Mining tourist information from user-supplied collections

Published 2 November 2009

Adrian Popescu, Gregory Grefenstette, Pierre-Alain Moëllic

Citations78

TL;DR

This work shows how to extract clean trip related information (what people visit, for how long, panoramic spots) from Flickr metadata.

Abstract

Tourist photographs constitute a large part of the images uploaded to photo sharing platforms. But filtering methods are needed before one can extract useful knowledge from noisy user-supplied metadata. Here we show how to extract clean trip related information (what people visit, for how long, panoramic spots) from Flickr metadata. We illustrate our technique on a sample of metadata and images covering 183 cities of different size and from different parts of the world.

Keywords

Computer Science

Mining interesting locations and travel sequences from GPS trajectories

1,915 Citations2009Yu Zheng, Lizhu Zhang +2 more

This work first model multiple individuals' location histories with a tree-based hierarchical graph (TBHG), and proposes a HITS (Hypertext Induced Topic Search)-based inference model, which regards an individual's access on a location as a directed link from the user to that location.

Trajectory pattern mining

1,009 Citations2007Fosca Giannotti, Mirco Nanni +2 more

This paper develops an extension of the sequential pattern mining paradigm that analyzes the trajectories of moving objects and introduces trajectory patterns as concise descriptions of frequent behaviours in terms of both space and time.

Mapping the world's photos

777 Citations2009David Crandall, Lars Bäckström +2 more

This work uses the spatial distribution of where people take photos to define a relational structure between the photos that are taken at popular places, and finds that visual and temporal features improve the ability to estimate the location of a photo, compared to using just textual features.

Towards automatic extraction of event and place semantics from flickr tags

475 Citations2007Tye Rattenbury, Nathaniel Good +1 more

An approach for extracting semantics of tags, unstructured text-labels assigned to resources on the Web, based on each tag's usage patterns, and shows that the Scale-structure Identification method outperforms the existing techniques.

Generating diverse and representative image search results for landmarks

390 Citations2008Lyndon S. Kennedy, Mor Naaman

This work uses a combination of context- and content-based tools to generate representative sets of images for location-driven features and landmarks, a common search task.

Mining, indexing, and querying historical spatiotemporal data

287 Citations2004Nikos Mamoulis, Huiping Cao +4 more

This work defines the spatiotemporal periodic pattern mining problem and proposes an effective and fast mining algorithm for retrieving maximal periodic patterns, and devise a novel, specialized index structure that can benefit from the discovered patterns to support more efficient execution of spatiotsemporal queries.

World-scale mining of objects and events from community photo collections

260 Citations2008Till Quack, Bastian Leibe +1 more

This paper describes an approach for mining images of objects from community photo collections in an unsupervised fashion, and demonstrates this approach on several urban areas, densely covering an area of over 700 square kilometers and mining over 200,000 photos, making it probably the largest experiment of its kind to date.

World explorer

247 Citations2007Shane Ahern, Mor Naaman +2 more

This work analyzes the tags associated with the geo-referenced Flickr images to generate aggregate knowledge in the form of "representative tags" for arbitrary areas in the world, and uses these tags to create a visualization tool, World Explorer, that can help expose the content of the data, using a map interface to display the derived tags and the original photo items.

Deducing trip related information from flickr

71 Citations2009Adrian Popescu, Gregory Grefenstette

This work deduces answers to questions like "how long does it take to visit a tourist attraction?" or "what can I visit in one day in this city?" by comparing the automatically obtained visit duration times to manual estimations.

Image and Vision ComputingImage classification using color, texture and regions

70 Citations2003Ya-Chun Cheng, Shu‐Yuan Chen

A new classification method using color, texture and regions to prune irrelevant database images for each query image and the k-NN rule is used to assign a class label to the query image to achieve effective classification results.