2 resultados para Citation classification schemes
em Universidad de Alicante
Resumo:
El análisis de citas bibliográficas que usa variaciones de métodos de conteo provoca deformaciones en la evaluación del impacto. Para enriquecer el cálculo de los factores de impacto se necesita entender el tipo de influencia de los aportes de un investigador sobre el autor que los menciona. Para ello, se requiere realizar análisis de contenido del contexto de las citas que permita obtener su función, polaridad e influencia. El presente artículo trata sobre la definición de un esquema de anotación tendiente a la creación de un corpus de acceso público que sea la base de trabajo colaborativo en este campo, con miras al desarrollo de sistemas que permitan llevar adelante tareas de análisis de contenido con el objetivo planteado.
Resumo:
Prototype Selection (PS) algorithms allow a faster Nearest Neighbor classification by keeping only the most profitable prototypes of the training set. In turn, these schemes typically lower the performance accuracy. In this work a new strategy for multi-label classifications tasks is proposed to solve this accuracy drop without the need of using all the training set. For that, given a new instance, the PS algorithm is used as a fast recommender system which retrieves the most likely classes. Then, the actual classification is performed only considering the prototypes from the initial training set belonging to the suggested classes. Results show that this strategy provides a large set of trade-off solutions which fills the gap between PS-based classification efficiency and conventional kNN accuracy. Furthermore, this scheme is not only able to, at best, reach the performance of conventional kNN with barely a third of distances computed, but it does also outperform the latter in noisy scenarios, proving to be a much more robust approach.