<script type="text/javascript">
<!--
document.write('<div id="oa_widget"></div>');
document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=undefined&type=result"></script>');
-->
</script>
Este trabajo explora la utilidad de la aplicación de un enfoque de adquisición de conocimientos al problema de la clasificación de objetos arqueológicos, particularmente fragmentos cerámicos; uno de los materiales más abundantes en el registro arqueológico y cuya gestión implica mayor costo en términos de tiempo y recursos en el marco de cualquier investigación. Para ello se construyó un modelo tipológico implementando diversas herramientas de los principales paradigmas del aprendizaje de máquinas (aprendizaje no-supervisado y supervisado). El modelo fue implementado sobre un conjunto de datos de cerámicas arqueológicas del caribe colombiano. Los resultados revelan que el modelo funciona mejor cuando se cuenta con tipologías de referencia bien definidas y criterios clasificatorios estandarizados que puedan servir para entrenar algoritmos en tareas de aprendizaje supervisado. Por el contrario, una de sus mayores debilidades es su poca utilidad comparativa. Esto no solo se debe a las herramientas utilizadas sino a la calidad de la información disponible en el contexto de estudio
This work explores a knowledge acquisition and data analysis approach to the problem of archaeological classification. Particularly classification of pottery sherds; one of the most abundant and challenging elements of the archaeological record. To this end, a machine learning- based typology was developed applying different classification schemas; starting with an unstructured dataset (non-supervised paradigm) until the training of different supervised classifiers to automate the process of classifying new instances. This process was applied to a dataset from the Caribbean coast of Colombia to evaluate the accuracy and performance of different algorithms. The results reveal that, the model Works better with supervised classification tasks with consistent typologies that can work as training models. Also, the model shows a decent performance working with a categorical multivariate dataset. Conversely one of his major weaknesses is his lack of comparative utility. That's not only because the tools and procedures that was implemented but for the quality of the data used.
citations This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |