| Título |
Enhancement of Performance of Document Clustering in the Authorship Identification Problem with a Weighted Cosine Similarity |
| Tipo |
Congreso |
| Sub-tipo |
SCOPUS |
| Descripción |
17th Mexican International Conference on Artificial Intelligence, MICAI 2018 |
| Resumen |
Distance and similarity measures are essential to solve many pattern recognition problems such as classification, information retrieval and clustering, where the use of a specific distance could led to a better performance than others. A weighted cosine distance is proposed considering a variation in the weights of exclusive attributes of the input vectors. An agglomerative hierarchical clustering of documents was used for the comparison between the traditional cosine similarity and the one proposed in this paper. This modified measure has outcome in an improvement in the formation of clusters. |
| Observaciones |
DOI: 10.1007/978-3-030-04497-8_4
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), v. 11289 |
| Lugar |
Guadalajara |
| País |
Mexico |
| No. de páginas |
49–56 |
| Vol. / Cap. |
|
| Inicio |
2018-10-22 |
| Fin |
2018-10-27 |
| ISBN/ISSN |
9783030044961 |
