Publication: T.E.F.G: generador de frecuencias de eventos temporales
Loading...
Files
Official URL
Full text at PDC
Publication Date
2018
Advisors (or tutors)
Editors
Journal Title
Journal ISSN
Volume Title
Publisher
Citations
Exportar
Abstract
Las redes sociales han supuesto un cambio en la forma de conectar con otras personas y compartir información: lo que se hace, comentarios sobre los que se ve en otros medios, pensamientos y opiniones, etc. A escala global, esto genera una ingente cantidad de datos que puede proporcionarnos información muy útil. Además, se ha despertado un gran interés por parte de grandes empresas de sectores como el análisis de opinión o el marketing, que utilizan los gustos y preferencias expresados a través de estas redes para orientar la toma de decisiones en la empresa.
En este proyecto proponemos analizar eventos con duración limitada, como partidos de fútbol o debates electorales, con objeto de extraer la información que mejor representa estos eventos desde el punto de vista de los usuarios de redes sociales. Para ello, comenzamos almacenando los datos generados en estos eventos para su posterior análisis. Posteriormente, realizamos un tratamiento exhaustivo buscando la información más relevante a partir de la frecuencia de mensajes en cada instante del evento. Para visualizar esta información recurrimos a gráficos que muestran la frecuencia de mensajes y los momentos clave, así como los mensajes que mejor representan estos momentos. Para acabar, el usuario final puede interactuar con la representación, por ejemplo, haciendo zoom de un intervalo de especial interés.
Además de mostrar la evolución del evento de forma gráfica, nuestra herramienta también proporciona un resumen cronológico textual de los acontecimientos más importantes del evento, mostrándonos de esta forma cómo se ha percibido el evento bajo el punto de vista de los usuarios de redes sociales.
Social networks have supposed a change in the way people connect each other and share information: what they do, comments about what you see in other media, thoughts and opinions, etc. On a global scale, this generates a huge amount of data that can provide very useful information. Moreover, there is a great interest on the part of big companies in sectors such as opinion analysis or marketing, which make use of tastes and preferences stated through networks to guide the business policy-making. In this project we suggest to analyze time-bound events, such as football matches or election debates, in order to retrieve the information which best represents these events under the social networks user’s point of view. Later, we perform a comprehensive treatment to look for the most relevant information from the messages frequency at every moment of the event. To visualize this information, we make use of graphs that show the messages frequency and the key moments, as well as the messages that best represent the moment. After that, the final user can interact with the graph, for example, by zooming into an interval of great interest. In addition to show the evolution of the event in a graphic way, our program also provides a textual timeline of the most important occurrences of the event, showing how the event has been perceived under the point of view of social network users.
Social networks have supposed a change in the way people connect each other and share information: what they do, comments about what you see in other media, thoughts and opinions, etc. On a global scale, this generates a huge amount of data that can provide very useful information. Moreover, there is a great interest on the part of big companies in sectors such as opinion analysis or marketing, which make use of tastes and preferences stated through networks to guide the business policy-making. In this project we suggest to analyze time-bound events, such as football matches or election debates, in order to retrieve the information which best represents these events under the social networks user’s point of view. Later, we perform a comprehensive treatment to look for the most relevant information from the messages frequency at every moment of the event. To visualize this information, we make use of graphs that show the messages frequency and the key moments, as well as the messages that best represent the moment. After that, the final user can interact with the graph, for example, by zooming into an interval of great interest. In addition to show the evolution of the event in a graphic way, our program also provides a textual timeline of the most important occurrences of the event, showing how the event has been perceived under the point of view of social network users.
Description
Universidad Complutense, Facultad de Informática, curso 2017/2018