matthias; Cavique, LuísIn a series of interviews, it was collected a heterogeneous set of several million relations of positive and negative knowledge that a group of thousands of people has about a set of circa twenty-two thousand Portuguese Proverbs. This is a unique source for socio-cultural analysis of the mechanisms of transmission of oral culture in geographic discontinuous spaces. We present in this article some results on the problem of finding a homomorphism between proverbial knowledge and geographical locations. To find this relation, we chose an approach based on the Analysis of social networks where the broadcast of oral culture, at least historically, could be interpreted as a trace of direct social contact between some of their users. We can simply give the Hamming Distance between two people by comparing their proverbial knowledge and, then, choose for every person only those relations to a peer where this distance is minimal. The resulting graph is analysed by a new Clique Analysis procedure, proposed in this work, design to work on very dense networks. The procedure was tested on a subset of data and we found that there are clusters where the neighbourhood relation inducted by the minimum Hamming Distance could be a reflex of the geographical distribution and of some migration flux of the Azorean population. When we compare the cliques with high geographic proximity, we found some proverbs which are good discriminators between the different clusters.