Mettina Veenstra

Learn More
Generating hypermedia presentations requires processing constituent material into coherent, unified presentations. One large challenge is creating a generic process for producing hypermedia presentations from the semantics of potentially unfamiliar domains. The resulting presentations must both respect the underlying semantics and appear as coherent,(More)
Predicting the interests of a user in information is an important process in personalized information systems. In this paper, we present a way to create prediction engines that allow prediction techniques to be easily combined into prediction strategies. Prediction strategies choose one or a combination of prediction techniques at the moment a prediction is(More)
For humans to gain comprehensive views of large amounts of repository contents, they need to have insight into the relations among information objects. It is a challenge to automatically generate presentations of repository contents, through, for example, search results, which reveal such relations to readers. Such presentations must reflect properties of(More)
Adaptive systems can be used to help people find their way through large amounts of information by taking into account the content itself and opinions of other people. Annotations that describe content are important in this process as is generally recognized. What is less obviously stated so far is that not only the annotations themselves, but also(More)
In this paper, we report on a study that explores the contribution of social tags, professional metadata and automatically generated metadata to the retrieval process. In this study, 194 participants tagged a total of 115 videos, while another 140 participants searched the video collection for answers to eight questions. The results show that in the current(More)
Semantic annotations of media repositories make relationships among the stored media and relevant concepts explicit. However, these relationships and the media they join are not directly presentable as hypermedia. Previous work shows how clustering over the annotations in the repositories can determine hypermedia presentation structure. Here we explore the(More)
An important step in providing personalized information, such as personalized electronic program guides, is predicting the level of interest for a specific user in information, such as TV programs. This paper describes a model that can be used to combine prediction techniques, which predict the level of interest, into prediction strategies. Our hypothesis(More)
Hybrid recommender systems are capable of providing better recommendations than non-hybrid ones. Our approach to hybrid recommenders is the use of prediction strategies that determine which prediction technique(s) should be used at the moment an actual prediction is required. In this paper, we determine whether case-based reasoning can provide more accurate(More)
Recommender systems help people to find information that is interesting to them. However, current recommendation techniques only address the user’s short-term and long-term interests, not their immediate interests. This paper describes a method to structure information (with or without using recommendations) taking into account the users’ immediate(More)
This paper explores the creation of smart information services for tourists using dynamic open data. Research is described, which uses physical sensors at the Van Gogh Museum in Amsterdam to retrieve data about queueing time at that location. This data is added to an open data framework called CitySDK and combined with other open data in the framework.