Keywords
hydrology, water, natural language generation, cyberinfrastructure, scotland
Start Date
1-7-2012 12:00 AM
Abstract
Increasing the effectiveness of how river level data are communicated to a range of stakeholders (individuals, groups and communities) with an interest in river level information is likely to result in greater use of data collected by regulatory agencies. A range of interest groups, like those involved in recreational pursuits such as angling and canoeing, require certain but different information on changes in river levels state to allow effective scheduling of their activities. A range of options have been developed for communicating river level data to different audiences, but those fail to address group heterogeneity and information demands. To a large extent, those problems derive from a lack of understanding of information demanded by river water users, as well as the failure to comprehend how they perceive river level change. We are working with river users who span land managers and farmers, hydropower generators, recreational users e.g. those involved in canoeing and fishing and broader local communities as well as the public authority (SEPA) responsible for hydrological monitoring and provision of this information and cyberinfrastructure in Scotland. Currently, river level data is provided to members of the public through a web site without any formal engagement with river users having taken place. In our research project called wikiRivers, we are working with the suppliers of river level data as well as the users of this data to explore and improve from the user perspective how river level data and information is made available online. We are focusing on the application of natural language generation technology to create textual summaries of river level data tailored for specific interest groups. These tailored textual summaries will be presented among other modes of information presentation (e.g. maps and visualizations) with the aim to increase communication effectiveness. Natural language generation involves developing computational models that use non-linguistic input data to produce natural language as their output. Acquiring accurate correct system knowledge for natural language generation is a key step in developing such an effective computer software system. In this paper we set out the needs for this project based on discussions with the stakeholder who supplies the river level data and current cyberinfrastructure and present a detailed stakeholder identification, engagement and cyberinfrastructure development plan.
Communicating river level data and information to stakeholders with different interests: the participative development of an interactive online service
Increasing the effectiveness of how river level data are communicated to a range of stakeholders (individuals, groups and communities) with an interest in river level information is likely to result in greater use of data collected by regulatory agencies. A range of interest groups, like those involved in recreational pursuits such as angling and canoeing, require certain but different information on changes in river levels state to allow effective scheduling of their activities. A range of options have been developed for communicating river level data to different audiences, but those fail to address group heterogeneity and information demands. To a large extent, those problems derive from a lack of understanding of information demanded by river water users, as well as the failure to comprehend how they perceive river level change. We are working with river users who span land managers and farmers, hydropower generators, recreational users e.g. those involved in canoeing and fishing and broader local communities as well as the public authority (SEPA) responsible for hydrological monitoring and provision of this information and cyberinfrastructure in Scotland. Currently, river level data is provided to members of the public through a web site without any formal engagement with river users having taken place. In our research project called wikiRivers, we are working with the suppliers of river level data as well as the users of this data to explore and improve from the user perspective how river level data and information is made available online. We are focusing on the application of natural language generation technology to create textual summaries of river level data tailored for specific interest groups. These tailored textual summaries will be presented among other modes of information presentation (e.g. maps and visualizations) with the aim to increase communication effectiveness. Natural language generation involves developing computational models that use non-linguistic input data to produce natural language as their output. Acquiring accurate correct system knowledge for natural language generation is a key step in developing such an effective computer software system. In this paper we set out the needs for this project based on discussions with the stakeholder who supplies the river level data and current cyberinfrastructure and present a detailed stakeholder identification, engagement and cyberinfrastructure development plan.
