Thesis ÁRBOLES GENEALÓGICOS MEDIANTE LÓGICA DIFUSA Y BASES DE DATOS DE GRAFOS
Date
2019-01
Authors
AVALO MIRANDA, JUAN FELIPE
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Una parte importante del estudio de la genealogía es la construcción de árboles genealógicos a partir de datos obtenidos desde diversos orígenes (registro civil, registros eclesiásticos, tradición familiar). Como provienen de fuentes heterogéneas es posible que detalles tales como nombres y apellidos tengan variantes que hacen más difícil una comparación directa. A su vez existe el problema de que los registros presenten errores de diversos tipos, tales como errores tipográficos, los cuales tienen como origen
la naturaleza humana al momento de ingresar datos. Para facilitar el trabajo del genealogista se diseñó e implementó una herramienta que usa lógica difusa para manejar la imprecision inherente en los datos genealógicos, y una base de datos de grafo (Neo4j) para su almacenamiento y posterior análisis. Se presenta un algoritmo de cuatro pasos que separa la búsqueda de árboles en una comparación por nombres, por tiempos, por relaciones de grafo y una posterior unión de registros hecha con el propósito de encontrar distintos individuos.
An important part of genealogic studies is getting a family tree starting from many diverse sources (civil registry, ecclesiastial registries, family tradition). Because of their heterogeneous quality it’s possible for registers like names and surnames to have variants that make a direct comparison difficult. There’s also the problem of data with errors of different types, like typographical errors that arise from human nature when entering them. To make the work of the genealogist easier a tool that uses fuzzy logic to manage the inherent imprecision in genealogical data, and uses a graph database (Neo4j) for their storage and analysis, will be designed and implemented. A four-step algorithm will be presented that separates the search for family trees in a name comparison, time comparison, graph relationship comparison and their following union designed to find different distinct individuals.
An important part of genealogic studies is getting a family tree starting from many diverse sources (civil registry, ecclesiastial registries, family tradition). Because of their heterogeneous quality it’s possible for registers like names and surnames to have variants that make a direct comparison difficult. There’s also the problem of data with errors of different types, like typographical errors that arise from human nature when entering them. To make the work of the genealogist easier a tool that uses fuzzy logic to manage the inherent imprecision in genealogical data, and uses a graph database (Neo4j) for their storage and analysis, will be designed and implemented. A four-step algorithm will be presented that separates the search for family trees in a name comparison, time comparison, graph relationship comparison and their following union designed to find different distinct individuals.
Description
Keywords
LOGICA DIFUSA , TEORIA DE GRAFOS , GENEALOGIA