Qualitative Spatial Representation Research Articles

Tracing Trajectories: Qualitative Visualizations of Migrant Death in South Texas, 1993–2020

Executive Summary This paper explores how qualitative information may be used to enhance understanding and inform recommendations for improved accounting of migrant deaths along the Texas-Mexico border. While border crossing related deaths affect jurisdictions throughout the US’s Southwestern border states and Northern Mexican states, Texas has unique challenges that merit specific examination. In short, the management and investigation of unidentified migrant decedents in Texas is severely fragmented and often noncompliant with Texas Criminal Code Procedure. I explore a way to improve accounting of migrant deaths leveraging qualitative information, local newspaper articles from South Texas and Northern Mexico, by using descriptive summaries coupled with Sankey Diagrams and a programmatic technique, qualitative spatial representation (QSR). QSR enabled me to identify under-recognized stakeholders (South Texas locals, Mexican consulates) and under-supported counties (i.e., those with migrant deaths that do not share a border directly with Mexico). I found that local English-language newspapers obscured the prevalence of migrant deaths and that their narrative tone of “business as usual” normalized the occurrence of such deaths. The Spanish-language articles better represented the diversity of agencies and individuals that were involved in the various aspects of migrant remains management and forensic investigation, most notably residents of South Texas themselves (or situational participants who first found the remains) and Mexican consulates. Finally, I noted that the collaborations among Texas counties and between Texan and Mexican jurisdictions, when they were described in the newspapers, highlighted that the phenomenon of migrant deaths penetrates beyond the dividing line of the Rio Grande/Río Bravo River. Migrant deaths must be accounted for in a way that reflects a regional experience in which people may perish as far as ~100 miles into Texas’ interior. Considering these observations, I have two policy recommendations. First, the Office of the Texas Governor should establish funds distinct from the Operation Lone Star program for the management of unidentified human remains. Currently, grants through the Operation Lone Star Program are the only funds available in Texas to support identification of migrant decedents. However, this program is explicitly designed to further border security operations in the state of Texas, which can contradict efforts around recovery and identification. The documentation and forensic investigation of the unidentified deceased can be eclipsed or neglected under the larger deterrence aims of Operation Lone Star. Second, to improve accounting and increase documentation of migrant deaths, a Regional Identification Center should be established in South Texas. Not every county in Texas has or has access to a Medical Examiner’s office, including the means to transport and pay for autopsy or other forensic services. The Center would provide training, store extra equipment such as mobile refrigerated morgues, and hire personnel to inventory cases, and report information to the state (e.g., vital statistics) and to foreign consulates. A Regional Identification Center would counteract the challenges highlighted in the local newspaper summaries and QSR by decreasing the isolation of counties experiencing migrant deaths and serve as a documentary and communication hub for stakeholders in Texas and Mexico.

Object-agnostic Affordance Categorization via Unsupervised Learning of Graph Embeddings

Acquiring knowledge about object interactions and affordances can facilitate scene understanding and human-robot collaboration tasks. As humans tend to use objects in many different ways depending on the scene and the objects’ availability, learning object affordances in everyday-life scenarios is a challenging task, particularly in the presence of an open set of interactions and objects. We address the problem of affordance categorization for class-agnostic objects with an open set of interactions; we achieve this by learning similarities between object interactions in an unsupervised way and thus inducing clusters of object affordances. A novel depth-informed qualitative spatial representation is proposed for the construction of Activity Graphs (AGs), which abstract from the continuous representation of spatio-temporal interactions in RGB-D videos. These AGs are clustered to obtain groups of objects with similar affordances. Our experiments in a real-world scenario demonstrate that our method learns to create object affordance clusters with a high V-measure even in cluttered scenes. The proposed approach handles object occlusions by capturing effectively possible interactions and without imposing any object or scene constraints.

Report of the 6th ACM SIGSPATIAL International Workshop on Geospatial Humanities (GeoHumanities 2022)

The field of Geospatial Humanities addresses the use of geographic information systems and other spatial technologies in humanities research, looking to address questions related to space and place. The field is constantly evolving, and currently it is posing very interesting challenges relating, for example, to the identification and analysis of real, vague, and imaginary space in humanities sources like archival manuscripts, maps, encyclopedias, newspapers, correspondence collections, and more. These kinds of documents pose new challenges for identifying and analyzing spatial information, and a strong emphasis is being put on new methodologies that leverage the standard tools from geographic information systems, as well as more advanced methods (e.g., based on machine learning) such as text- and image-based geographical analysis, this way combining GIS, NLP, Deep Mapping, Computer Vision, and Qualitative Spatial Representation, among others.

Towards a Cognitive System that Can Recognize Spatial Regions Based on Context

In order to collaborate with people in the real world, cognitive systems must be able to represent and reason about spatial regions in human environments. Consider the command "go to the front of the classroom". The spatial region mentioned (the front of the classroom) is not perceivable using geometry alone. Instead it is defined by its functional use, implied by nearby objects and their configuration. In this paper, we define such areas as context-dependent spatial regions and present a cognitive system able to learn them by combining qualitative spatial representations, semantic labels, and analogy. The system is capable of generating a collection of qualitative spatial representations describing the configuration of the entities it perceives in the world. It can then be taught context-dependent spatial regions using anchor pointsdefined on these representations. From this we then demonstrate how an existing computational model of analogy can be used to detect context-dependent spatial regions in previously unseen rooms. To evaluate this process we compare detected regions to annotations made on maps of real rooms by human volunteers.

Qualitative case-based reasoning and learning

The development of autonomous agents that perform tasks with the same dexterity as performed by humans is one of the challenges of artificial intelligence and robotics. This motivates the research on intelligent agents, since the agent must choose the best action in a dynamic environment in order to maximise the final score. In this context, the present paper introduces a novel algorithm for Qualitative Case-Based Reasoning and Learning (QCBRL), which is a case-based reasoning system that uses qualitative spatial representations to retrieve and reuse cases by means of relations between objects in the environment. Combined with reinforcement learning, QCBRL allows the agent to learn new qualitative cases at runtime, without assuming a pre-processing step. In order to avoid cases that do not lead to the maximum performance, QCBRL executes case-base maintenance, excluding these cases and obtaining new (more suitable) ones. Experimental evaluation of QCBRL was conducted in a simulated robot-soccer environment, in a real humanoid-robot environment and on simple tasks in two distinct gridworld domains. Results show that QCBRL outperforms traditional RL methods. As a result of running QCBRL in autonomous soccer matches, the robots performed a higher average number of goals than those obtained when using pure numerical models. In the gridworlds considered, the agent was able to learn optimal and safety policies.

Text mining and semantic triples: Spatial analyses of text in applied humanitarian forensic research

The methods and tools of Geographic Information Sciences (GIScience)—spatial analysis, spatial statistics, and geographic information technologies—are increasingly being used in forensic humanitarian projects. In this article we explore ways to parse and analyze social and media releases from the United States Customs and Border Patrol (CBP) to gain an understanding of the death of migrants at the Texas-Mexico border. The methods we used include corpus linguistic (CL)/natural language processing (NLP) and Qualitative Spatial Representation (QSR) and Semantic Triples (ST). Our results indicate that CL/NLP and QSR/ST have the potential to increase and improve deceased migrant case identification by providing a framework for searching for key terms or themes throughout multiple textual sources. In the specific case examined, however, CL/NLP showed that CBP social media focused on drug confiscation and general patrolling activities and were of limited use for tabulating incidences of migrant death. On the other hand, QSR/ST visualizations showed which CBP Stations most frequently reported deceased migrant recoveries (i.e., search and collection of human remains) and with whom they collaborated.We believe these methods are part of the methodological toolkit needed to lay the ground for what we call Humanitarian GIS—the application of spatial analytical perspectives and tools to genocide studies, spatial forensics, and, in general, human rights topics and events. Within this toolkit, CL/NLP and QSR/ST highlight spatial relationships that are not necessarily mappable in a traditional GIS setting and allow researchers to detect patterns across large corpora of heterogeneous information. The mixing of methodologies and the combination of qualitative and quantitative data in Humanitarian GIS may change our understanding of an ongoing humanitarian crisis and aid in improving and increasing response to such crises.

Qualitative Spatial Representation for the Humanities

‘Qualitative spatial reasoning and representation’ is a range of techniques developed in Artificial Intelligence to meet the need for a computational treatment of qualitative spatial relations. Examples of such relations include ‘next to’, ‘overlapping’, ‘to the left of’, ‘separate from’, ‘including’, and so on. These relations occur within the data found in the spatial humanities, but the computational techniques described here do not appear to have been used in connection with this context. While Geographical Information Systems (GIS) are widely used as a means of visualizing and exploring material in the spatial humanities, GIS technology is acknowledged to be ill-suited to information that is vague, uncertain, ambiguous, imprecise or having other qualities that in a scientific setting could be regarded as imperfections. In the humanities such ‘imperfections’ are of course important, and qualitative spatial relations are one source of data that challenges scientifically based GIS. This article reviews the origin of qualitative spatial reasoning and representation in A. N. Whitehead's mereotopology and argues for exploring how these methods could complement GIS as a computational technique in the humanities. Qualitative representation is applicable to modelling spatial arrangements in many domains, not just geographical space. This is demonstrated through an example of spatial relations in lines of printed text.

Geographies of the Holocaust: Experiments in GIS, QSR, and Graph Representations

This article responds to the widespread and oft-noted challenges digital humanists face in working with data that is uncertain and characterised by complex narratives. Using an example drawn from the vast archives of post-war interviews with Holocaust survivors, it draws on approaches developed in Qualitative Spatial Representation (QSR) to explore how two key spatial aspects of survivor's narratives – their uncertain wartime trajectories and the slippage in scales as these are retold – can be represented. Spatial information in narratives tends to not provide the exact coordinates necessary to store the information in geospatial databases. Instead, narratives rely much more on often less precise qualitative spatial relations such as ‘near’, ‘next to’, ‘at the corner of’ without precise geometric interpretations. Given that relational databases are ill-equipped to store this kind of relational spatial knowledge from natural language sources, the article argues for a need for digital humanists to return to first principles and reconsider database design. Descriptive triple-based graph representations, which have been devised to accommodate this kind of highly irregular, semi-structured relational knowledge, have the potential to work with, rather than against, the grain of the narrative sources that underlie the work of much of the digital humanities.

Spatial Humanities 3.0: Qualitative Spatial Representation and Semantic Triples as New Means of Exploration of Complex Indigenous Spatial Representations in Sixteenth Century Early Colonial Mexican Maps

For some time now, the field of Spatial Humanities has acknowledged the need for a system capable of the spatial exploration of historical and archaeological phenomena beyond Geographic Information Systems (GIS). This idea comes from the need to analyse holistically spatial information, including that which is not geographic (i.e. vague, symbolic and imaginary space). In addition, this need becomes more apparent when dealing with traditions that do not conform to the Modern/European/Cartesian conception of space in which GIS is rooted. This article, explores the use of Qualitative Spatial Representation (QSR) and Semantic Triples as possible alternative means to model complex and diverse expressions of spatial information, including social and symbolic conceptions in 16th century Mexican maps. Using as case study the map from the region of Atengo-Misquiahuala (Hidalgo) which combine the Mesoamerican and European traditions, we explore how these approaches might open new venues of research, potentially shedding light to long discussed and problematic Mesoamerican spatial categories. Focusing on a contained and partial example, we examine from a theoretical perspective and as a starting point, the possible future implementation of these approaches for historical and archaeological research.

Qualitative Spatial Representation: A Special Section

Qualitative Spatial Representation: A Special Section

Formal representation of qualitative direction

ABSTRACTThis paper reviews formal approaches to representing spatial knowledge about qualitative direction. Unlike geometric direction information, qualitative information does not employ numerical values but relies on comparison. The qualitative approach is often regarded as suitable for capturing commonsense concepts and thus is relevant to human-centered interfaces for spatial information systems. To establish a context for the work on qualitative direction, we preset a brief history of the development of qualitative temporal and spatial representations from different scientific perspectives. We identify main focal areas of these representations of spatial direction and propose a taxonomy. In the light of more than three decades of fruitful research, we obtain a map of formal representations that reveal interrelationships between different research strands in the field.

A qualitative spatial representation of string loops as holes

This research note contains an extension of a previous work by Cabalar and Santos (2011) that formalised several spatial puzzles formed by strings and holes. That approach explicitly ignored some configurations and actions that were irrelevant for the studied puzzles but are physically possible and may become crucial for other spatial reasoning problems. In particular, the previous work did not consider the formation of string loops or the situations where a holed object is partially crossed by another holed object. In this paper, we remove these limitations by treating string loops as dynamic holes that can be created or destroyed by a pair of elementary actions, respectively picking or pulling from strings. We explain how string loops can be recognised in a data structure representing the domain states and define a notation to represent crossings through string loops. The resulting formalism is dual in the sense that it also allows understanding any hole as a kind of (sometimes rigid) closed string loop.

Bridging qualitative spatial constraints and feature-based parametric modelling: Expressing visibility and movement constraints

We present a concept for integrating state-of-the-art methods in geometric and qualitative spatial representation and reasoning with feature-based parametric modelling systems. Using a case-study involving a combination of topological, visibility, and movement constraints, we demonstrate the manner in which a parametric model may be constrained by the spatial aspects of conceptual design specifications and higher-level semantic design requirements. We demonstrate the proposed methodology by applying it to architectural floor plan layout design, where a number of spaces with well defined functionalities have to be arranged such that particular functional design constraints are maintained. The case-study is developed by an integration of the declarative spatial reasoning systemCLP(QS) (CLP(QS) – a declarative spatial reasoning system. www.spatial-reasoning.com.) with the parametric CAD system FreeCAD.

A metaheuristic approach for efficient and effective sketch-to-metric map alignment

ABSTRACTSketching as a natural mode for human communication and creative processes presents opportunities for improving human–computer interaction in geospatial information systems. However, to use a sketch map as user input, it must be localized within the underlying spatial data set of the information system, the base metric map. This can be achieved by a matching process called qualitative map alignment in which qualitative spatial representations of the two input maps are used to establish correspondences between each sketched object and one or more objects in the metric map. The challenge is that, to the best of our knowledge, no method for matching qualitative spatial representations suggested so far is applicable in realistic scenarios due to excessively long runtimes, incorrect algorithm design or the inability to use more than one spatial aspect at a time. We address these challenges with a metaheuristic algorithm which uses novel data structures to match qualitative spatial representations of a pair of maps. We present the design, data structures and performance evaluation of the algorithm using real-world sketch and metric maps as well as on synthetic data. Our algorithm is novel in two main aspects. Firstly, it employs a novel system of matrices known as local compatibility matrices, which facilitate the computation of estimates for the future size of a partial alignment and allow several types of constraints to be used at the same time. Secondly, the heuristic it computes has a higher accuracy than the state-of-the-art heuristic for this task, yet requires less computation. Our algorithm is also a general method for matching labelled graphs, a special case of which is the one involving complete graphs whose edges are labelled with spatial relations. The results of our evaluation demonstrate practical runtime performance and high solution quality.

The cognition of change: scaling deformations in mind and spatial theories

The dynamics of the earth and its inhabitants have become a core topic and focus in research in the spatial sciences. The spatio-temporal data avalanche challenges researchers to provide efficient and effective means to process spatio-temporal data. It is of vital importance to develop mechanisms that allow for the transition of data not only into information but also into knowledge. Knowledge representation techniques from artificial intelligence play an important role in laying the foundations for theories dealing with spatio-temporal data. Specifically, the advances in the area of qualitative spatial representation and reasoning (QSTR) have led to promising results. Categorical distinctions of spatio-temporal information identified by QSTR calculi potentially correspond to those relevant to humans. This article presents the first behavioral evaluation of qualitative calculi modeling geographic events associated with scaling deformations of entities, that is, changes in size by either expansion or contraction. Examples of such dynamics include a lake flooding its surroundings or an expanding oil spill in the ocean. We compare four experiments using four different semantic domains. Each domain consists of two spatially extended entities: one entity is undergoing scaling deformations while the other is static. We kept the formal QSTR characterization, which are paths through a topologically defined conceptual neighborhood graph, identical across all semantic domains. Our results show that for geographic events associated with scaling deformations (a) topological relations are not equally salient cognitively; (b) domain semantics has an influence on the conceptual salience of topological relations.

Global machine learning for spatial ontology population

Understanding spatial language is important in many applications such as geographical information systems, human computer interaction or text-to-scene conversion. Due to the challenges of designing spatial ontologies, the extraction of spatial information from natural language still has to be placed in a well-defined framework. In this work, we propose an ontology which bridges between cognitive–linguistic spatial concepts in natural language and multiple qualitative spatial representation and reasoning models. To make a mapping between natural language and the spatial ontology, we propose a novel global machine learning framework for ontology population. In this framework we consider relational features and background knowledge which originate from both ontological relationships between the concepts and the structure of the spatial language. The advantage of the proposed global learning model is the scalability of the inference, and the flexibility for automatically describing text with arbitrary semantic labels that form a structured ontological representation of its content. The machine learning framework is evaluated with SemEval-2012 and SemEval-2013 data from the spatial role labeling task.

A survey of qualitative spatial representations

AbstractRepresentation and reasoning with qualitative spatial relations is an important problem in artificial intelligence and has wide applications in the fields of geographic information system, computer vision, autonomous robot navigation, natural language understanding, spatial databases and so on. The reasons for this interest in using qualitative spatial relations include cognitive comprehensibility, efficiency and computational facility. This paper summarizes progress in qualitative spatial representation by describing key calculi representing different types of spatial relationships. The paper concludes with a discussion of current research and glimpse of future work.

Image retrieval through qualitative representations over semantic features

This paper proposes a qualitative knowledge-driven semantic modelling approach for image understanding and retrieval. The similarity measure is calculated for each query by exploiting the notion of conceptual neighbourhood--a measure of closeness between qualitative relations. The relative similarity of two images is proportional to the qualitative similarity measure value. The approach is motivated by the need to bridge the semantic gap between a human user and that of CBIR systems and enable semantic querying in such systems. Local semantic concepts, such as sky, grass, of an image are used to obtain a semantic image description. Four kinds of qualitative spatial representations have been applied to these semantic concepts. This allows for representation and reasoning of an image's content structures at a more abstract level than pixels or other low level features and provides a higher level, semantic basis for image understanding. We investigate whether such a representation of an image's visual content also provides an effective and natural way to provide content-oriented querying. We also investigate whether querying based on multiple representations is effective, and report on three voting schemes used to retrieve images in this way. The test data set having hand-assigned class labels has been used to have a metric evaluation of the retrieval accuracy. The results compare favourably with a non-qualitative representation based on the same semantic features which simply compares the percentage of each feature in pairs of images.

Spatial reasoning with rectangular cardinal relations

Qualitative spatial representation and reasoning plays a important role in various spatial applications. In this paper we introduce a new formalism, we name RCD calculus, for qualitative spatial reasoning with cardinal direction relations between regions of the plane approximated by rectangles. We believe this calculus leads to an attractive balance between efficiency, simplicity and expressive power, which makes it adequate for spatial applications. We define a constraint algebra and we identify a convex tractable subalgebra allowing efficient reasoning with definite and imprecise knowledge about spatial configurations specified by qualitative constraint networks. For such tractable fragment, we propose several polynomial algorithms based on constraint satisfaction to solve the consistency and minimality problems. Some of them rely on a translation of qualitative networks of the RCD calculus to qualitative networks of the Interval or Rectangle Algebra, and back. We show that the consistency problem for convex networks can also be solved inside the RCD calculus, by applying a suitable adaptation of the path-consistency algorithm. However, path consistency can not be applied to obtain the minimal network, contrary to what happens in the convex fragment of the Rectangle Algebra. Finally, we partially analyze the complexity of the consistency problem when adding non-convex relations, showing that it becomes NP-complete in the cases considered. This analysis may contribute to find a maximal tractable subclass of the RCD calculus and of the Rectangle Algebra, which remains an open problem.

Ontology based Image Retrieval Framework Using Qualitative Semantic Image Descriptions

This research proposes an ontology based image retrieval framework from a corpus of natural scene images by imparting human cognition in the retrieval process. The proposed architecture addresses the issues of keyword based image retrieval and content-based image retrieval through the use of qualitative spatial representations over semantic image annotations. Domain ontology has been developed to model qualitative semantic image descriptions and retrieval, thereafter can be accomplished either using a natural language description of an image containing semantic concepts and spatial relations, or in a query by example fashion. A psychophysical evaluation has also been carried out to evaluate the effectiveness of our approach and results of different experiments are quite promising in terms of retrieval accuracy and relevance of retrieved images.