Abstract. Ontologies evolve over time, when they are altered to correct errors, to accommodate ne... more Abstract. Ontologies evolve over time, when they are altered to correct errors, to accommodate new information or to adjust the representation of the domain. Hence there is a need for methods and means to manage the ontology evolution in order to ensure that applications using different versions of an ontology remain compatible with respect each other and metadata annotated. In this paper an ontology versioning framework is presented to enable ontology versioning.
Abstract We present a national ontology development and service framework being developed in Finl... more Abstract We present a national ontology development and service framework being developed in Finland in 2003-2007. Our goal is to initiate and support collaborative ontology development processes of various expert groups now developing keyword thesauri. The framework is based on a set of related core ontologies, most notably on a national upper ontology based on the commonly used Finnish General Thesaurus YSA. The framework implements three ontology services by a web-based system ONKI.
Abstract This paper presents a semantic recommender method and a system for a personalized access... more Abstract This paper presents a semantic recommender method and a system for a personalized access to digital cultural heritage through context-aware user profiling. Given annotation knowledge-bases, explicit background knowledge in the form of ontologies, a user model capturing the user's behavior and context, the system produces recommendations. Ontology-based user profiling can be used to reduce cold-start, sparsity and over-specialization problems.
Abstract. The Linked Open Data University of Münster (LODUM) project establishes a university-wid... more Abstract. The Linked Open Data University of Münster (LODUM) project establishes a university-wide infrastructure to publish university data as Linked Open Data. The main goals are to increase visibility and accessibility of data produced and collected at the university, and to facilitate effective reuse of these data. This includes the goal to ease the development of applications and mashups based on the data, so that the common user can benefit from the LODUM data.
Abstract: In this paper, we argue that sensors provide a better understanding of geographic event... more Abstract: In this paper, we argue that sensors provide a better understanding of geographic events. They produce observations that reflect the natural events taking place at a particular location. The essential part of deriving information about geographic events from sensor observations is to formalize the relations between them. In this spirit, we develop an ontology to capture the relations between weather events and properties observed by sensors.
Abstract. Tropical deforestation is an example of geochange with massive impacts to the environme... more Abstract. Tropical deforestation is an example of geochange with massive impacts to the environment locally and globally. In the Brazilian Amazon deforestation has prevailed owing mostly to private investments in agricultural expansion, associated with large-scale cattle ranching, smallscale subsistence farming and soybeans expansion. Data on deforestation have been relied mostly on satellite remote sensing, mapping the extent of forest loss.
Place names and their geographical coverage change in time. This causes problems when retrieving ... more Place names and their geographical coverage change in time. This causes problems when retrieving information content related to different times. Geo-content is usually indexed using place names of the time of indexing (eg a photo of the 1968 upraise of Czechoslovakia indexed then) or of the time that the content has been used or created (eg a spear used in the Punic Wars in 146 bc in Carthago but indexed at a later time using place names of that time).
This dissertation discusses the problems and the methods of creating and using ontologies in the ... more This dissertation discusses the problems and the methods of creating and using ontologies in the area of digital cultural heritage. One of the problems is that content annotations in semantic cultural heritage portals commonly make spatiotemporal references to historical regions and places using names whose meanings are different in different times.
Metadata for scientific publications contain various explicit and implicit spatio-temporal refere... more Metadata for scientific publications contain various explicit and implicit spatio-temporal references. Data on conference locations as well as author and editor affiliations–both changing over time–enable insights into the geographic distribution of scientific fields and particular specializations. At the same time, these byproducts of scientific bibliographies offer a great opportunity to integrate data across different bibliographies to get a more complete picture of a domain.
Abstract. We propose to use the trustworthiness of features in Open-StreetMap as a proxy function... more Abstract. We propose to use the trustworthiness of features in Open-StreetMap as a proxy function for data quality, where a feature's trust value is computed solely from its history. The trustworthiness is based on the different contribution patterns that can be found in a feature's history, such as rollbacks or deletions. We argue that these patterns influence the trustworthiness of a feature and its contributors' reputation. We discuss different potential implications of the patterns on trust and reputation. Keywords.
Abstract. A lot of functionality is needed when an application, such as a museum cataloguing syst... more Abstract. A lot of functionality is needed when an application, such as a museum cataloguing system, is extended with semantic capabilities, for example ontological indexing functionality or multi-facet search. To avoid duplicate work and to enable easy and cost-efficient integration of information systems with the Semantic Web, we propose a mash-up approach. Here, data sources are combined with functionality into readyto-use software components that allow adding semantic functionality to systems with just a few lines of code.
Abstract. Different techniques, including remote sensing, are used to collect data about the defo... more Abstract. Different techniques, including remote sensing, are used to collect data about the deforestation in the Brazilian Amazon rainforest. However, a key problem for integrating all the resulting datasets for statistical analysis is the semantic diversity among these datasets. In this paper we identify key semantic interoperability challenges for the Brazilian Amazon rainforest statistics.
Abstract. This paper presents the national level cross-domain ontology and ontology service infra... more Abstract. This paper presents the national level cross-domain ontology and ontology service infrastructure ONKI used in Finland. The novelty of ONKI is based on two ideas. First, the core ontologies are developed collaboratively by experts transforming thesauri into mutually aligned lightweight ontologies, based on a large top ontology that is extended by various domain specific ontologies. Second, the National Ontology Service ONKI has been implemented for publishing ontologies cost-efficiently as ready to use services.
Abstract. Ontologies evolve over time, when they are altered to correct errors, to accommodate ne... more Abstract. Ontologies evolve over time, when they are altered to correct errors, to accommodate new information or to adjust the representation of the domain. Hence there is a need for methods and means to manage the ontology evolution in order to ensure that applications using different versions of an ontology remain compatible with respect each other and metadata annotated. In this paper an ontology versioning framework is presented to enable ontology versioning.
Abstract We present a national ontology development and service framework being developed in Finl... more Abstract We present a national ontology development and service framework being developed in Finland in 2003-2007. Our goal is to initiate and support collaborative ontology development processes of various expert groups now developing keyword thesauri. The framework is based on a set of related core ontologies, most notably on a national upper ontology based on the commonly used Finnish General Thesaurus YSA. The framework implements three ontology services by a web-based system ONKI.
Abstract This paper presents a semantic recommender method and a system for a personalized access... more Abstract This paper presents a semantic recommender method and a system for a personalized access to digital cultural heritage through context-aware user profiling. Given annotation knowledge-bases, explicit background knowledge in the form of ontologies, a user model capturing the user's behavior and context, the system produces recommendations. Ontology-based user profiling can be used to reduce cold-start, sparsity and over-specialization problems.
Abstract. The Linked Open Data University of Münster (LODUM) project establishes a university-wid... more Abstract. The Linked Open Data University of Münster (LODUM) project establishes a university-wide infrastructure to publish university data as Linked Open Data. The main goals are to increase visibility and accessibility of data produced and collected at the university, and to facilitate effective reuse of these data. This includes the goal to ease the development of applications and mashups based on the data, so that the common user can benefit from the LODUM data.
Abstract: In this paper, we argue that sensors provide a better understanding of geographic event... more Abstract: In this paper, we argue that sensors provide a better understanding of geographic events. They produce observations that reflect the natural events taking place at a particular location. The essential part of deriving information about geographic events from sensor observations is to formalize the relations between them. In this spirit, we develop an ontology to capture the relations between weather events and properties observed by sensors.
Abstract. Tropical deforestation is an example of geochange with massive impacts to the environme... more Abstract. Tropical deforestation is an example of geochange with massive impacts to the environment locally and globally. In the Brazilian Amazon deforestation has prevailed owing mostly to private investments in agricultural expansion, associated with large-scale cattle ranching, smallscale subsistence farming and soybeans expansion. Data on deforestation have been relied mostly on satellite remote sensing, mapping the extent of forest loss.
Place names and their geographical coverage change in time. This causes problems when retrieving ... more Place names and their geographical coverage change in time. This causes problems when retrieving information content related to different times. Geo-content is usually indexed using place names of the time of indexing (eg a photo of the 1968 upraise of Czechoslovakia indexed then) or of the time that the content has been used or created (eg a spear used in the Punic Wars in 146 bc in Carthago but indexed at a later time using place names of that time).
This dissertation discusses the problems and the methods of creating and using ontologies in the ... more This dissertation discusses the problems and the methods of creating and using ontologies in the area of digital cultural heritage. One of the problems is that content annotations in semantic cultural heritage portals commonly make spatiotemporal references to historical regions and places using names whose meanings are different in different times.
Metadata for scientific publications contain various explicit and implicit spatio-temporal refere... more Metadata for scientific publications contain various explicit and implicit spatio-temporal references. Data on conference locations as well as author and editor affiliations–both changing over time–enable insights into the geographic distribution of scientific fields and particular specializations. At the same time, these byproducts of scientific bibliographies offer a great opportunity to integrate data across different bibliographies to get a more complete picture of a domain.
Abstract. We propose to use the trustworthiness of features in Open-StreetMap as a proxy function... more Abstract. We propose to use the trustworthiness of features in Open-StreetMap as a proxy function for data quality, where a feature's trust value is computed solely from its history. The trustworthiness is based on the different contribution patterns that can be found in a feature's history, such as rollbacks or deletions. We argue that these patterns influence the trustworthiness of a feature and its contributors' reputation. We discuss different potential implications of the patterns on trust and reputation. Keywords.
Abstract. A lot of functionality is needed when an application, such as a museum cataloguing syst... more Abstract. A lot of functionality is needed when an application, such as a museum cataloguing system, is extended with semantic capabilities, for example ontological indexing functionality or multi-facet search. To avoid duplicate work and to enable easy and cost-efficient integration of information systems with the Semantic Web, we propose a mash-up approach. Here, data sources are combined with functionality into readyto-use software components that allow adding semantic functionality to systems with just a few lines of code.
Abstract. Different techniques, including remote sensing, are used to collect data about the defo... more Abstract. Different techniques, including remote sensing, are used to collect data about the deforestation in the Brazilian Amazon rainforest. However, a key problem for integrating all the resulting datasets for statistical analysis is the semantic diversity among these datasets. In this paper we identify key semantic interoperability challenges for the Brazilian Amazon rainforest statistics.
Abstract. This paper presents the national level cross-domain ontology and ontology service infra... more Abstract. This paper presents the national level cross-domain ontology and ontology service infrastructure ONKI used in Finland. The novelty of ONKI is based on two ideas. First, the core ontologies are developed collaboratively by experts transforming thesauri into mutually aligned lightweight ontologies, based on a large top ontology that is extended by various domain specific ontologies. Second, the National Ontology Service ONKI has been implemented for publishing ontologies cost-efficiently as ready to use services.
Uploads
Papers by Tomi Kauppinen