No Arabic abstract
In decision-support systems, the visual component is important for On Line Analysis Processing (OLAP). In this paper, we propose a new approach that faces the visualization problem due to data sparsity. We use the results of a Multiple Correspondence Analysis (MCA) to reduce the negative effect of sparsity by organizing differently data cube cells. Our approach does not reduce sparsity, however it tries to build relevant representation spaces where facts are efficiently gathered. In order to evaluate our approach, we propose an homogeneity criterion based on geometric neighborhood of cells. The obtained experimental results have shown the efficiency of our method.
The principal result of this work is the freeness in the $ overline{mathbb Z}_l$-cohomology of the Lubin-Tate tower. The strategy is of global nature and relies on studying the filtration of stratification of the perverse sheaf of vanishing cycles of some Shimura varieties of Kottwitz-Harris-Taylor types, whose graduates can be explicited as some intermediate extension of some local system constructed in the book of Harris andTaylor. The crucial point relies on the study of the difference between such extension for the two classical $t$-structures $p$ and $p+$. The main ingredients use the theory of derivative for representations of the mirabolic group.
We give a new definition, simpler but equivalent, of the abelian category of Banach-Colmez spaces introduced by Colmez, and we explain the precise relationship with the category of coherent sheaves on the Fargues-Fontaine curve. One goes from one category to the other by changing the t-structure on the derived category. Along the way, we obtain a description of the pro-etale cohomology of the open disk and the affine space, of independent interest.
Bitmap indexes are frequently used to index multidimensional data. They rely mostly on sequential input/output. Bitmaps can be compressed to reduce input/output costs and minimize CPU usage. The most efficient compression techniques are based on run-length encoding (RLE), such as Word-Aligned Hybrid (WAH) compression. This type of compression accelerates logical operations (AND, OR) over the bitmaps. However, run-length encoding is sensitive to the order of the facts. Thus, we propose to sort the fact tables. We review lexicographic, Gray-code, and block-wise sorting. We found that a lexicographic sort improves compression--sometimes generating indexes twice as small--and make indexes several times faster. While sorting takes time, this is partially offset by the fact that it is faster to index a sorted table. Column order is significant: it is generally preferable to put the columns having more distinct values at the beginning. A block-wise sort is much less efficient than a full sort. Moreover, we found that Gray-code sorting is not better than lexicographic sorting when using word-aligned compression.
The Knowledge Discovery in Database (KDD) process permits the detection of pattern in databases, where this analysis may be compromised if database is not consistent, making necessary the use of data cleaning techniques. This paper presents a tool based in similarity functions to help the preprocessing of databases and it behaved efficiently in the standardization of a System of Public Security of the State of Para database and may be reused with other databases and other data mining projects.
We introduce an algebra given by quadratic relations in an algebra of polynomials in an infinite number of variables. Using this algebra, we prove some explicit formulas for the Sturm sequence of a polynomial.