In order to facilitate the analysis of user-generated contents (always enormous in number), we studied novel topic detection and tracking (TDT) techniques to analyze and follow the evolution of the information expressed by a social network. In order to do that, we studied novel metrics to identify the relationships that exist among users, pages and contents, and therefore we map these information in a social graph where it is possible to follow the most emergent topics (by considering temporal conditions) expressed by a social network community.
During the last years, the number of text documents in digital form has grown enormously in size. For this, it is of practical importance to be able to automatically organize these very dynamic information in order to provide novel mechanisms for an efficient exploration and retrieval process. I investigated these problems developing innovative statistical-based data mining methodologies for exploration purposes; in particular, we take into account user preferences, specific content domain properties and aggregations techniques to help the user explore the text data. We also studied novel approaches for efficiently represent the data and help the user identify hidden semantic relationships that exist among them.
Considering the enormous amount of multimedia contents in digital form, there is an emerging need for representing this information in navigable graphs that can help analyse and explore the information expressed by them. Moreover, considering that these graphs are often too large to be explored, it is also important to summarize the data in order to help reduce the size of the considered graphs preserving as much as possible the relevant information and reducing the overall redundancy. In order to do that, we developed novel graph manipulation methods and new summarization/anonymization approaches to formalize the content expressed by the multimedia documents by minimizing the loss of information and the time requested by this process.
During my Ph.D. activities, I studied how, in a Web 2.0 reality, cultural heritage institutions and museums communicate with their visitors and how these visitors understand the message vehiculated by them. In particular, my work cope with the following questions: which topics are debated by the visitors in the web and could we compare them with those disseminated by museums web sites? How will social web have an impact on cultural institutions communication? Thus, my work tried to investigate those research questions through a quantitative content analysis of a sample of museums institutional web sites and related blogosphere’s contents (see also Cooperare Project).
