%0 Conference Paper %1 reyes-pena_document_2020 %A Reyes-Peña, Cecilia %A Tovar Vidal, Mireya %A Lavalle Martínez, José de Jesús %B Proceedings of the Future Technologies Conference (FTC) 2019 %C Cham %D 2020 %E Arai, Kohei %E Bhatia, Rahul %E Kapoor, Supriya %I Springer International Publishing %K clustering %P 610--617 %R 10.1007/978-3-030-32520-6_44 %T Document clustering by relevant terms : an approach %X In this work, a document clustering based on relevant terms into an untagged medical text corpus approach is presented. To achieve this, to create a list of documents containing each word is necessary. Then, for relevant term extraction, the frequency of each term is obtained in order to compute the word weight into the corpus and into each document. Finally, the clusters are built by mapping using main concepts from an ontology and the relevant terms (only subjects), assuming that if two words appear in the same documents these words are related. The obtained clusters have a category corresponding to ontology concepts, and they are measured with cluster from K-Means (assuming the k-Means cluster were well formed) using the Overlap Coefficient and obtaining 70\% of similarity among the clusters. %@ 978-3-030-32520-6

@inproceedings{reyes-pena_document_2020, abstract = {In this work, a document clustering based on relevant terms into an untagged medical text corpus approach is presented. To achieve this, to create a list of documents containing each word is necessary. Then, for relevant term extraction, the frequency of each term is obtained in order to compute the word weight into the corpus and into each document. Finally, the clusters are built by mapping using main concepts from an ontology and the relevant terms (only subjects), assuming that if two words appear in the same documents these words are related. The obtained clusters have a category corresponding to ontology concepts, and they are measured with cluster from K-Means (assuming the k-Means cluster were well formed) using the Overlap Coefficient and obtaining 70\% of similarity among the clusters.}, added-at = {2019-11-04T11:36:28.000+0100}, address = {Cham}, author = {Reyes-Peña, Cecilia and Tovar Vidal, Mireya and Lavalle Martínez, José de Jesús}, biburl = {https://www.bibsonomy.org/bibtex/2e7241d84d3257f17d92a73cf5dfd35f7/lepsky}, booktitle = {Proceedings of the {Future} {Technologies} {Conference} ({FTC}) 2019}, doi = {10.1007/978-3-030-32520-6_44}, editor = {Arai, Kohei and Bhatia, Rahul and Kapoor, Supriya}, interhash = {ca026f58d1d704be2555708f04a2feda}, intrahash = {e7241d84d3257f17d92a73cf5dfd35f7}, isbn = {978-3-030-32520-6}, keywords = {clustering}, language = {en}, pages = {610--617}, publisher = {Springer International Publishing}, series = {Advances in {Intelligent} {Systems} and {Computing}}, shorttitle = {Document {Clustering} by {Relevant} {Terms}}, timestamp = {2019-11-04T11:36:28.000+0100}, title = {Document clustering by relevant terms : an approach}, year = 2020 }