Integrating archaeological literature into resource discovery interfaces using natural language processing and name authority services

摘要

There exists a large and underutilized resource of archaeological literature, both formal, such as scholarly journals and less formal in the form of `grey literature'. In the archaeological domain the vast majority of this literature contains some geo-spatial element as well as the expected temporal information and therefore its ease of discovery would be greatly enhanced were it accessible via a geo-spatially enabled search mechanism. As a result of this, geo-referencing these types of material and integrating them with other resources, such as monument inventories, is seen as a desirable enhancement for digital archives serving the archaeological research community. This paper provides an overview of a number of the approaches to the integration of such legacy literature into geospatial search mechanisms in an archaeological context. In particular efforts to achieve this via the Archaeotools e-Science project and its use of natural language processing and a geo-spatial cross-walk service are discussed as well as potential future enhancements to the process.