Evaluation of georeferencing

Tobin, Richard; Grover, Claire; Byrne, Kate; Reid, James; WALSH, J. R.

doi:10.1145/1722080.1722089

Cited by 38 publications

(31 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Edinburgh Parser Tobin et al 2010) is an end-to-end system, developed in-house and partitioned into two subsystems, a geotagger and a geocoder. The geotagger is a multistep rule-based NER which makes use of lists of place names and person names for personal and location entity recognition.…”

Section: Yahoo!placespottermentioning

confidence: 99%

What’s missing in geographical parsing?

Gritta

Pilehvar

Limsopatham

et al. 2017

Lang Resources & Evaluation

100

124

View full text Add to dashboard Cite

Geographical data can be obtained by converting place names from freeformat text into geographical coordinates. The ability to geo-locate events in textual reports represents a valuable source of information in many real-world applications such as emergency responses, real-time social media geographical event analysis, understanding location instructions in auto-response systems and more. However, geoparsing is still widely regarded as a challenge because of domain language diversity, place name ambiguity, metonymic language and limited leveraging of context as we show in our analysis. Results to date, whilst promising, are on laboratory data and unlike in wider NLP are often not cross-compared. In this study, we evaluate and analyse the performance of a number of leading geoparsers on a number of corpora and highlight the challenges in detail. We also publish an automatically geotagged Wikipedia corpus to alleviate the dearth of (open source) corpora in this domain.

show abstract

Section: Yahoo!placespottermentioning

confidence: 99%

What’s missing in geographical parsing?

Gritta

Pilehvar

Limsopatham

et al. 2017

Lang Resources & Evaluation

100

124

View full text Add to dashboard Cite

show abstract

“…Note that we see geoparsing as the task of location-based extraction from text (NER), and gecoding as the NED complement, the mapping of references to real-world counterparts [17]. Similar to current NER approaches, we can identify three different branches of methods [4]: Gazetteer-based lookup methods [18], Rule-based approaches (e.g. GATEs ANNIE module [19]) by using a set of symbolic rules to encode the decision procedure (Definite Clause Grammars via Prolog) [20].…”

Section: Related Workmentioning

confidence: 99%

Market Intelligence: Linked Data-driven Entity Resolution for Customer and Competitor Analysis

Waltinger

Tecuci

Picioroaga

et al. 2013

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper, we present a linked data-driven method for named entity recognition and disambiguation which is applied within an industry customer and competitor analysis application. The proposed algorithm primarily targets the domain of geoparsing and geocoding, but it can easily be adapted to other problems such duplicate detection. The contributions of this paper are three fold: First, we want to give an overview of Market Intelligence, a customer and competitor analysis application developed for Siemens Energy, which allows users to pose questions and queries on regularly crawled websites, emails and RSS feeds, to detect and respond to competitor, customer, and market trends more effectively. Second, we describe the UIMA-based processing architecture that builds the framework for analyzing and converting unstructured heterogeneous documents into a structured and semantically-enhanced knowledge representation. Third, we propose a novel algorithm that is used within the framework for content analysis and entity disambiguation. The performed evaluation shows with an accuracy of up to 91.69% that the proposed method for named entity recognition and disambiguation is very effective, while at the same time relying on Linked Data only.

show abstract

“…Clough [4] and Tobin [5] reported their geo-referencing systems which firstly used information extraction techniques to identify place names in textual documents and to resolve the place names against gazetteers. This approach is simple and allows efficient implementations, however, with a loss of precision in toponym extraction.…”

Section: A Geo-parsingmentioning

confidence: 99%

A tool to analyse spatial distribution of science research activities based on toponym resolution in text

Ma²,

Shaoxiong³

et al. 2011

2011 19th International Conference on Geoinformatics

View full text Add to dashboard Cite

Abstract-Spatial distribution of research activities in geosciences is related not only to the distribution of authors but also to the distribution of research area. In the present work we developed a tool to analyze spatial distribution of science research activities based on toponym resolution in text of scientific papers. The idea is to identify and annotate toponym referred in research papers automatically, then to geo-code them, at last to carry out an analysis of the distribution of research area and authors in related subjects in geo-sciences. We carried out some experiment with the tool. Firstly, we extracted geographical names from research articles in a pre-built Chinese documentary database on some scientific subjects with words segment software and gazetteer. Secondly, combining with Google map's geocoding API to get coordinates of the places, we exported the place names and coordinates into arcGIS. Consequently, we tried to analyze the spatial distribution of authors and research area in ecological footprint in China. The results show that, based on toponym resolution from large-scale article collection, we can analyze hot areas and blank areas of scientific research on some subjects and the distribution of the researchers of the subjects. And we presented a redesign of the system framework. The improvement is focused on leveraging a geographical knowledge database and some rules for disambiguation, integrating Conditional Random Fields in toponym resolution in Chinese text.

show abstract

Evaluation of georeferencing

Cited by 38 publications

References 8 publications

What’s missing in geographical parsing?

What’s missing in geographical parsing?

Market Intelligence: Linked Data-driven Entity Resolution for Customer and Competitor Analysis

A tool to analyse spatial distribution of science research activities based on toponym resolution in text

Contact Info

Product

Resources

About