In Information Extraction (IE), processing of named entities in text has traditionally been seen as a two-step process comprising a flat text span recognition sub-task and an atomic classification sub-task; relating the text span to a model of the world has been ignored by evaluations such as DARPA/NIST's MUC or ACE. However, spatial and temporal expressions refer to events in space-time, and the grounding of events is a precondition for accurate reasoning. Thus, automatic grounding can improve many applications such as automatic map drawing (e.g. for choosing a focus) and question answering (e.g., for questions like How far is London from Edinburgh , given a story in which both occur and can be resolved). Whereas temporal grounding has received considerable attention in the recent Past [2, 3], robust spatial grounding has long been neglected. Concentrating on geographic names for populated places, I define the task of automatic Toponym Resolution (TR) as computing the mapping from occurrences of names for places as found in a text to a representation of the extensional semantics of the location referred to (its referent), such as a geographic latitude/longitude footprint. The task of mapping from names to locations is hard due to insufficient and noisy databases, and a large degree of ambiguity: common words need to be distinguished from proper names (geo/non-geo ambiguity), and the mapping between names and locations is ambiguous London can refer to the capital of the UK or to London, Ontario, Canada, or to about forty other Londons on earth). In addition, names of places and the boundaries referred to change over time, and databases are incomplete.

Recognizing spatial language in text documents, termed geoparsing , is useful for many applications, because together with mapping such language to lat/long values, also known as geocoding , it enables the connection of the unstructured textual realm with the structured realm of Geographic Information Systems (GIS) [11]. For example, news stories about events happening in a particular location can be explored on a map for a spatial understanding of these events, as implemented by applications like the European Media Monitor (EMM) [18] and NewsStand [13, 20]. Web pages, blogs, encyclopedia articles, news stories, tweets and travel reports can all benefit from such interlinking with maps, which requires the recognition of spatial language. Note that geoparsing can be considered as a more specific application of the task of Named Entity Recognition and Classification (NERC): NERC is concerned with automatically recognizing proper nouns of any kind, often meant to include monetary amounts, dates, and other types, while geoparsing is the NERC task applied to locations specifically. Geoparsing is also known by many names in the literature, including geotagging, georecognition , and toponym recognition , but for consistency, here we will refer only to geoparsing. In this paper, we provide an overview of the challenges related to geoparsing, several families of geoparsing methods, existing systems and data collections available for performing geoparsing, and open research questions related to geoparsing.

No abstract

The task of named entity annotation of unseen text has recently been successfully automated with near-human performance. But the full task involves more than annotation, i.e. identifying the scope of each (continuous) text span and its class (such as place name). It also involves grounding the named entity (i.e. establishing its denotation with respect to the world or a model). The latter aspect has so far been neglected.In this paper, we show how geo-spatial named entities can be grounded using geographic coordinates, and how the results can be visualized using off-the-shelf software. We use this to compare a "textual surrogate" of a newspaper story, with a "visual surrogate" based on geographic coordinates. § ¤ 3 45, 2, 345¦ ;

No abstract