Discovery, Analysis, and Retrieval of Multimodal Environmental Information

Environmental conditions are considered of utmost importance for human life. Citizens are increasingly aware of the important role that environmental data (i.e. weather forecast, air quality and pollen concentration) play on health issues (e.g. allergies), as well as to a variety of outdoor activities (e.g. agriculture, trip planning). Given the fact that ensembling information from several environmental providers can generate more reliable measurements, there is a need to combine environmental data from multiple resources, in order to facilitate retrieval of environmental information and support personalized services (Wanner et al., 2012). In this context, this article analyzes the aforementioned needs and challenges (Figure 1) by discussing the application of techniques from the information technologies domain on environmental data. First, we address the discovery of environmental web resources (referred to as environmental nodes) as a domainspecific search problem. Then, we provide insights into the presentation formats of the environmental resources, as well as information extraction techniques that could be applied. Finally, we discuss indexing and retrieval of environmental information. The article is structured as follows. First we present the background and basic definitions regarding the environmental information. Then, an empirical study on the presentation of environmental data is realized. In the following sections, the approaches for environmental data discovery, content extraction, as well as indexing and retrieval are reported. Finally, we present future trends and conclusions. BACKGROUND