The application of adaptive partitioned random search in feature selection problem

Feature selection is one of important and frequently used techniques in data preprocessing. It can improve the efficiency and the effectiveness of data mining by reducing the dimensions of feature space and removing the irrelevant and redundant information. Feature selection can be viewed as a global optimization problem of finding a minimum set of M relevant features that describes the dataset as well as the original N attributes. In this paper, we apply the adaptive partitioned random search strategy into our feature selection algorithm. Under this search strategy, the partition structure and evaluation function is proposed for feature selection problem. This algorithm ensures the global optimal solution in theory and avoids complete randomness in search direction. The good property of our algorithm is shown through the theoretical analysis.

Predictors of word-level literacy amongst Grade 3 children in five diverse languages

Groups of Grade 3 children were tested on measures of word-level literacy and undertook tasks that required the ability to associate sounds with letter sequences and that involved visual, auditory and phonological-processing skills. These groups came from different language backgrounds in which the language of instruction was Arabic, Chinese, English, Hungarian or Portuguese. Similar measures were used across the groups, with tests being adapted to be appropriate for the language of the children. Findings indicated that measures of decoding and phonological-processing skills were good predictors of word reading and spelling among Arabic- and English-speaking children, but were less able to predict variability in these same early literacy skills among Chinese- and Hungarian-speaking children, and were better at predicting variability in Portuguese word reading than spelling. Results were discussed with reference to the relative transparency of the script and issues of dyslexia assessment across languages. Overall, the findings argue for the need to take account of features of the orthography used to represent a language when developing assessment procedures for a particular language and that assessment of word-level literacy skills and a phonological perspective of dyslexia may not be universally applicable across all language contexts. Copyright (C) 2008 John Wiley & Sons, Ltd.

The mu-way intersection problem for m-cycle systems

An m-cycle system of order upsilon is a partition of the edge-set of a complete graph of order upsilon into m-cycles. The mu -way intersection problem for m-cycle systems involves taking mu systems, based on the same vertex set, and determining the possible number of cycles which can be common to all mu systems. General results for arbitrary m are obtained, and detailed intersection values for (mu, m) = (3, 4), (4, 5),(4, 6), (4, 7), (8, 8), (8, 9). (For the case (mu, m)= (2, m), see Billington (J. Combin. Des. 1 (1993) 435); for the case (Cc,m)=(3,3), see Milici and Quattrochi (Ars Combin. A 24 (1987) 175. (C) 2001 Elsevier Science B.V. All rights reserved.