International audienceGeoparsing and geocoding are two essential middleware services to facilitate final user applications such as location-aware searching or different types of location-based services. The objective of this work is to propose a method for establishing a processing chain to support the geoparsing and geocoding of text documents describing events strongly linked with space and with a frequent use of fine-grain toponyms. The geoparsing part is a Natural Language Processing approach which combines the use of part of speech and syntactico-semantic combined patterns (cascade of transducers). However, the real novelty of this work lies in the geocoding method. The geocoding algorithm is unsupervised and takes profit of clustering techniques to provide a solution for disambiguating the toponyms found in gazetteers, and at the same time estimating the spatial footprint of those other fine-grain toponyms not found in gazetteers. The feasibility of the proposal has been tested with a corpus of hiking descriptions in French, Spanish and Italian

Geographic text analysis (GTA) research in the digital humanities has focused on projects analyzing modern English-language corpora. These projects depend on temporally specific lexicons and gazetteers that enable place name identification and georesolution. Scholars working on the early modern period (1400–1800) lack temporally appropriate geoparsers and gazetteers and have been reliant on general purpose linked open data services like Geonames. These anachronistic resources introduce significant information retrieval and ethical challenges for early modernists. Using the geography entries of the canonical eighteenth-century Encyclopédie, we evaluate rule-based named entity recognition (NER) systems to pinpoint areas where they would benefit from adjustments for processing historical corpora. As we demonstrate, annotating nested and extended place information is one way to improve early modern GTA. Working with Enlightenment sources also motivates a critique of the landscape of digital geospatial data.

Considerable amounts of geographical data are still collected not in form of GIS data but just as natural language texts. This paper proposes an approach for the automatic geocoding of itineraries described in natural language. This approach needs as an input a text annotated with part-of-speech and geo-semantic tags. The proposed method is divided into three main steps. First, we build a complete graph where vertices represent locations, and all vertices are connected to each other by undirected edges. We assign a weight to all the edges of the complete graph using a multi-criteria analysis approach. Then we compute a minimum spanning tree to obtain an undirected acyclic graph connecting all vertices. And finally, we transform this graph into a partially directed acyclic graph in order to identify the sequence of waypoints and build an approximation of a plausible footprint of the itinerary described. Additionally, the rationale of the proposed approach has been verified with a set of experiments on a corpus of hiking descriptions.

Geocoding aims to assign unambiguous locations (i.e., geographic coordinates) to place names (i.e., toponyms) referenced within documents (e.g., within spreadsheet tables or textual paragraphs). This task comes with multiple challenges, such as dealing with referent ambiguity (multiple places with a same name) or reference database completeness. In this work, we propose a geocoding approach based on modeling pairs of toponyms, which returns latitude-longitude coordinates. One of the input toponyms will be geocoded, and the second one is used as context to reduce ambiguities. The proposed approach is based on a deep neural network that uses Long Short-Term Memory (LSTM) units to produce representations from sequences of character n-grams. To train our model, we use toponym co-occurrences collected from different contexts, namely textual (i.e., co-occurrences of toponyms in Wikipedia articles) and geographical (i.e., inclusion and proximity of places based on Geonames data). Experiments based on multiple geographical areas of interest—France, United States, Great-Britain, Nigeria, Argentina and Japan—were conducted. Results show that models trained with co-occurrence data obtained a higher geocoding accuracy, and that proximity relations in combination with co-occurrences can help to obtain a slightly higher accuracy in geographical areas with fewer places in the data sources.

International audienceOur project involves building a platform able to retrieve, map and analyze the occurrences of place names in fictional novels published between 1800 and 1914 and whose action occurs wholly or partly in Paris. We describe a proof of concept using queries made via the TXM textual analysis platform for the extraction of street names. Then, we propose a fully automatic process using the named entity recognition (NER) components of the PERDIDO platform. This paper describes some encouraging initial results obtained by combining NLP approaches (NER methods) with textometric tools for the automated geoparsing of street names