Combining social-based data mining techniques to extract collective trends from twitter
Social Networks have become an important environment for Collective Trends extraction. The interactions amongst users provide information of their preferences and relationships. This information can be used to measure the influence of ideas, or opinions, and how they are spread within the Network. C...
| Autores: | , , , |
|---|---|
| Formato: | artículo |
| Fecha de publicación: | 2014 |
| País: | España |
| Recursos: | Universidad Autónoma de Madrid |
| Repositorio: | Biblos-e Archivo. Repositorio Institucional de la UAM |
| Idioma: | inglés |
| OAI Identifier: | oai:repositorio.uam.es:10486/666288 |
| Acesso em linha: | http://hdl.handle.net/10486/666288 |
| Access Level: | acceso abierto |
| Palavra-chave: | Collective Trends Social Network Classification Clustering Informática |
| Resumo: | Social Networks have become an important environment for Collective Trends extraction. The interactions amongst users provide information of their preferences and relationships. This information can be used to measure the influence of ideas, or opinions, and how they are spread within the Network. Currently, one of the most relevant and popular Social Networks is Twitter. This Social Network was created to share comments and opinions. The information provided by users is especially useful in different fields and research areas such as marketing. This data is presented as short text strings containing different ideas expressed by real people. With this representation, different Data Mining techniques (such as classification or clustering) will be used for knowledge extraction to distinguish the meaning of the opinions. Complex Network techniques are also helpful to discover influential actors and study the information propagation inside the Social Network. This work is focused on how clustering and classification techniques can be combined to extract collective knowledge from Twitter. In an initial phase, clustering techniques are applied to extract the main topics from the user opinions. Later, the collective knowledge extracted is used to relabel the dataset according to the clusters obtained to improve the classification results. Finally, these results are compared against a dataset which has been manually labelled by human experts to analyse the accuracy of the proposed method. |
|---|