Geographic Information Retrieval Research Articles

Empowering geoportals HCI with task-oriented chatbots through NLP and deep transfer learning

ABSTRACT In the past ten years, chatbot development has matured to become one of the most well-distinguished outcomes of artificial intelligence. Despite some criticism, Bing AI, ChatGPT and other natural language processing (NLP) products of similar nature are becoming more popular. The creation of chatbots can close several gaps in geographic information retrieval as well. This research introduces and successfully implements, for the first time, a model for integrating task-oriented chatbots into geoportals, with the goal of easing user requests, improving access to geospatial services, and fostering human-computer interactions (HCI). Additionally, it presents a novel recommendation solution for matching the most appropriate volunteer to the user’s geospatial needs based on expertise similarity, semantic similarity, and community feedback. The three categories of finding map services, discovering geoprocessing services, and volunteer expert recommendations were shown to be the most significant geoportal bot intents. Depending on the requirement, each intent additionally includes various entities such as time, place, description, skill, etc. The notion of deep transfer learning (DTL) was then put into practice by customizing a pre-trained BERT (Bidirectional Encoder Representations from Transformers) model for our particular aim and creating a task-oriented conversational agent. According to the results, effective intent classification and entity recognition in the geospatial domain could arise from this approach. We performed the training process with 200 sample data, 20% of which were utilized in a stratified manner for testing, and we obtained f1-scores of at least 0.75. Finally, a pilot Geoportal Chabot that combines crowdsourcing and conversational agents’ approaches was put into use and tested with success. In keeping with SDI technical purposes, the system might direct users to common geospatial web services, namely WMS and WPS, in addition to including natural language understanding (NLU) and natural language generation (NLG) capabilities. Result of user-centered evaluation indicated that the integration of a chatbot significantly reduces the average time required to access geospatial data and processing services by more than 50%. Notably, this effect is even more pronounced when locating an expert, with a fivefold decrease in the time required. Finally, overall user satisfaction rose from 86% to 94%.

A deep neural network model for Chinese toponym matching with geographic pre-training model

ABSTRACT Multiple tasks within the field of geographical information retrieval and geographical information sciences necessitate toponym matching, which involves the challenge of aligning toponyms that share a common referent. The multiple string similarity approaches struggle when confronted with the complexities associated with unofficial and/or historical variants of identical toponyms. Also, current state-of-the-art approaches/tools to supervised machine learning rely on labeled samples, and they do not adequately address the intricacies of character replacements either from transliterations or historical shifts in linguistic and cultural norms. To address these issues, this paper proposes a novel matching approach that leverages a deep neural network model empowered by geographic language representation model, known as GeoBERT, which stands for geographic Bidirectional Encoder Representations from Transformers (BERT). This model harnesses the groundbreaking capabilities of the GeoBERT framework by extending a generalized Enhanced Sequential Inference Model architecture and integrating multiple features to enhance the accuracy and robustness of the toponym matching. We present a comprehensive evaluation of the proposed method’s performance using three extensive datasets. The findings clearly illustrate that our approach outperforms the individual similarity metrics used in previous studies.

Get Spatial from Non-Spatial Information: Inferring Spatial Information from Textual Descriptions by Conceptual Spaces

With the rapid growth of social media, textual content is increasingly growing. Unstructured texts are a rich source of latent spatial information. Extracting such information is useful in query processing, geographical information retrieval (GIR), and recommender systems. In this paper, we propose a novel approach to infer spatial information from salient features of non-spatial nature in text corpora. We propose two methods, namely DCS and RCS, to represent place-based concepts. In addition, two measures, namely the Shannon entropy and the Moran’s I, are proposed to calculate the degree of geo-indicativeness of terms in texts. The methodology is compared with a Latent Dirichlet Allocation (LDA) approach to estimate the accuracy improvement. We evaluated the methods on a dataset of rental property advertisements in Iran and a dataset of Persian Wikipedia articles. The results show that our proposed approach enhances the relative accuracy of predictions by about 10% in case of the renting advertisements and by 13% in case of the Wikipedia articles. The average distance error is about 13.3 km for the advertisements and 10.3 km for the Wikipedia articles, making the method suitable to infer the general region of the city in which a property is located. The proposed methodology is promising for inferring spatial knowledge from textual content that lacks spatial terms.

Location Reference Recognition from Texts: A Survey and Comparison

A vast amount of location information exists in unstructured texts, such as social media posts, news stories, scientific articles, web pages, travel blogs, and historical archives. Geoparsing refers to recognizing location references from texts and identifying their geospatial representations. While geoparsing can benefit many domains, a summary of its specific applications is still missing. Further, there is a lack of a comprehensive review and comparison of existing approaches for location reference recognition, which is the first and core step of geoparsing. To fill these research gaps, this review first summarizes seven typical application domains of geoparsing: geographic information retrieval, disaster management, disease surveillance, traffic management, spatial humanities, tourism management, and crime management. We then review existing approaches for location reference recognition by categorizing these approaches into four groups based on their underlying functional principle: rule-based, gazetteer matching–based, statistical learning-–based, and hybrid approaches. Next, we thoroughly evaluate the correctness and computational efficiency of the 27 most widely used approaches for location reference recognition based on 26 public datasets with different types of texts (e.g., social media posts and news stories) containing 39,736 location references worldwide. Results from this thorough evaluation can help inform future methodological developments and can help guide the selection of proper approaches based on application needs.

TopoBERT: a plug and play toponym recognition module harnessing fine-tuned BERT

ABSTRACT Extracting precise geographical information from the textual content, referred to as toponym recognition, is fundamental in geographical information retrieval and crucial in a plethora of spatial analyses, e.g. mining location-based information from social media, news reports, and surveys for various applications. However, the performance of existing toponym recognition methods and tools is deficient in supporting tasks that rely on extracting fine-grained geographic information from texts, e.g. locating people sending help requests with addresses through social media during disasters. The emerging pretrained language models have revolutionized natural language processing and understanding by machines, offering a promising pathway to optimize toponym recognition to underpin practical applications. In this paper, TopoBERT, a uniquely designed toponym recognition module based on a one-dimensional Convolutional Neural Network (CNN1D) and Bidirectional Encoder Representation from Transformers (BERT), is proposed and fine-tuned. Three datasets are leveraged to tune the hyperparameters and discover the best strategy to train the model. Another seven datasets are used to evaluate the performance. TopoBERT achieves state-of-the-art performance (average f1-score = 0.854) compared to the seven baseline models. It is encapsulated into easy-to-use python scripts and can be seamlessly applied to diverse toponym recognition tasks without additional training.

NeuroSPE: A neuro‐net spatial relation extractor for natural language text fusing gazetteers and pretrained models

AbstractSpatial relations are frequently described and used in natural language texts, and relations play a core role in a range of applications—from supporting geographic information retrieval in natural language texts to locating people and objects in natural disaster response situations. In this article, we present a neuro‐net spatial extraction model (NeuroSPE) designed to address various language irregularities (i.e., a variety of sentence structures) that occur in natural language texts. We also propose a two‐stage workflow to generate a training dataset based on a collection of words and their associated frequencies. The first stage of the proposed workflow focuses on processing the words in the input data and their associated frequencies; then, the words are segmented into a set of groups and used to accelerate model training. The second stage automatically generates a variety of sentences that include two geographic entities and related spatial relation terms through deep learning iteration based on a unigram language model. We evaluate our method both qualitatively and quantitatively using a real dataset. The experimental results demonstrate that the proposed two‐stage workflow effectively extracts spatial relations from natural language texts and outperforms other current state‐of‐the‐art approaches.

Flexible Trip-Planning Queries

The current practice of users searching for different types of geo-resources in a geographic area and wishing to identify the most convenient routes for visiting the most relevant ones, requires the iterative formulation of several queries: first to identify the more interesting resources and then to select the best route to visit them. In order to simplify this process, in this paper a novel functionality for a geographic information retrieval (GIR) system is proposed, which retrieves and ranks several routes for visiting a number of relevant georeferenced resources as a result of a single query, named flexible trip-planning query. An original retrieval model is defined to identify the relevant resources and to rank the most convenient routes by taking into account personal user preferences. To this end, a graph-based algorithm is defined, exploiting prioritized aggregation to optimize the routes’ identification and ranking. The proposed algorithm is applied in the proof-of-concept of a Smart cOmmunity-based Geographic infoRmation rEtrievAl SysTem (SO-GREAT) designed to strengthen local communities: it collects and manages open data from regional authorities describing categories of authoritative territorial resources and services, such as schools, hospitals, etc., and from volunteered geographic services (VGSs) created by citizens to offer services in their neighbourhood.

Geographic Named Entity Recognition by Employing Natural Language Processing and an Improved BERT Model

Toponym recognition, or the challenge of detecting place names that have a similar referent, is involved in a number of activities connected to geographical information retrieval and geographical information sciences. This research focuses on recognizing Chinese toponyms from social media communications. While broad named entity recognition methods are frequently used to locate places, their accuracy is hampered by the many linguistic abnormalities seen in social media posts, such as informal sentence constructions, name abbreviations, and misspellings. In this study, we describe a Chinese toponym identification model based on a hybrid neural network that was created with these linguistic inconsistencies in mind. Our method adds a number of improvements to a standard bidirectional recurrent neural network model to help with location detection in social media messages. We demonstrate the results of a wide-ranging evaluation of the performance of different supervised machine learning methods, which have the natural advantage of avoiding human design features. A set of controlled experiments with four test datasets (one constructed and three public datasets) demonstrates the performance of supervised machine learning that can achieve good results on the task, significantly outperforming seven baseline models.

Detecting common features from point patterns for similarity measurement using matrix decomposition

ABSTRACT Similarity of point patterns is critical to geographic information retrieval. Many methods depend on measuring the similarity between point patterns within the spatial database. However, previous researches mainly focus on point density which is only one aspect of point patterns. A point distribution can be complex by its numerous alignments of point groups, which usually imply different geographical meanings in reality. In this paper, we propose a new method that uses image analysis techniques to comprehensively consider the characteristics of a point pattern. Specifically, given a set of point datasets falling in the same region, our method first generates the point intensity surfaces to transform the original vector space to raster space; then, the method constructs a matrix to describe all the pattern-related information. Finally, the point pattern similarity is calculated by decomposing this matrix into the lower-order representation and the factorized basis features. Due to the use of matrix decomposition, the proposed method has the merits that it can eliminate noises from the original data and assess the similarity of two patterns with emphasis on their major features. As a case study, our method is effective in discovering regularity from the taxi pick-up/drop-off point datasets.

Spatial Relationship Extraction of Geographic Entities Based on BERT Model

Geographic entity relationship extraction from text is an important way to acquire geographic knowledge. Entity relations in Chinese text are difficult to discover because of implicit representations of relations between entities in Chinese text. Therefore, using existing pattern matching and machine learning methods to extract entity relations from Chinese text often has problems such as insufficient artificial features, poor generality, inability to resolve word polysemy, and difficulty in making full use of contextual information. However, deep learning methods can better solve the above problems. This paper takes the spatial relation between geographic entities as the main research object to build a geographic entity spatial relation corpus, and propose a geographic entity relationship extraction method that combines BERT and attention mechanism. The method is trained and tested on the corpus. The results show that the F1 value of the BERT-BiGRU-Attention model proposed in this paper reaches 85.2% on the test set, which is a great improvement in relation extraction ability compared with other baseline models. The geographic entity relationship extraction method based on the BERT model can effectively learn the context information in Chinese text sentences, improve the accuracy of relationship extraction, and provide support for geographic knowledge graph construction and geographic entity information retrieval.

New Era for Geo-Parsing to Obtain Actual Locations: A Novel Toponym Correction Method Based on Remote Sensing Images

Geo-parsing, one of the key components of geographical information retrieval, is a process to recognize and geo-locate toponyms mentioned in texts. Such a process can obtain locations contained in toponyms successfully with consistent updating of neural network models and multiple contextual features. The significant offset distance between the geo-parsed locations and the actual occurrence locations still remains. This is because the geo-parsed locations sourced from toponyms in texts always point to the centers of cities, counties, or towns, and cannot directly represent the actual occurrence locations such as factories, farms, and activity areas. Consequently, The significant offset distances between the geo-parsed locations and the actual occurrence locations limit text mining applications in micro-scale geographic discoveries. This research aims at decreasing offset distances of geo-parsed locations by proposing a novel Toponym Correction Method based on satellite Remote Sensing Images (TC-RSI). The TC-RSI method uses satellite remote sensing images to provide extra detailed spatial information that can be associated with the sentence toponym by corresponding attributes. The TC-RSI method was validated in a case study of the forest ecological pattern dataset of An’hui province from visual, statistical, and robustness assessments. The correction results show that the TC-RSI method dramatically decreases the offset distances from about 50 km to about 1 km and promotes geographical discoveries on smaller scales. A series of analyses indicated that the TC-RSI is a valid, effective, and promising method to improve the accuracy of geo-parsed locations, which allows text mining to find more accurate geographical discoveries with lower offset distances. Moreover, toponym correction promotes the use of more diverse spatial data sources, such as Lidar, domain gazetteers, Wikimedia, and streetscapes, which are expected to usher in a new era of geo-parsing with toponym corrections.

DeepSSN: A deep convolutional neural network to assess spatial scene similarity

AbstractSpatial‐query‐by‐sketch is an intuitive tool to explore human spatial knowledge about geographic environments and to support communication with scene database queries. However, traditional sketch‐based spatial search methods perform inadequately due to their inability to find hidden multiscale map features from mental sketches. In this research, we propose a deep convolutional neural network, namely the Deep Spatial Scene Network (DeepSSN), to better assess the spatial scene similarity. In DeepSSN, a triplet loss function is designed as a comprehensive distance metric to support the similarity assessment. A positive and negative example mining strategy is designed to ensure a consistently increasing distinction of triplets during the training process. Moreover, we develop a prototype spatial scene search system using the proposed DeepSSN, in which the users input spatial queries via sketch maps and the system can automatically augment the sketch training data. The proposed model is validated using multisource conflated map data including 131,300 labeled scene samples after data augmentation. The empirical results demonstrate that the DeepSSN outperforms baseline methods including k‐nearest neighbors, the multilayer perceptron, AlexNet, DenseNet, and ResNet using mean reciprocal rank and precision metrics. This research advances geographic information retrieval studies by introducing a novel deep learning method tailored to spatial scene queries.

Geometric Similarity Measurement Method for Micro Scene Generalization

Geometric similarity plays an important role in geographic information retrieval, map matching, and data updating. Many approaches have been developed to calculate the similarity between simple features. However, complex group objects are common in map and spatial database systems. With a micro scene that contains different types of geographic features, calculating similarity is difficult. In addition, few studies have paid attention to the changes in a scene’s geometric similarity in the process of generalization. In this study, we developed a method for measuring the geometric similarity of micro scene generalization based on shape, direction, and position. We calculated shape similarity using the hybrid feature description, and we constructed a direction Voronoi diagram and a position graph to measure the direction similarity and position similarity. The experiments involved similarity calculation and quality evaluation to verify the usability and effectiveness of the proposed method. The experiments showed that this approach can be used to effectively measure the geometric similarity between micro scenes. Moreover, the proposed method accounts for the relationships amongst the geometrical shape, direction, and position of micro scenes during cartographic generalization. The simplification operation leads to obvious changes in position similarity, whereas delete and merge operations lead to changes in direction and position similarity. In the process of generalization, the river + islands scene changed mainly in shape and position, the similarity change in river + lakes occurred due to the direction and location, and the direction similarity of rivers + buildings and roads + buildings changed little.

Location Based Navigation Service Technology: Development of an Optimization Algorithm for Multivariate Contextual Cartographic Content in Location Based Navigation Service

User response or reaction to navigation applications is influenced by relevance in geographic information, in terms of cartographic context and content delivered within a definite time, providing a direct impact to outcome or consequence based on decision making and hence user reaction. Location Based Navigation Services (LBNS) have continuously advanced in cartographic visualization, making maps interpretation easy and ubiquitous to any user, as compared to pre-historic times when maps were a preserve of a few. Despite rapid growth in LBNS, there exist challenges that may be characterized as technical and non-technical challenges, among them being process of conveying geospatial information to user. LBNS system deliver appropriate information to a user through smartphone (mobile device) for effective decision making and response within a given time span. This research focuses on optimization of cartographic content for contextual information in LBNS to users, based on prevailing circumstances of various components that constitute it. The research looks into Geographic Information Retrieval (GIR), as a technical challenge centered on a non-technical issue of social being of user satisfaction, leading to decision making in LBNS, hence response and outcome. Though advanced technologically, current LBNS on information sourcing depends on user manual web pages navigation and maneuver, this can be painstaking and time consuming that it may cause unnecessary delay in information delivery, resulting to delayed information response time (DIRT). This in turn may lead to unappropriate decision making with erroneous reaction or response being taken, resulting in loss of opportunity, resources, time and even life. Optimization in LBNS is achieved by a mathematical relationship developed between user status, mobile device variables against cartographic content. The relationship is in turn applied in LBNS android application to fulfill optimization solution for user consumption.

The Integration of Linguistic and Geospatial Features Using Global Context Embedding for Automated Text Geocoding

Geocoding is an essential procedure in geographical information retrieval to associate place names with coordinates. Due to the inherent ambiguity of place names in natural language and the scarcity of place names in textual data, it is widely recognized that geocoding is challenging. Recent advances in deep learning have promoted the use of the neural network to improve the performance of geocoding. However, most of the existing approaches consider only the local context, e.g., neighboring words in a sentence, as opposed to the global context, e.g., the topic of the document. Lack of global information may have a severe impact on the robustness of the model. To fill the research gap, this paper proposes a novel global context embedding approach to generate linguistic and geospatial features through topic embedding and location embedding, respectively. A deep neural network called LGGeoCoder, which integrates local and global features, is developed to solve the geocoding as a classification problem. The experiments on a Wikipedia place name dataset demonstrate that LGGeoCoder achieves competitive performance compared with state-of-the-art models. Furthermore, the effect of introducing global linguistic and geospatial features in geocoding to alleviate the ambiguity and scarcity problem is discussed.

Crowdsourced Data Relevance Analysis for Crowd-assisted Flood Disaster Management

<p class="Default">The recent climate changes have significantly increased the number and intensity of natural disasters around the world. This includes floods that cause a great deal of damage to properties, and more importantly, to the lives of the people. The reporting of current disasters has changed from official media to public reporters through social media and crowdsourcing technologies which have guaranteed the availability and up-to-date nature of the reported data. However, crowdsourced data (CSD) are often questioned due to issues in reliability and relevancy, heterogeneity or bias, bad structure, and un-professionalism. As a result of this, disaster responders are reluctant to use such data for their critical decision making actions. Using Natural Language Processing (NLP) and Geographic Information Retrieval (GIR) techniques, this study evaluated the quality of CSD, focusing on the thematic relevance. The study examined a proof of concept on relevance assessment based on an improved set of user queries utilizing crowdsourced messages from the 2011 Australian floods (Ushahidi Crowd-map). The findings show that the approach was effective in generating a thematically rated list of CSD messages for post-flood disaster managers to confidently take actions. The study's future work will consider thematic and geographic specificities and semantic context of the modified queries. Moreover, it is expected to test the approach with similar geospatial crowd-map data, and finally to check the possibility of integrating the derived information with authoritative datasets such as Spatial Data Infrastructures (SDIs).

Assessing the influence of point-of-interest features on the classification of place categories

Points of interest (POIs) digitally represent real-world amenities as point locations. POI categories (e.g. restaurant, hotel, museum etc.) play a prominent role in several location-based applications such as social media, navigation, recommender systems, geographic information retrieval tools, and travel-related services. The majority of user queries in these applications center around POI categories. For instance, people often search for the closest pub or the best value-for-money hotel in an area. To provide valid answers to such queries, accurate and consistent information on POI categories is an essential requirement. Nevertheless, category-based annotations of POIs are often missing. The task of annotating unlabeled POIs in terms of their categories — known as POI classification — is commonly achieved by means of machine learning (ML) models, often referred to as classifiers. Central to this task is the extraction of known features from pre-labeled POIs in order to train the classifiers and, then, use the trained models to categorize unlabeled POIs. However, the set of features used in this process can heavily influence the classification results. Research on defining the influence of different features on the categorization of POIs is currently lacking. This paper contributes a study of feature importance for the classification of unlabeled POIs into categories. We define five feature sets that address operation based, review-based, topic-based, neighborhood-based, and visual attributes of POIs. Contrary to existing studies that predominantly use multi-class classification approaches, and in order to assess and rank the influence of POI features on the categorization task, we propose both a multi-class and a binary classification approach. These, respectively, predict the place category among a specified set of POI categories, or indicate whether a POI belongs to a certain category. Using POI data from Amsterdam and Athens to implement and evaluate our study approach, we show that operation based features, such as opening or visiting hours throughout the day, are the most important place category predictors. Moreover, we demonstrate that the use of feature combinations, as opposed to the use of individual features, improves the classification performance by an average of 15%, in terms of F1-score.

Event Geoparser with Pseudo-Location Entity Identification and Numerical Argument Extraction Implementation and Evaluation in Indonesian News Domain

Geoparser is a fundamental component of a Geographic Information Retrieval (GIR) geoparser, which performs toponym recognition, disambiguation, and geographic coordinate resolution from unstructured text domain. However, geoparsing of news articles which report several events across many place-mentions in the document are not yet adequately handled by regular geoparser, where the scope of resolution is either toponym-level or document-level. The capacity to detect multiple events and geolocate their true coordinates along with their numerical arguments is still missing from modern geoparsers, much less in Indonesian news corpora domain. We propose an event geoparser model with three stages of processing, which tightly integrates event extraction model into geoparsing and provides precise event-level resolution scope. The model casts the geotagging and event extraction as sequence labeling and uses LSTM-CRF inferencer equipped with features derived using Aggregated Topic Model from a large corpus to increase the generalizability. Throughout the proposed workflow and features, the geoparser is able to significantly improve the identification of pseudo-location entities, resulting in a 23.43% increase for weighted F1 score compared to baseline gazetteer and POS Tag features. As a side effect of event extraction, various numerical arguments are also extracted, and the output is easily projected to a rich choropleth map from a single news document.

Mapping Mediascapes: Exploring the Semantic and Geographic Spaces of Catalonia’s 2015 Independence Movement through Digital News

Digital media merges disparate places, communities, and topics in reporting about current events, with both local and international impacts. This study maps the semantic spaces of reporting on Catalonia’s independence movement in 2015 onto geographic space, forming a “mediascape” of interwoven themes and places. It uses topic modeling to determine semantic themes and geographic information retrieval to map the geographies of those themes. A global pattern of distance decay from Barcelona, the center of the Catalonian independence movement, is apparent. This study also compares the particular media themes of professional sport, which is a globalizing force in Catalonia’s geopolitical narrative, and the Scottish independence referendum of 2014. It examines the particular geographies that appear in news articles pertaining to each theme. Contrasting the geographies of Catalonian independence and its relation to soccer and Scottish independence through the media produces a spatial–semantic illustration of how mediascapes form from and influence global geopolitical connections.

Measuring the similarity between multipolygons using convex hulls and position graphs

ABSTRACT Polygon similarity can play an important role in geographic information retrieval, map matching and updating, and spatial data mining applications. Geographic information science (GIS) represents various spatial objects as polygons, including simple polygons and polygons with holes, as well as multipolygons. Spatial objects of multipolygons possess complex structure which makes it difficult to assess their similarity. This study develops a method based on convex hulls and position graphs to measure the similarity between multipolygons. The proposed method first finds correspondences between subpolygons in the two multipolygons based on a control polygon. Thereafter, the method constructs a position graph to denote the distribution of these subpolygons and applies a turning function to compute the similarity between various graphs. Fourier transformation and moment invariants were combined to characterize the different matching relationships among subpolygons. The experiments involve three different kinds multipolygons to verify the effectiveness and robustness of proposed method. The experiments show that this approach effectively measures similarity between multipolygons. Moreover, the proposed method accounts for the relationships across the entire complex geometrical shape and components of multipolygon during measuring similarity.