Location Extraction from Social Media

Abstract

Location extraction, also called “toponym extraction,” is a field covering geoparsing, extracting spatial representations from location mentions in text, and geotagging, assigning spatial coordinates to content items. This article evaluates five “best-of-class” location extraction algorithms. We develop a geoparsing algorithm using an OpenStreetMap database, and a geotagging algorithm using a language model constructed from social media tags and multiple gazetteers. Third-party work evaluated includes a DBpedia-based entity recognition and disambiguation approach, a named entity recognition and Geonames gazetteer approach, and a Google Geocoder API approach. We perform two quantitative benchmark evaluations, one geoparsing tweets and one geotagging Flickr posts, to compare all approaches. We also perform a qualitative evaluation recalling top N location mentions from tweets during major news events. The OpenStreetMap approach was best (F1 0.90+) for geoparsing English, and the language model approach was best (F1 0.66) for Turkish. The language model was best (F1@1km 0.49) for the geotagging evaluation. The map database was best (R@20 0.60+) in the qualitative evaluation. We report on strengths, weaknesses, and a detailed failure analysis for the approaches and suggest concrete areas for further research.

nLORE: A Linguistically Rich Deep-Learning System for Locative-Reference Extraction in Tweets

“…locative preposition + proper nouns) and/or lexical resources (e.g. gazetteers of place names or lists of locative cues) together with supportive natural language processing (NLP) tasks, such as the part-of-speech (POS) tagging of n-grams [3,11,12]. Other systems are based on probabilistic frameworks, using Conditional Random Fields (CRF) or neural networks such as Convolutional Neural Networks (CNN) or Recurrent Neural Networks (RNN) with or without additional linguistic feature engineering [7,13].…”

“…Because of the lack of geotagged data, e.g. geotagged tweets represent only around 1% of tweets [3], and its recent sharing restrictions, it becomes necessary to turn to other geospatial evidence, such as that found in tweets in the form of locative references, which are usually much more frequent than geotagged data [4]. Twitter has in fact been exploited in many geolocation systems that handle real-life scenarios, e.g.…”

Comparing Methods to Retrieve Tweets: a Sentiment Approach

“…In some of these cases researchers require information that is fully geolocalized: this happens, for example, monitoring socio-demographic phenomena (Jashinsky et al, 2014), in disaster management (de Bruijn et al, 2017) or in transportation planning studies (Paule et al, 2019). In this framework, one of the biggest problem is that tweets with a geographical information are just a small fraction of the total (Middleton et al, 2018). Moreover, this information is not always reliable or nicely structured (Middleton et al, 2018, Zheng et al, 2018, mostly because self reported by users.…”

“…In this framework, one of the biggest problem is that tweets with a geographical information are just a small fraction of the total (Middleton et al, 2018). Moreover, this information is not always reliable or nicely structured (Middleton et al, 2018, Zheng et al, 2018, mostly because self reported by users. There are currently several methods used to overtake these limits: location extraction (Ozdikis et al, 2017, de Bruijn et al, 2017, Zheng et al, 2018 or statistical models and machine learning methods are used to assign spatial coordinates to media items basing on tweets content (Zola et al, 2019, Han et al, 2014.…”

Extracting and modeling geographic information from scientific articles

“…Geographical information permeates the written world, appearing as place names or place descriptions in texts including news articles, blog posts, social media content, historical documents, and scientific articles. Research on extracting geographical information from text has often focused on news articles [1][2][3] and social media content [4][5][6], with surprisingly limited attention being directed towards the increasing number of published scientific articles. Indeed, with each passing year, scientists face an ever-growing stack of scientific articles to sort through, read, understand, and build upon.…”