English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

GIScience 2016 Short Paper Proceedings Multi-Scale Extraction of Regular Activity Patterns in Spatio-Temporal Events Databases: A Study Using Geolocated Tweets from Central Mexico Pablo Lopez-Ramirez 1 and Oscar Sanchez-Siordia 1 Centro de Investigaci´on en Geograf´ia y Geom´atica Ing. Jorge L. Tamayo, CentroGeo Abstract This paper proposes a new technique for the extraction of regular activity patterns at different scales, mined from the micro-blogging platform Twitter. The approach is based on the recursive application of the DBSCAN clustering algorithm to the geolocated Twitter feed. This technique includes a novel way to obtain averaged regular activity zones based on the rasterization and aggregation of the Concave Hull of the clusters identified at each resolution level. Since the proposed technique uses only the spatio-temporal characteristics of the geolocated Twitter feed and it does not depend on the messages, it can be extended to work with different spatio-temporal event sources such as mobile telephone records. An experiment was carried out to demonstrate the effectiveness of the technique in the extraction of known activity patterns in the Mexico Central Region. Introduction The digital breadcrumbs left by social media users have proven to be a valuable source of geographic insights. In the GIS field, they have been used for the detection of events (Atefeh and Khreich) or the characterization of zones through social media activity (Frias-Martinez and Frias-Martinez; Lee et al.), among other things. In most cases, extraction and characterization of regular activity patterns is of great importance. In this paper we propose a technique for the extraction of such patterns that relies only on the spatio-temporal properties of the Twitter feed. The purpose of this is, on the one hand, to improve on the current available techniques (Lee et al.; Frias-Martinez and Frias-Martinez) and, on the other hand, to be as independent from the nature of the Twitter feed as possible. The proposed technique is based on the observation that human activity patterns exhibit a wide range of scales (Arcaute et al.), and that the current methods for determining this activity from Twitter messages do not consider this. The proposed approach is based on the recursive application of a clustering algorithm. This approach demands the development of a novel way for averaging the spatial patterns extracted from the data. Methodology The idea behind the technique proposed in this paper, is that activity patterns may exhibit a range of scales, and that this scales cannot be represented by a flat tessellation. To overcome this limitation, we propose the use of a recursive clustering strategy that is able to extract the structures present at different scales in the database. Recursive Clustering Several techniques produce a hierarchical representation of point samples, the focus being often on extracting the most significant clusters across all scales. In the case of our study, we need to find clusters that are representative at each resolution level. For this, we will recursively apply DBSCAN (Ester et al.) to the data and thus obtain a hierarchy of clusters across resolution levels. Initially eps 0 and MinPoints are selected using the k-dist plot (Ester et al.) for the whole sample in the corresponding time slice. Then for each iteration of the algorithm, the value for eps 0 is halved while MinPoints