92 research outputs found
Uniformly Generating Distribution Functions for Discrete Random Variables
No full textAn algorithm is presented which efficiently solves the problem of uniformity generating distribution functions for an n-valued random variabl
Nomen est Omen: Analyzing the Language of Function Identifiers
No full textThe identifiers chosen by programmers as function names contain valuable information. They are often the starting point for the program understanding activities, especially if high level views, like the call graph, are available.
In this paper the lexical, syntactical and semantical structure of function identifiers is analyzed by means of a segmentation technique, a regular language and a conceptual classification. The application of these analyses to a database of procedural programs suggested some potential uses of the results, ranging from the support to program understanding, to the evolution toward a standard and more maintainable for
Restructuring Program Identifier Names
No full textThe identifiers chosen by the programmers as entity names contain valuable information. They are often the starting point for the program understanding activities, especially when high level views, like the call graph, are available.
In this paper an approach for the restructuring of program identifier names is proposed, aimed at improving their meaningfulness. It considers two forms of standardization, associated respectively to the lexicon of the composing terms and to the syntax of their arrangement. Automatic and semiautomatic techniques are described which can help the restructuring intervention. Their application to a real world case study is also presente
Circuit Line- Problema 1. Studio preliminare
No full textIl lavoro qui esposto si inquadra nella collaborazione recentemente avviata con la ditta Circuit Line di Verona, e si riferisce, in particolare al problema di determinare una trasformazione rigida in grado di correggere le difformita' tra le posizioni nominali e quelle effettive dei punti di test su PCB (Problema 1). Il problema viene innanzitutto succintamente esposto, introducendo la necessaria nomenclatura. Una procedura di risoluzione viene quindi proposta e delineata, ed i singoli passi descritti in dettaglio sulla base delle prove effettuate e delle evidenze sperimentali raccolt
Highlighting Hard Patterns via AdaBoost Weights Evolution
No full textThe dynamical evolution of weights in the Adaboost algorithm contains useful information about the rôle that the associated data points play in the built of the model. In particular, the dynamics induces a bipartition of the data set into two (easy/hard) classes. Easy points are ininfluential in the making of the model, while the varying relevance of hard points can be gauged in terms of an entropy value associated to their evolution. Smooth approximations of entropy highlight regions where classification is most uncertain. Promising results are obtained when methods proposed are applied in the Optimal Sampling framewor
2D Deformable Models for Visual Speech Analysis
No full textA scheme for describing the mouth of the speaker in colour image sequence is proposed which is based on a parametric 2D model of the lips. Key information for the parameters estimation is extracted from chrominance analysis. A detailed description of the techniques employed is given, and some preliminary results are show
The Dynamics of AdaBoost Weights Tells You What's hard to Classify
No full textThe dynamical evolution of weights in the AdaBoost algorithm contains useful information about the role that the associated data points play in the built of the AdaBoost model. In particular, the dynamics induces a bipartition of the data set into two (easy/hard) classes. Easy points are ininfluential in the making of the model, while the varying relevance of hard points can be gauged in terms of an entropy value associated to their evolution. Smooth approximations of entropy highlight regions where classification is most
uncertain. Promising results are obtained when methods proposed are applied in the Optimal Sampling framewor
Exact Bagging with k-Nearest Neighbour Classifiers
No full textA formula is derived for the exact computation of Bagging classifiers when the base model adopted is k-Nearest Neighbour (k-NN). The formula, that holds in any dimension and does not require the extraction of bootstrap replicates, proves that Bagging cannot improve 1-Nearest Neighbour. It also proves that, for k > 1, Bagging has a smoothing effect on k-NN. Convergence of empirically bagged k-NN predictors to the exact formula is also considered. Efficient approximations to the exact formula are derived, and their applicability to practical cases is illustrate
Exact Bagging with k-Nearest Neighbour Classifiers
No full textA formula is exhibited for the exact computation of Bagging classifiers when the base model adopted is k-Nearest Neighbour (k-NN). The formula holds in any dimension, does not require the extraction of bootstrap replicates, and yields an implementation of Bagging that is as fast as the computation of a single k-NN classifier. It also shows that Bagging with 1-Nearest Neighbour is perfectly equivalent to plain 1-N
- …
