On 28th of May I will be a presentor at the EU workshop “Summarization - Key to information overload”
Personalized News Summarization in the Financial Economic domain.
In automated summarization, articles are automatically shortened and the most important information of the article is retained in the summary. The underlying assumption is that there is one optimal summary to represent the original contents. In many use cases however, the process of summarization is more about informing users of things (s)he should know about. In this situation there may well be relevant information in an article, that do not constitute the main points of the article and will not become part of the summary. In order to represent this type of information as well, summarization techniques need to be personalized.
In this presentation I will talk about my research into personalized summarization at FDMG; the leading information provider in the financial economic domain in the Netherlands. For FDMG we have researched the application of state of the art deep-learning based abstractive summarization methods in combination with recommender technology in order to create personalized summaries.