nÃo hà / A desigualdade de renda no Brasil diminuiu nos Ãltimos anos. DÃvidas, entretanto, pairam sobre a universalizaÃÃo dessa queda em regiÃes ou Ãreas especÃficas, especialmente naquelas onde se verificam elevados Ãndices de analfabetismo, parcela significativa da populaÃÃo à pobre e/ou as oportunidades de empregos sÃo escassas sobretudo para mÃo de obra nÃo qualificada e o consequente fraco desenvolvimento econÃmico, ambiente propÃcio para a mà distribuiÃÃo de renda. Esta pesquisa se propÃs mostrar a evoluÃÃo da desigualdade na distribuiÃÃo da renda domiciliar per capita, medida pelo Ãndice de Gini e detectar quais foram as parcelas do rendimento mensal domiciliar que contribuÃram para a desigualdade, alÃm de determinar a contribuiÃÃo percentual dos componentes do rendimento domiciliar. Utilizou-se como metodologia a decomposiÃÃo do Ãndice de Gini em fontes de rendimentos. A base de dados utilizada teve como fonte a Pesquisa Nacional por Amostra de DomicÃlios (PNAD), do perÃodo de 1997 a 2007. O meio rural da RegiÃo Nordeste nÃo metropolitana, do Estado do Cearà e da RegiÃo Metropolitana de Fortaleza foram as Ãreas de estudo desta pesquisa por se enquadrarem nesse contexto. Os resultados mostraram que a desigualdade de renda caiu, nÃo de forma permanente, na RegiÃo Nordeste. Nos outros dois espaÃos analisados, ela oscilou ao longo do perÃodo sem muitas mudanÃas. As parcelas como Outros Trabalhos, Aposentadorias e PensÃes Oficiais reforÃaram a desigualdade nas trÃs Ãreas geogrÃficas e os programas sociais de transferÃncia de renda aos mais carentes tiveram significativa participaÃÃo nas reduÃÃes contÃnuas do Ãndice de Gini, especialmente no Nordeste rural. / Income inequality in Brazil declined in recent years. Doubts, however, weaken the universalization of this fall in regions or specific areas, especially where there are high rates of illiteracy, a significant portion of the population is poor and / or job opportunities are scarce especially for labor unskilled and the consequent weak economic development, enabling environment for the maldistribution of income. This research aims to show the evolution of inequality in the distribution of household income per capita, measured by the Gini index and to detect what were the shares of household income contributed to inequality, and to determine the percentage contribution of the components of household income. Was used as a methodology to decompose the Gini index of income sources. The database used as a source the National Household Sample Survey (PNAD), the period from 1997 to 2007. The rural areas of the Northeast not metropolitan, the State of Cearà and the Metropolitan Region of Fortaleza were the areas of this research, since they fit in this context. The results showed that income inequality has fallen, not permanently, in the Northeast. In the other two areas studied, it varied throughout the period without many changes. The plots and Other Works, Retirement and Pensions officials reinforced the inequality in three geographical areas and social programs to transfer income to the poorest countries have had significant participation in the continual reduction of the Gini index, especially in the rural Northeast.

Made available in DSpace on 2016-04-04T18:36:44Z (GMT). No. of bitstreams: 1 Jose Jorge Meschiatti Nogueira.pdf: 407232 bytes, checksum: 2e359a68104f4d773b63582ba4e7c86a (MD5) Previous issue date: 2004-02-17 / This dissertation describes the connection of Information Science with Economics in the environment of social exclusion, outlining the historical scenario of these two sciences. It inserts the economic environment of social and information exclusion in a context, and classifies them under various denominations provided by literature. Finally, it applies the empiric methodology based on econometric, which uses data from technical magazines and books, provided by IBGE, through Family Budget Research Pesquisa de Or?amento Familiar (POF). In order to do so, it takes as an evolution comparative basis the 1987 and 1996 POF results, in which it is focused from several perspectives the distribution and concentration of information under the income and education viewpoints. It measures the consumption patterns of the wealthier and poorer levels, as well as the education levels. / Esta disserta??o retrata a liga??o da Ci?ncia da Informa??o com a Economia no ambiente da exclus?o social, tra?ando um panorama hist?rico do encontro destas duas ci?ncias. Contextualiza o ambiente econ?mico da exclus?o social e informacional, conceituando-as sob as diversas denomina??es fornecidas na literatura. Por fim, aplica metodologia emp?rica baseada na econometria, utilizando-se dos dados de consumo de livros e revistas t?cnicas, fornecidos pelo IBGE, por interm?dio da sua Pesquisa de Or?amento Familiar (POF). Para tanto, toma como base comparativa de evolu??o a POF de 1987 e a de 1996, enfocando sob diversos aspectos a distribui??o e concentra??o de informa??o sob a ?tica da renda e da instru??o, na mensura??o das diferen?as de consumo entre os estratos mais ricos e os mais pobres, os mais instru?dos e os menos instru?dos.

La notion de pauvreté a fait l’objet de nombreux débats à travers le monde. Les premières analyses ont mis l’accent sur le caractère unidimensionnel basé essentiellement sur une approche monétaire (revenu ou dépense de consommation). C’est à la faveur des travaux de certains auteurs, tels que Townsend et Sen, que le caractère multidimensionnel est mis en lumière, au regard de la difficulté de quantification de certaines variables qui traduisaient l’idée de manque. l ressort des nombreuses études empiriques que le secteur rural reste le secteur le plus touché par le phénomène de pauvreté.Dans le cas de la Côte d’Ivoire, la pauvreté reste également un phénomène rural et la plupart des études réalisées se sont focalisées sur l’approche monétaire. Cette étude aborde l’analyse de la pauvreté en se focalisant sur le milieu rural agricole et appréhende le phénomène de la pauvreté à partir de trois (03) approches : (i) monétaire ; (ii) privation relative et (iii) patrimoine.Les résultats montrent que la pauvreté reste importante dans ce secteur avec un taux plus élevé pour l’indicateur de privation relative. En outre, il existe une inégalité monétaire plus importante que les autres types de pauvreté quel que soit l’année (2002 et 2008).L’identification des facteurs explicatifs de l’appartenance ou non à la classe des pauvres, montre que les variables liées au genre, au type de religion et à la classe d’âge sont les plus communes aux différentes années et aux différentes approches.Sur la base des résultats obtenus, les recommandations suivantes sont formulées : (i) à l’endroit du gouvernement ivoirien, utiliser les approches monétaires et non-monétaires dans les prochaines analyses sur la pauvreté en Côte d’Ivoire ; mettre l’accent sur la construction de nouvelles infrastructures et l’achat de nouveaux équipements ; améliorer la communication relatives aux actions du gouvernement ; renforcer les capacités des producteurs en matière d’utilisation d’intrants améliorés et d’outils pertinents ; (ii) à l’endroit des producteurs agricoles, adopter les technologies et techniques agricoles et des intrants de qualité, accepter d’adhérer à des entreprises coopératives ; (iii) à l’endroit des coopératives, il faut rechercher des débouchés pour ses membres, négocier de meilleures rémunérations des productions agricoles, transformer les agriculteurs en véritables entrepreneurs agricoles. / The notion of poverty has been the subject of much debate around the world. Previous analyses have emphasized the one-dimensional character based essentially on a monetary approach (income or consumption expenditure). Thanks to the work of some authors such as Townsend and Sen, the multidimensional character is highlighted, considering the difficulty of quantifying certain variables that translated the idea of lack. In addition, many empirical studies show that the rural sector remains the most affected by poverty.In the case of Côte d'Ivoire, poverty is also a rural phenomenon and most studies have focused on the monetary approach.This study addresses the analysis of poverty by focusing on rural farming and apprehends the phenomenon of poverty from three (03) approaches: (i) monetary; (ii) relative deprivation and (iii) wealth.The results show that poverty remains important in this sector with a high rate for the indicator of the relative deprivation. In addition, there is greater monetary inequality than other types of poverty whatever the year (2002 and 2008).The identification of the explanatory factors of the membership or not in the class of the poor shows that the variables related to the gender, to the type of religion and the age group are the most common in the various years and various approaches.Based on the obtained results , the following recommendations are formulated: (i) towards the Ivorian government, use monetary and non-monetary approaches in next analyses on poverty in Côte d’Ivoire; emphasize the construction of new infrastructures and the purchase of new equipment; improve communication about government actions; build the capacity of producers to use improved inputs and relevant tools; (ii) to agricultural producers, adopt agricultural technologies and techniques and quality inputs, accept to join cooperative enterprises; (iii) for cooperatives, it is necessary to look for outlets for its members, to negotiate better remunerations for agricultural productions, to transform farmers into real agricultural entrepreneurs.

Nous proposons dans cette thèse la méthode STREE de construction d'arbres de décision avec des données symboliques. Ce type de données permet de caractériser des individus de niveau supérieur qui peuvent être des classes ou catégories d’individus ou des concepts au sens des treillis de Galois. Les valeurs des variables, appelées variables symboliques, peuvent être des ensembles, des intervalles ou des histogrammes. Le critère de partitionnement récursif est une combinaison d'un critère par rapport aux variables explicatives et d'un critère par rapport à la variable à expliquer. Le premier critère est la variation de la variance des variables explicatives. Quand il est appliqué seul, STREE correspond à une méthode descendante de classification non supervisée. Le second critère permet de construire un arbre de décision. Il s'agit de la variation de l'indice de Gini si la variable à expliquer est nominale et de la variation de la variance si la variable à expliquer est continue ou bien est une variable symbolique. Les données classiques sont un cas particulier de données symboliques sur lesquelles STREE peut aussi obtenir de bons résultats. Il en ressort de bonnes performances sur plusieurs jeux de données UCI par rapport à des méthodes classiques de Data Mining telles que CART, C4.5, Naive Bayes, KNN, MLP et SVM. STREE permet également la construction d'ensembles d'arbres de décision symboliques soit par bagging soit par boosting. L'utilisation de tels ensembles a pour but de pallier les insuffisances liées aux arbres de décisions eux-mêmes et d'obtenir une décision finale qui est en principe plus fiable que celle obtenue à partir d'un arbre unique. / In this thesis, we propose the STREE methodology for the construction of decision trees with symbolic data. This data type allows us to characterize individuals of higher levels which may be classes or categories of individuals or concepts within the meaning of the Galois lattice. The values of the variables, called symbolic variables, may be sets, intervals or histograms. The criterion of recursive partitioning is a combination of a criterion related to the explanatory variables and a criterion related to the dependant variable. The first criterion is the variation of the variance of the explanatory variables. When it is applied alone, STREE acts as a top-down clustering methodology. The second criterion enables us to build a decision tree. This criteron is expressed as the variation of the Gini index if the dependant variable is nominal, and as the variation of the variance if thedependant variable is continuous or is a symbolic variable. Conventional data are a special case of symbolic data on which STREE can also get good results. It has performed well on multiple sets of UCI data compared to conventional methodologies of Data Mining such as CART, C4.5, Naive Bayes, KNN, MLP and SVM. The STREE methodology also allows for the construction of ensembles of symbolic decision trees either by bagging or by boosting. The use of such ensembles is designed to overcome shortcomings related to the decisions trees themselves and to obtain a finaldecision that is in principle more reliable than that obtained from a single tree.