Combining Design Patterns and Topic Modeling to Discover Regions That Support Particular Functionality

Abstract

The problem of discovering regions that support particular functionalities in an urban setting has been approached in literature using two general methodologies: top-down, encoding expert knowledge on urban planning and design and discovering regions that conform to that knowledge; and bottom-up, using data to train machine learning models, which can discover similar regions. Both methodologies face limitations, with knowledge-based approaches being criticized for scalability and transferability issues and data-driven approaches for lacking interpretability and depending heavily on data quality. To mitigate these disadvantages, we propose a novel framework that fuses a knowledge-based approach using design patterns and a data-driven approach using latent Dirichlet allocation (LDA) topic modeling in three different ways: Functional regions discovered using either approach are evaluated against each other to identify cases of significant agreement or disagreement; knowledge from patterns is used to adjust topic probabilities in the learning model; and topic probabilities are used to adjust pattern-based results. The proposed methodologies are demonstrated through the use case of identifying shopping-related regions in the Los Angeles metropolitan area. Results show that the combination of pattern-based discovery and topic modeling extraction helps uncover discrepancies between the two approaches and smooth inaccuracies caused by the limitations of each approach.

Driving Environment Inference from POI of Navigation Map: Fuzzy Logic and Machine Learning Approaches

“…Theoretically, one can handcraft a reference POI occurrence pattern using the expert knowledge derived from widely acceptable data sources, such as dictionaries, encyclopedias, and the design and planning standards of a local government [29]. Alternatively, one can also experimentally derive a reference POI occurrence pattern from a set of training samples [49]. However, the first method may face the following challenges in our application:…”

A comparative assessment of machine learning methods in extracting place functionality from textual content

“…Therefore, functional characteristics of places in terms of activities that a person may perform in a place or place affordance can complement the structural elements in the formalization of place and are considered a primary dimension to determine place characteristics (Alazzawi et al, 2012). To address the challenge of discovering functional areas, place‐based research has followed mainly two independent approaches (Papadakis et al, 2019): (1) top‐down; and (2) bottom‐up.…”

“…Conceptualization of place differs between objective (e.g., location) and subjective (e.g., attachment) aspects of place, as well as between individual (e.g., sense of place) and social (e.g., functional differentiation) meaning of place (Hamzei et al, 2020). In this case, GIS should link functionality and space using data and knowledge about human activity and experience to answer questions such as “What can I do here?” or “How can I find places that provide a particular functionality?” In simpler terms, GIS should detect the place functionality (Papadakis et al, 2019) that does not necessarily exist simply and clearly in the stored data. On the other hand, due to the dramatic increase in the use of social networks, many users share textual content.…”

“…or "How can I find places that provide a particular functionality?" In simpler terms, GIS should detect the place functionality (Papadakis et al, 2019) that does not necessarily exist simply and clearly in the stored data. On the other hand, due to the dramatic increase in the use of social networks, many users the various predicted functionalities, the algorithm's efficiency is higher in Food Places.…”

Emotive facets of place meet urban analytics

“…This is yielding a pool of questions, models, and methods to study place-based concepts, which enriches the literature, but also complicates unified data collection and conceptualizations, essential for modeling them computationally (Hamzei, Li, et al, 2020). Within (geographic) information science, scholars focused on place-based research are mainly investigating exploratory, prototypical methods of analysis (Jenkins et al, 2016;Papadakis et al, 2019;Westerholt et al, 2018b), methods to visualize places (Westerholt, Resch, et al, 2018), theoretical differences between place and space (Goodchild & Li, 2011), and development of place-based ontologies (Ballatore, 2016;Purves et al, 2019;Scheider & Janowicz, 2014). The mentioned works are only a sample of the recent broad attention for the notion of "place" in GIScience.…”