On the Use of Consensus Clustering for Incremental Learning of Topic Hierarchies

Using Contextual Information from Topic Hierarchies to Improve Context-Aware Recommender Systems

“…In this paper, we first propose to use an unsupervised learning method, called BC 2 (Buckshot Consensus Clustering) [4], Fig. 1: Hierarchical structure of the contextual attribute "period of the year" [5] to generate topic hierarchies from textual data, that can be viewed as contextual information that characterize the items.…”

“…In this sense, we use an approach for topic hierarchy construction based on the consensus clustering called BC 2 (Buckshot Consensus Clustering) [4]. In BC 2 approach, it is possible to combine solutions of different topic extraction algorithms in a single consensual solution.…”

“…Thus, in these situations, one possible solution is to use unsupervised learning methods. In this sense, topic hierarchies are efficient models to capture the context of textual data in order to organize them [4]. These models allow the organization of items into topics and subtopics, providing an intuitive way to explore contextual information at different levels of granularity.…”

Interactive textual feature selection for consensus clustering

“…In this sense, consensus clustering allows the combination of different clustering solutions into a unique and more robust clustering solution [10,13,14]. Thus, if a document is mistakenly allocated to a particular clustering solution, the same document will not necessarily be mistakenly allocated in other clustering solutions -i.e., eventual errors can be corrected in the final solution obtained by consensus clustering [15].…”

Extração de tópicos baseado em agrupamento de regras de associação

“…Já a complexidade das estratégias divisivas crescem exponencialmente em relação ao tamanho do conjunto de dados, proibindo sua aplicação em conjuntos de dados grandes (Xu and Wunsch, 2008). No entanto, nosúltimos anos foram propostos algoritmos de agrupamento hierárquico divisivos com complexidade semelhante aos aglomerativos, possibilitando sua aplicação em conjuntos de dados maiores, inclusive em coleções textuais (Steinbach et al, 2000;Marcacini et al, 2012b).…”