ThinkSpatial: Armin Haller

thinkspatial_log<strong></strong>oThe UCSB forum on spatial technologies presents

What Are Links in Linked Open Data? A Characterization and Evaluation of Links between Knowledge Graphs on the Web

Armin Haller

Australian National University

3:00 p.m. (PT)

Thursday, April 8, 2021

The recorded video of this talk can be found here.

Slides can be found here.

 

Abstract. Linked Open Data promises to provide guiding principles to publish interlinked knowledge graphs on the Web in the form of findable, accessible, interoperable, and reusable datasets. I will argue that while as such, Linked Data may be viewed as a basis for instantiating the FAIR principles, there are still a number of open issues that cause significant data quality issues even when knowledge graphs are published as Linked Data. In this talk I will first define the boundaries of what constitutes a single coherent knowledge graph within Linked Data, i.e., present a principled notion of what a dataset is and what links within and between datasets are. I will also define different link types for data in Linked datasets and present the results of our empirical analysis of linkage among the datasets of the Linked Open Data cloud. Recent results from our analysis of Wikidata, which has not been part of the Linked Open Data Cloud, will also be presented.

Bio. Armin Haller is an Associate Professor of Business Information Systems at the Australian National University. His research interests include knowledge graph engineering, ontology engineering, linked data, the Internet-of-Things and the semantic Web in general. Haller has received funding from organisations including: the Department of Finance, to develop an ontology framework for government information and an open-source software that allows domain experts to maintain knowledge graphs called Schímatos; and from the Australian Research Council’s Linkage Program, for a project with Sydney Trains. He has published in the premier journals and conferences in his field, including Semantic Web Journal, Journal of Web Semantics, Applied Ontology, World Wide Web Conference and the International Semantic Web Conference. Haller has been an active contributor to several W3C working groups and chaired the Semantic Sensor Network Ontology working group. This work has resulted in an international standard for the description of sensors and actuators on the Internet-of-Things. Haller is a passionate evangelist of Open Government Data and, to this end, has chaired the Australian Government Linked Data Working Group for the last 7 years.

The objectives of the ThinkSpatial brown-bag presentations are to exchange ideas about spatial perspectives in research and teaching, to broaden communication and cooperation across disciplines among faculty and graduate students, and to encourage the sharing of tools and concepts.

Please contact Marcela Suarez (amsuarez@ucsb.edu), or Karen Doehner (kdoehner@spatial.ucsb.edu)), to review and schedule possible discussion topics or presentations that share your disciplinary interest in spatial thinking.

Follow spatial@ucsb on Twitter | Google+ | Google Calendar

NSF Convergence Accelerator Series Tracks A&B: Natasha Noy

NSF Convergence Accelerator Tracks A&B Speaker Series

 

Google Dataset Search: Building an Open Ecosystem for Dataset Discovery

Natasha Noy

Google Research

Thursday, March 18, 2021. 9:00 a.m. (PT)

The recorded video of this talk can be found here.

 

The National Science Foundation’s (NSF) tracks A and B of the Convergence Accelerator program are proud to present Natasha Noy in its 2021/2022 speaker series on Open Knowledge Networks. The series features researchers and practitioners widely recognized for their contribution to knowledge graphs, knowledge engineering, and FAIR data.

Abstract. There are thousands of data repositories on the Web, providing access to millions of datasets. National and regional governments, scientific publishers and consortia, commercial data providers, and others publish data for fields ranging from social science to life science to high-energy physics to climate science and more. Access to this data is critical to facilitating reproducibility of research results, enabling scientists to build on others’ work, and providing data journalists easier access to information and its provenance. In this talk, we will discuss recently launched Dataset Search by Google, which provides search capabilities over potentially all dataset repositories on the Web. I will talk about the open ecosystem for describing datasets that we hope to encourage.

Bio: Natasha Noy is a senior staff scientist at Google Research where she works on making structured data accessible and useful. She leads the team building Dataset Search, a search engine for all the datasets on the Web. Prior to joining Google, Noy worked at Stanford Center for Biomedical Informatics Research where she made major contributions in the areas of ontology development and alignment, and collaborative ontology engineering. Noy is a Fellow of the Association for the Advancement of Artificial Intelligence (AAAI). She served as president of the Semantic Web Science Association from 2011 to 2017.

Please contact us for follow-up questions.

Follow spatial@ucsb on Twitter | Google+ | Google Calendar

2021 Lightning Talks

The Center for Spatial Studies’ Lightning Talks are designed to be serious or funny, as long as the mantra is followed: “​Enlighten us, but make it quick​.

This year the talks were held on ​Tuesday, March 16, 2021. This annual series of ​3-minute​ lightning talks typically brings together speakers from across the UCSB campus and the local community to enlighten the crowd on thought-provoking ​spatial topics of all kinds​. Since this year the event was held online, it had much broader participation from across the globe with an exciting lineup of speakers from the spatial community worldwide. See the list of selected speakers and the titles of their talks below. Recorded videos can be found here.

Special note: ​The 2021 annual Center for Spatial Studies ​Lightning Talks​ were dedicated to the memory of Paul Wilson, who was one of the Center’s most avid and constant participants of all things spatial. Paul’s presence at our ongoing activities will be sorely missed. See his 2014 Lighting Talk on Whale Traffic Control: https://escholarship.org/uc/item/6t4627c5

Contact:
Rui Zhu ruizhu@geog.ucsb.edu

NSF Convergence Accelerator Series Tracks A&B: Denny Vrandečić

NSF Convergence Accelerator Tracks A&B Speaker Series

 

Knowledge beyond the Graph: Toward a Multilingual Wikipedia

 

Denny Vrandečić

Wikimedia Foundation

Thursday, Feb 11, 2021. 9:00 a.m. (PT)

The recorded video of this talk can be found here.

 

The National Science Foundation’s (NSF) tracks A and B of the Convergence Accelerator program are proud to present the next speaker in its 2021/22 speaker series on Open Knowledge Networks. The series will feature researchers and practitioners widely recognized for their contribution to knowledge graphs, knowledge engineering, and FAIR data

Abstract. Wikipedia’s vision is a world in which everyone can share in the sum of all knowledge. In its first two decades, this vision has been very unevenly achieved. One of the largest hindrances is the sheer number of languages Wikipedia needs to cover in order to achieve that goal. We argue that we need a new approach to tackle this problem more effectively, a multilingual Wikipedia where content can be shared between language editions.

We have started a new project where we separate this goal into two parts: creating and maintaining content in an abstract notation within a project called Abstract Wikipedia, and creating a new project called Wikifunctions that can translate this notation to natural language. Both parts are fully owned and maintained by the community. This architecture will make more encyclopedic content available to more people in their own language, and at the same time allow more people to contribute knowledge and reach more people with their contributions, no matter what their respective language backgrounds.

Bio: Denny Vrandečić has joined Wikimedia Foundation as Head of Special Projects in order to lead the Abstract Wikipedia project. He obtained his Ph.D. from Karlsruhe Institute of Technology (KIT) in 2012 where he co-founded Semantic MediaWiki, and thereafter launched Wikidata at Wikimedia Deutschland. He then joined Google as an ontologist on the Google Knowledge Graph and later worked as a researcher on the topic of knowledge representation.

Please contact us for follow-up questions.

Follow spatial@ucsb on Twitter | Google+ | Google Calendar

NSF Convergence Accelerator Series Tracks A&B: Markus Krötzsch

NSF Convergence Accelerator Tracks A&B Speaker Series

 

Knowledge Graphs for AI: Wikidata and Beyond

 

Markus Krötzsch

Technische Universität Dresden

Wednesday, Feb 3, 2021. 9:00 a.m. (PT)

The recorded video of this talk can be found here.

 

The National Science Foundation’s (NSF) tracks A and B of the Convergence Accelerator program are proud to present the first speaker in their 2021/22 speaker series on Open Knowledge Networks. The series will feature researchers and practitioners widely recognized for their contribution to knowledge graphs, knowledge engineering, and FAIR data

Abstract. Wikidata, the knowledge graph of Wikimedia, has successfully grown from an experimental “data wiki” to a well-organized reference knowledge base with a large and active editor community as well as many academic and industrial uses. It is also a key ingredient of popular AI applications, most prominently of intelligent agents such as Apple’s Siri or Amazon’s Alexa. Of course, human knowledge is fully expected to be in high demand in this time of rapidly advancing AI. And yet, the fact that modern AI relies on the manual labor of thousands of human knowledge modelers is in stark contrast to the common narrative of AI in popular media, which tells us that methods of pattern recognition and statistical function approximation can produce intelligent behavior from unstructured data without much human intervention. However, Wikidata is not a singular exception to the trend but rather a specific solution to a general need of AI: the need for knowledge that is understandable to humans and accessible to computers. Almost every major AI application incorporates such knowledge, and organizations long have realized the need to acquire and develop knowledge resources as part of their AI strategy. The next frontier in AI is the ability of systems to explain and justify their behavior. There, too, we can see the need for knowledge-based technologies as a bridge between human understanding and computational mechanisms, but the task goes far beyond the realms of knowledge representation or machine learning, and will require the effort of all of AI and maybe all of computer science. In my talk, I will give an overview of Wikidata and outline some ongoing research efforts that combine knowledge representation with other methods towards the creation of (more) understandable and accountable AI.

Bio: Markus Krötzsch is a full professor at the Faculty of Computer Science of TU Dresden, where he is holding the chair for Knowledge-Based Systems. He obtained his Ph.D. from Karlsruhe Institute of Technology (KIT) in 2010, and thereafter worked at the Department of Computer Science of the University of Oxford until October 2013. He has contributed to the concept and design of Wikidata, as one of the most prominent examples of applied knowledge representation today. His research made many further contributions to the development and analysis of knowledge modeling languages (including the W3C OWL standard), inference methods, and automated reasoners. Krötzsch is a member of the Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI) and of the Center for Perspicuous Computing (CPEC).

Please contact us for follow-up questions.

Follow spatial@ucsb on Twitter | Google+ | Google Calendar

Spatial Data Science Hangout Series: Reminder

Our Spatial Data Hangout will take place tomorrow (Thursday), December 3 from 10:00 to 11:00 am (Pacific Time). Please join us for this special session focused on Graph Data and Networks, and led by Rui Zhu, Su Burtner, Gengchen Mai, and Mike Johnson.

Zoom Meeting Link: https://ucsb.zoom.us/j/88693830517?pwd=emUyejFWa0s5K0EyUXNaZVI4bnBEdz09

Meeting ID: 886 9383 0517
Passcode: geohangout

We look forward to seeing you!

Spatial Data Science Hangout Series: Call for Speakers

The Center for Spatial Studies will again be hosting a special Spatial Data Hangout on Thursday, December 3 at 10:00 a.m. All grad students are invited to attend! We are fully aware that your schedule is already crowded with  Zoom meetings, but we are hopeful that this will be  an excuse for you to hang out with other geography grads and  learn a bit about the cool projects or ideas they have been working on this year.

This spatial data science hangout will focus on learning and discussing all sorts of graph data (and analytics that involve graphs) used to do research in our department. This includes everything related to representing and reasoning on data. This is  an opportunity for you to: (i) teach others about the tools you use to represent data for spatial reasoning, –this might include some coding, (ii) discuss common graphs for spatial analytics in your research field, (iii) discuss the process you use to create a graph that you are proud of, or (iv) present early-stage ideas, projects that you might not complete yet or learned lessons from “failed” projects. This is meant to be a safe space for people to talk through ideas, and learn from each other.  Some potential topics include spatial networks in their broadest sense, such as: 

  • Social networks
  • Transportation networks 
  • Biological networks
  • Other types of networks in your domain (stream networks)
  • Knowledge graphs

We are looking for folks who would like to lead/co-lead sessions! Depending on the number of speakers, you might be able to participate with a lightning talk or a 15-minute talk. If you are interested in participating in this, please contact Marcela Suárez at amsuarez@ucsb.edu by November 27 (preferably sooner), and otherwise mark your calendar to attend.

ThinkSpatial: Gengchen Mai

thinkspatial_log<strong></strong>oThe UCSB forum on spatial technologies presents

Space2Vec: Multi-Scale Representation Learning for Spatial Feature Distributions using Grid Cells

Gengchen Mai

STKO Lab

University of California, Santa Barbara

11:30 a.m. (PT)

Tuesday, Dec 1, 2020

Zoom:  https://ucsb.zoom.us/meeting/register/tZUtcu6rrzosE9VEnfMGptAxbxe4zhNILoHY

Upon registration, you will receive access to this link

Abstract. Unsupervised text encoding models have recently fueled substantial progress in NLP. The key idea is to use neural networks to convert words in texts to vector space representations (embeddings) based on word positions in a sentence and their contexts, which are suitable for end-to-end training of downstream tasks. We see a strikingly similar situation in spatial analysis, which focuses on incorporating both absolute positions and spatial contexts of geographic objects such as POIs into models. A general-purpose representation model for space is valuable for a multitude of tasks. However, no such general model exists to date beyond simply applying discretization or feed-forward nets to coordinates, and little effort has been put into jointly modeling distributions with vastly different characteristics, which commonly emerges from GIS data. Meanwhile, Nobel Prize-winning Neuroscience research shows that grid cells in mammals provide a multi-scale periodic representation that functions as a metric for location encoding and is critical for recognizing places and for path-integration. Therefore, we propose a representation learning model called Space2Vec to encode the absolute positions and spatial relationships of places. We conduct experiments on two real-world geographic data for two different tasks: (1) predicting types of POIs given their positions and context, (2) image classification leveraging their geo-locations. Results show that because of its multi-scale representations, Space2Vec outperforms well-established ML approaches such as RBF kernels, multi-layer feed-forward nets, and tile embedding approaches for location modeling and image classification tasks. Detailed analysis shows that all baselines can at most well handle distribution at one scale but show poor performances in other scales. In contrast, Space2Vec ’s multi-scale representation can handle distributions at different scales.

 

Bio. Gengchen Mai is a Ph.D. candidate at the Space and Time for Knowledge Organization Lab in the Department of Geography, University of California, Santa Barbara. His Ph.D. adviser is Prof. Krzysztof Janowicz. His research interests include Machine Learning/Deep Learning, GIScience, Geographic Question Answering, NLP, Geographic Information Retrieval, Knowledge Graph, and Semantic Web. Currently, Mai’s research is highly focused on Geographic Question Answering and Spatially-Explicit Machine Learning models. He received his B.S. Degree in Geographic Information System from Wuhan University. Thus far, he has completed four AI/ML research-based internships at Esri Inc., SayMosaic Inc., Apple Map, and Google X. He now serves as a machine learning consultant/advisor for Google X.

The objectives of the ThinkSpatial brown-bag presentations are to exchange ideas about spatial perspectives in research and teaching, to broaden communication and cooperation across disciplines among faculty and graduate students, and to encourage the sharing of tools and concepts.

Please contact Marcela Suarez (amsuarez@ucsb.edu), or Karen Doehner (kdoehner@spatial.ucsb.edu)), to review and schedule possible discussion topics or presentations that share your disciplinary interest in spatial thinking.

Follow spatial@ucsb on Twitter | Google+ | Google Calendar

ThinkSpatial: Patricia Murrieta-Flores

thinkspatial_logo
On Tuesday, November 24, 2020, the UCSB forum on spatial thinking presents

Subaltern Spatial Thinking: Towards a decolonial approach to spatial technologies

 

Patricia Murrieta-Flores

Lancaster University

11:30 a.m. Tuesday, November 24, 2020 | Zoom*

Abstract:

The quick emergence of digital technologies and the rise of the use of computational approaches and tools in all disciplines have substantially changed the ways that we do research, expediting and placing at our fingertips datasets and information at a scale that was impossible before. While our societies have enthusiastically embraced this change, technologies are neither unbiased nor innocent. In the wake of Geographic Information Systems (GIS), a now-ubiquitous tool, disciplines such as history and archaeology have adopted this technology for the study of the past. This is well represented in the creation of the field called the Spatial Humanities. Rooted in a modern, western and Cartesian conception of space, while GIS has proved invaluable to advance both research and discussions related to geography, space and place in these fields, there is the need for critical reflection regarding its use, and to work towards the development of more inclusive spatial tools. Given GIS’ increasing popularity in Humanities research worldwide, and especially its emergence in the Global South, I will present a case study from research produced in the project “Digging into Early Colonial Mexico: A large-scale computational analysis of 16th century historical sources”, aiming to showcase how through the analysis of Mesoamerican spatial thinking, we can highlight the need of carefully considering the use of particular technologies in historical research, and discuss a decolonial approach to technology.

Bio:

Patricia Murrieta-Flores is Senior Lecturer and Co-Director of Digital Humanities at Lancaster University. Her interest lies in the application of technologies for Humanities and her primary research area is the Spatial Humanities. Her main focus is the investigation of different aspects of space, place and time using a range of technologies including GIS, NLP, Machine Learning and Corpus Linguistics approaches. Patricia is the PI on the Transatlantic Platform (T-AP) funded project ‘Digging into Early Colonial Mexico: A large-scale computational analysis of 16th century historical sources’, and also collaborator and Co-I in multiple projects funded by the ERC, ESRC, AHRC, HERA, and the Paul Mellon Centre among others. She has edited and contributed to multiple books on Digital Humanities, Cultural Heritage, the use of GIS and other technologies in Archaeology, History, and Literature, and she has published multiple articles exploring theories and methodologies related to space and place. She is currently Executive Board Secretary Elect of the Alliance of Digital Humanities Organizations (ADHO).

The objectives of the ThinkSpatial brown-bag presentations are to exchange ideas about spatial perspectives in research and teaching, to broaden communication and cooperation across disciplines among faculty and graduate students, and to encourage the sharing of tools and concepts.

Please contact Marcela Suarez (amsuarez@ucsb.edu), Karen Doehner (kdoehner@spatial.ucsb.edu), or Emmanuel Papadakis (epd@ucsb.edu), to review and schedule possible discussion topics or presentations that share your disciplinary interest in spatial thinking.

*This talk is a part of the spatial series Knowledge Representation and GeoHumanities; upon registration, you can access all the talks of the series using the provided link.

Follow spatial@ucsb on Twitter | Google+ | Google Calendar

SpatialTech: Bruno Martins

On Tuesday, November 17, 2020, The UCSB forum on spatial technology presents

Challenges in resolving place names over text

 

Bruno Martins

University of Lisbon

11:30 a.m. Tuesday, November 17, 2020 | Zoom*

Abstract:

Toponym resolution concerns the disambiguation of place names in textual documents, envisioning the support for applications such as geographical search or the mapping of textually encoded information. Place names are first recognized through a named entity recognition model, and the disambiguation is then achieved by associating each of the place references to a unique position on the Earth’s surface, e.g., through the assignment of geospatial coordinates. The toponym resolution task is particularly challenging, given that place references are highly ambiguous (i.e., distinct locations can share the same place name, and multiple names can be used to refer to the same place). In this talk, I will discuss techniques for toponym resolution, with a particular emphasis on a novel deep learning approach. Contrarily to most previous methods, the novel approach does not involve matching references in the text against entries in a gazetteer, instead directly predicting geospatial coordinates. In brief, the neural network architecture considers multiple inputs (e.g.,the toponym to disambiguate together with the surrounding words), leveraging pre-trained contextual word embeddings for modeling the textual data. The intermediate representations are then used to predict a probability distribution over possible geospatial regions, and finally to predict the coordinates for the input toponym. I will present evaluation results over different types of corpora (e.g., modern newswire text or historical documents), and I will discuss the impact of model extensions related to (i) the use of external information concerning geophysical terrain properties, including information on terrain development or elevation, among others, and (ii) additional training data collected from Wikipedia articles, to guide and further help with model training.

Bio:

Bruno Martins is an assistant professor at the Computer Science and Engineering Department of Instituto Superior Técnico of the University of Lisbon (IST/UL), and a researcher at the Information and Decision Support Systems Lab of INESC-ID, where he works on problems related to the general areas of information retrieval, text mining, and the geographical information sciences. He received his MSc and PhD degrees from the Faculty of Sciences of the University of Lisbon, both in Computer Science. Bruno has been involved in several research projects related to geospatial aspects in information access and retrieval, and he has accumulated significant expertise in addressing challenges at the intersection of language technologies, machine learning, and the geographical information sciences. He and his students have worked on many different application areas, and he is proudest of the many PhD/MSc students who have graduated under his supervision and are now building wonderful careers.

Material:
SpatialTech-Bruno-Martins

The objectives of the Spatial Technology presentations are to exchange ideas and promote discussion and interaction within the spatial technology community

Please contact Karen Doehner, kdoehner@spatial.ucsb.edu or Emmanuel Papadakis epd@ucsb.edu, to review and schedule possible discussion topics or presentations that share your disciplinary interest in spatial technologies.

*This talk is a part of the spatial series Knowledge Representation and GeoHumanities; upon registration, you can access all the talks of the series using the provided link.

Follow spatial@ucsb on Twitter | Google+ | Google Calendar