- ZeitschriftenartikelRemember to Correct the Bias When Using Deep Learning for Regression!(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Igel, Christian; Oehmcke, StefanWhen training deep learning models for least-squares regression, we cannot expect that the training error residuals of the final model, selected after a fixed training time or based on performance on a hold-out data set, sum to zero. This can introduce a systematic error that accumulates if we are interested in the total aggregated performance over many data points (e.g., the sum of the residuals on previously unseen data). We suggest adjusting the bias of the machine learning model after training as a default post-processing step, which efficiently solves the problem. The severeness of the error accumulation and the effectiveness of the bias correction are demonstrated in exemplary experiments.
- ZeitschriftenartikelGeoAI as Collaborative Effort(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Richter, Kai-Florian; Scheider, Simon; Tuia, Devis
- ZeitschriftenartikelGeoAI and Beyond(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Scheider, Simon; Richter, Kai-Florian; Janowicz, Krzysztof
- ZeitschriftenartikelPragmatic GeoAI: Geographic Information as Externalized Practice(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Scheider, Simon; Richter, Kai-FlorianCurrent artificial intelligence (AI) approaches to handle geographic information (GI) reveal a fatal blindness for the information practices of exactly those sciences whose methodological agendas are taken over with earth-shattering speed. At the same time, there is an apparent inability to remove the human from the loop, despite repeated efforts. Even though there is no question that deep learning has a large potential, for example, for automating classification methods in remote sensing or geocoding of text, current approaches to GeoAI frequently fail to deal with the pragmatic basis of spatial information, including the various practices of data generation, conceptualization and use according to some purpose. We argue that this failure is a direct consequence of a predominance of structuralist ideas about information. Structuralism is inherently blind for purposes of any spatial representation, and therefore fails to account for the intelligence required to deal with geographic information. A pragmatic turn in GeoAI is required to overcome this problem.
- ZeitschriftenartikelAutomated Interpretation of Place Descriptions: Determining Entity Types for Querying OSM(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Yousaf, Madiha; Schwartz, Tobias; Wolter, DiedrichThis paper is concerned with interpretation of natural language place descriptions, as they are a rich source of geographic information. A place description is interpreted by matching geographic entities occurring in the text against the OpenStreetMap (OSM) database. This paper is mainly concerned with interpretation of paraphrased places, i.e., entities for which no name is given and which may only by described. Our objective is to determine suitable entity types that allow querying the OpenStreetMap database for the respective place. For example, if we wish to identify a place to eat , we have to check for entities of an a-priori unknown type (cafe, restaurant, etc.). Challenges arise from the open-endedness of language, its ambiguity, and context-sensitivity as well as from mismatches between human conceptualization of place and database ontologies. The contributions of this paper are, first, to present a hard problem that is key to geo-information retrieval beyond named entities. Second, we propose context-sensitive methods for identifying place types based on semantic word similarity. We evaluate the methods on text extracted from Wikipedia and travel blogs, revealing their contribution to advancing automated interpretation of place descriptions to paraphrased places.
- ZeitschriftenartikelGeoAI(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Scheider, Simon; Richter, Kai-Florian
- ZeitschriftenartikelIdentifying Landscape Relevant Natural Language using Actively Crowdsourced Landscape Descriptions and Sentence-Transformers(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Baer, Manuel F.; Purves, Ross S.Natural language has proven to be a valuable source of data for various scientific inquiries including landscape perception and preference research. However, large high quality landscape relevant corpora are scare. We here propose and discuss a natural language processing workflow to identify landscape relevant documents in large collections of unstructured text. Using a small curated high quality collection of actively crowdsourced landscape descriptions we identify and extract similar documents from two different corpora ( Geograph and WikiHow ) using sentence-transformers and cosine similarity scores. We show that 1) sentence-transformers combined with cosine similarity calculations successfully identify similar documents in both Geograph and WikiHow effectively opening the door to the creation of new landscape specific corpora, 2) the proposed sentence-transformer approach outperforms traditional Term Frequency - Inverse Document Frequency based approaches and 3) the identified documents capture similar topics when compared to the original high quality collection. The presented workflow is transferable to various scientific disciplines in need of domain specific natural language corpora as underlying data.
- ZeitschriftenartikelIn Search of Basement Indicators from Street View Imagery Data: An Investigation of Data Sources and Analysis Strategies(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Vo, Anh Vu; Bertolotto, Michela; Ofterdinger, Ulrich; Laefer, Debra F.Street view imagery databases such as Google Street View, Mapillary, and Karta View provide great spatial and temporal coverage for many cities globally. Those data, when coupled with appropriate computer vision algorithms, can provide an effective means to analyse aspects of the urban environment at scale. As an effort to enhance current practices in urban flood risk assessment, this project investigates a potential use of street view imagery data to identify building features that indicate buildings’ vulnerability to flooding (e.g., basements and semi-basements). In particular, this paper discusses (1) building features indicating the presence of basement structures, (2) available imagery data sources capturing those features, and (3) computer vision algorithms capable of automatically detecting the features of interest. The paper also reviews existing methods for reconstructing geometry representations of the extracted features from images and potential approaches to account for data quality issues. Preliminary experiments were conducted, which confirmed the usability of the freely available Mapillary images for detecting basement railings as an example type of basement features, as well as geolocating the features.
- ZeitschriftenartikelURWalking: Indoor Navigation for Research and Daily Use(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) Ludwig, Bernd; Donabauer, Gregor; Ramsauer, Dominik; Subari, Karema alIn this report, we present the project URWalking conducted at the University of Regensburg. We describe its major outcomes: Firstly, an indoor navigation system for pedestrians as a web application and as an Android app with position tracking of users in indoor and outdoor environments. Our implementation showcases that a variant of the $$A^*$$ A ∗ -algorithm by Ullmann (tengetriebene optimierung präferenzadaptiver fußwegrouten durch gebäudekomplexe https://epub.uni-regensburg.de/43697/ , 2020) can handle the routing problem in large, levelled indoor environments efficiently. Secondly, the apps have been used in several studies for a deeper understanding of human wayfinding. We collected eye tracking and synchronized video data, think aloud protocols, and log data of users interacting with the apps. We applied state-of-the-art deep learning models for gaze tracking and automatic classification of landmarks. Our results indicate that even the most recent version of the YOLO image classifier by Redmon and Farhadi (olov3: An incremental improvement. arXiv, 2018) needs finetuning to recognize everyday objects in indoor environments. Furthermore, we provide empirical evidence that appropriate machine learning models are helpful to bridge behavioural data from users during wayfinding and conceptual models for the salience of objects and landmarks. However, simplistic models are insufficient to reasonably explain wayfinding behaviour in real time—an open issue in GeoAI. We conclude that the GeoAI community should collect more naturalistic log data of wayfinding activities in order to build efficient machine learning models capable of predicting user reactions to routing instructions and of explaining how humans integrate stimuli from the environment as essential information into routing instructions while solving wayfinding tasks. Such models form the basis for real-time wayfinding assistance.
- ZeitschriftenartikelNews(KI - Künstliche Intelligenz: Vol. 37, No. 1, 2023) null