The problem of real-time extraction of meaningful patterns from time- changing data streams is important in financial applications, telecommunication data management, web applications, surveillance, monitoring patient health, and many others. The paper proposes an efficient incremental algorithm for learning regression and model trees from unbounded, high-speed, and time-changing data streams. The algorithm performs on-line and in real-time, observes each example only once at the speed of arrival, and maintains at any-time an accurate and ready-to-use model tree.
COBISS.SI-ID: 24179751
The paper investigates the role of outliers in literature-based knowledge discovery. It shows that detecting interesting outlier documents can help the expert to find implicit relationships among concepts of different domains. The proposed approach contributes to cross-context link discovery by proving the utility of outlier detection for finding bisociative links in the process of autism literature exploration, as well as by uncovering implicit relationships in the articles from the migraine domain.
COBISS.SI-ID: 1621243
This is an application of decision-support methodology to sustainable development in protected areas. Based on the analysis of the infrastructure of the Triglav National Park (TNP) in Slovenia, we developed a model for the assessment of sustainability of mountain huts in the Alps. The assessment results are of particular interest for decision makers in protected areas, such as Alpine national parks managers and administrators. The model helps to minimize the ecological footprint of tourists in environmentally sensitive and undeveloped mountain areas and contributes to mountain ecotourism.
COBISS.SI-ID: 1549563
Lemmatisation is the process of finding the normalised forms of words appearing in text. The paper presents a new lemmatisation system, LemmaGen, which was trained to generate accurate and efficient lemmatisers for twelve different languages. To our knowledge, LemmaGen is the most efficient publicly available lemmatiser trained on large lexicons of multiple languages, whose learning engine can be retrained to effectively generate lemmatisers of other languages.
COBISS.SI-ID: 23897895
Knowledge workers are central to an organization’s success, yet their information management tools often hamper their productivity. The paper addresses this challenge through an integrated knowledge management workspace that reduces information overload by significantly improving the mechanisms for creating, managing, and using information. The approach follows three themes: sharing information through tagging, wikis, and ontologies; prioritizing information delivery by understanding users’ current-task context; and leveraging informal processes that are learned from user behavior.
COBISS.SI-ID: 24216871