International Journal of Geographical Information Science

The Theme and Scope

In an age inundated with semi- and unstructured natural language texts—ranging from news articles and scientific publications through historical records and literary works to social media—a hidden treasure trove of geographic information awaits exploration. This vast array of texts, often containing toponyms, place attributes, and complex location descriptions, offers a unique lens through which to view our world. The extraction and analysis of this geographic information hold immense potential across various fields, ranging from sociolinguistics and spatial humanities to geographic search, disaster management, urban planning, disease surveillance, landscape characterization, and tourism planning. For instance, extracting geographic information from texts has facilitated the timely mapping of situational information during disasters, offered insights into how demographics, environment, and biology influence the emergence of global infectious diseases, and unveiled the spatial structures of narrative within fictional novels. Researchers in geographic information science, information retrieval, and natural language processing have made significant strides in developing methods to understand the geographic focus of documents, extract geographic references from unstructured and heterogeneous texts, and resolve these references unambiguously to specific locations or continuous gridded representations of the Earth’s surface.

In 2008 IJGIS published a special issue on Geographic Information Retrieval, a field specifically concerned with extracting and analysing geographic information from unstructured text with a focus on information retrieval and search. In that special issue, Jones and Purves (2008) set out a range of challenges almost all of which are still relevant today. In the intervening years, natural language has come to be seen as an important source of information for a wide range of geographically relevant research questions, with special issues exploring, for example, the geospatial humanities (Murrieta-Flores and  Martins, 2019) and spatial cognition (Stock et al., 2022) taking advantage of methods developed in Geographic Information Retrieval to extract and analyse information for a range of purposes beyond search.

Despite these notable advances, ensuring the reliability and applicability of methods continues to pose significant challenges. For example, the scarcity of digital resources and language complexity hinders extracting geographic information from minority languages. Similarly, in regions with limited coverage in geographic gazetteers, the accurate resolution of geographic references remains a formidable task and can increase existing biases. Moreover, resolving historical or ancient toponyms requires interdisciplinary approaches, considering place name evolution, spelling variations, and limited historical data. Understanding complex location descriptions, like 'behind the large mural, two blocks from City Hall, near the farmers market,' is particularly difficult due to their reliance on implicit local knowledge and spatial language ambiguity.

The recent emergence of Large Language Models (LLMs) such as GPT-4 offers promising advancements but also faces significant hurdles, including lack of transparency, inherent bias in training data, high computational needs, and difficulties in maintaining consistency across diverse languages and cultures. Lastly, there is a critical need for varied and sophisticated methodologies tailored to specific applications, which may include balancing accuracy with processing speed, offering multilingual support, or adapting to constraints in computational resources.

This special issue provides a central platform for scholars and researchers to convene and explore the extraction and analysis of geographic information found in texts. While we emphasize LLMs' pivotal role, our primary focus is to facilitate inclusive discussions, promote innovative research, and invite valuable contributions from experts active across diverse fields in this domain.

Topics Include, But Are Not Limited To:

• New approaches in document geocoding, toponym recognition, and toponym resolution to enhance reliability and applicability
• Scalability and resource-efficient use of LLMs in geographic information extraction and analysis
• Optimization of resource utilization and computational efficiency in geographic information extraction and analysis pipelines
• Multimodal data fusion for method improvement
• Geographic information extraction from underrepresented languages and regions
• Interdisciplinary approaches for historical and ancient toponym interpretation
• Advanced techniques for disambiguating complex location descriptions and spatial relations
• Representation and analysis of geographic information in texts
• Novel applications of text-based geographic information

 

References

Jones, C. B., & Purves, R. S. (2008). Geographical information retrieval. International Journal of Geographical Information Science, 22(3), 219-228.

Murrieta-Flores, P., & Martins, B. (2019). The geospatial humanities: past, present and future. International Journal of Geographical Information Science, 33(12), 2424-2429.

Stock, K., Jones, C. B., & Tenbrink, T. (2022). Speaking of location: a review of spatial language research. Spatial Cognition & Computation, 22(3-4), 185-224.

This IJGIS special issue welcomes submissions by scholars from all disciplines. Interested authors should first submit a short abstract (250 words max) to Xuke Hu ([email protected]) before April 30, 2024. Guest editors will review the submitted abstracts and evaluate whether the submissions fit the themes of this SI. Authors of abstracts with suitable topics will be invited to submit full manuscripts, while the invitation does not guarantee acceptance to the SI.

Full manuscripts, including any supporting materials, should be submitted using the journal's submission portal available at  http://mc.manuscriptcentral.com/ijgis  by September 30, 2024, and the authors should specify this SI as the target during their submission. For more information see the Instructions for Authors.   Please pay special attention to the required data and codes availability statement.

The International Journal of Geographical Information Science considers all manuscripts on the strict condition that they have been submitted only to the International Journal of Geographical Information Science, that they have not been published already, nor are they under consideration for publication or in press elsewhere. IJGIS exercises double-anonymous peer reviews. Authors should only deposit anonymous manuscripts to online depositories. IJGIS cannot consider manuscripts with author information available online.

Important Dates

• Abstract (no more than 250 words) Deadline: April 30, 2024
• Full Manuscripts Deadline: September 30, 2024

Instructions for AuthorsSubmit an Article