Logo des Repositoriums

Auflistung nach Autor:in "Baer, Manuel F."

1 - 3 von 3
  • Zeitschriftenartikel
    Augmented future: tracing the trajectory of location-based augmented reality gaming for the next ten years
    (i-com: Vol. 23, No. 2, 2024) Laato, Samuli; Söbke, Heinrich; Baer, Manuel F.
    Location-based games are a highly technology-dependent game genre that has witnessed an exponential increase in popularity with the democratisation of smartphones as well as ubiquitous mobile data and access to satellite navigation. Moving forward into the future, location-based games can be expected to evolve as the technologies underlying the genre improve. In this conceptual work, we review the current state of the art in location-based games, and identify key trajectories and trends. We discovered 12 trends, based on which we jump ten years into the future and evaluate how current technology trends may end up influencing location-based gaming. For example, we propose that in the year 2035 through improvements in map data services and sensor data coverage, we will see locative games that are increasingly connected to elements in the physical world. We also expect to see gameplay that moves away from solely taking place on a smartphone screen to the adoption of multiple forms of interactions with location-based game worlds, especially as head-mounted displays and other wearables become more commonplace.
  • Zeitschriftenartikel
    Identifying 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.
  • Zeitschriftenartikel
    Identifying 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.