Student research opportunities

Making Sense of Visual Folksonomy: a Flickr Study

Project Code: CECS_649

This project is available at the following levels:
Honours, Summer Scholar, Masters, PhD

Keywords:

image, flickr, semantics, taxonomy, folksonomy, artificial intelligence, computer vision, multimedia

Supervisor:

Dr Lexing Xie

Outline:

A taxonomy is a way of classifying things; a folksonomy is the spontaneous cooperation of a group of people to organize information into categories, i.e. a user-generated taxonomy, such as tags on Delicious, Youtube and Flickr.
A taxonomy as inherent relationships that can be used to organize content and search, a folksonomy are created by millions of users that will automatically scale. How do we find synergies between the two -- Can we use folksonomy to derive relationships that are otherwise unknown? Can we create a loosely organized visual taxonomy in the order of millions of nodes?

Goals of this project

Flickr is a living lab of millions of user-created photos, this provides an open platform for research experimentation. Steps in this project can be as follows:
+ Collect large amounts of tag data from online sources such as Flickr.
+ Derive relationships from online folksonomy.
+ Analyze which relationships in the folksonomy comforms and which ones violates common taxonomy.
+ Design algorithms that uses folksonomy-derived taxonomy for search and content organization.

Requirements/Prerequisites

+ Basic understanding of digital images. A course in computer vision or image processing a plus.
+ Linear algebra and probability. Knowledge in statistical machine learning or data mining a plus.
+ Knowledge in taxonomy and semantic web a plus.
+ Familiar with one or more programming/scripting language.

Student Gain

Practical:
- Hands-on knowledge in web2.0 data and programming protocols such as Flickr API.
- Hands-on knowledge in image processing, computer vision, and data mining algorithms.
- Hands-on knowledge in large-scale processing of web2.0 data.

Theoretical/research:
- Gain a deep knowledge in semantic structures, and existing taxonomies in english and images.
- Knowledge in state-of-the-art semantic ontology and visual taxonomy research.
- Publish in top international conferences and contribute to the research community.

Background Literature

Lexing Xie, Apostol Natsev, Matthew L. Hill, John R. Smith, Alex Phillips: The accuracy and value of machine-generated image tags: design and user evaluation of an end-to-end image tagging system. CIVR 2010: 58-65

Lexing Xie, Rong Yan, Jelena Tesic, Apostol Natsev, John R. Smith: Probabilistic visual concept trees. ACM Multimedia 2010: 867

J. Deng, K. Li, M. Do, H. Su, L. Fei-Fei, Construction and Analysis of a Large Scale Image Ontology. In Vision Sciences Society (VSS), 2009.

Cameron Marlow, Mor Naaman, Danah Boyd, and Marc Davis. 2006. HT06, tagging paper, taxonomy, Flickr, academic article, to read. In Proceedings of the seventeenth conference on Hypertext and hypermedia (HYPERTEXT '06). ACM, New York, NY, USA, 31-40.

Harris Wu, Mohammad Zubair, and Kurt Maly. 2006. Harvesting social knowledge from folksonomies. In Proceedings of the seventeenth conference on Hypertext and hypermedia (HYPERTEXT '06). ACM, New York, NY, USA, 111-114.

Thomas Gruber, Ontology of Folksonomy: A Mash-up of Apples and Oranges, International Journal on Semantic Web & Information Systems, Vol. 3, No. 2. (2007), pp. 1-11.