Waikato University crest

Department of
Computer Science
Tari Rorohiko

Computing and Mathematical Sciences

Upcoming Seminars

 

Towards Linked Data Enrichment using Open Information Extraction

Dr Amal Zouaq
Royal Military College of Canada, Ontario, Canada
Tuesday 21 October 2014
11:00 am
G.1.15

The Semantic Web and Linked Data movements aim at converting the Web unstructured and semi-structured content into a "web of data". The purpose is to help users (both human and software agents) to find, share, and combine information more easily and ultimately to enable large scale machine reasoning, using knowledge bases accessible on the Web. The last decade has witnessed the development of several knowledge bases such as DBPedia, Yago and Freebase. These knowledge bases are generally extracted from Web semi-structured content such as Wikipedia Infoboxes. By contrast, our work is mostly focused on the extraction of knowledge from unstructured textual content.

This talk will explore how open information extraction (a domain-independent and unsupervised knowledge extraction paradigm) can contribute to a) the extraction of structured content (entities, topics, relations) from free-text domain corpora and b) the enrichment of current Linked Data knowledge bases. In particular, we will focus on open relation extraction, which aims at linking (Web) entities with meaningful relations. Experiments and challenges around the quality of the extracted entities and relations and the measurement of such quality will be briefly discussed.

 

The Kukui Cup and Open Power Quality: creating smart consumers for the smart grid

Philip Johnson
University of Hawaii at Manoa, Honolulu
Tuesday 11 November 2014
11:00 am
G.1.15
The environmental, political, and economic problems with fossil fuels as an energy source have encouraged work on a next generation “smart” grid: an electrical infrastructure which can incorporate distributed, renewable energy generation sources such as solar and wind power. However, the smart grid cannot achieve its full potential without “smart” consumers: people who are literate about energy issues and who modify their behavior to support more efficient and effective energy generation, storage, and use. In this talk I describe two research initiatives to create “smart” consumers: the Kukui Cup and Open Power Quality.

The Kukui Cup Project (http://kukuicup.org) combines techniques from community-based social marketing, serious games, and educational pedagogy to investigate sustained change in energy-related behaviors. Kukui Cup challenges blend real world and online activities, all tied together through game mechanics. In the real world, players participate in workshops, excursions, and creative events. They compete to win prizes, and in the process, learn about their behavior and its impact on energy consumption. The online game environment allows players to earn points, achieve badges, increase their energy “literacy” through readings and videos, and use social networking mechanisms. I will discuss lessons learned from seven Kukui Cup challenges, and the open source software technologies created from this project.

Open Power Quality (http://openpowerquality.org) develops open source hardware, software, and data for low cost, crowd-sourced power quality monitoring, storage, and analysis. OPQ fills an important “data gap” in many current grids: the quality of power as experienced by end users. OPQ accomplishes this through custom-designed, low cost “OPQBoxes” that plug in to household electrical outlets, monitor power quality, and upload power quality “events” via WiFi to our OPQHub service. By crowd-sourcing the collection and analysis of power quality data, we hope to better understand the impact of distributed renewables on grid stability, distinguish grid-level problems from building-level problems, and increase consumer awareness of smart grid issues. I will discuss the current status of Open Power Quality and results from our recent pilot study with our first generation hardware and software.

Bio:
Philip Johnson is a Professor and Associate Chair in the Department of Information and Computer Sciences at the University of Hawaii. He received B.S. degrees in both Biology and Computer Science from the University of Michigan in 1980, and a Ph.D. in Computer Science from the University of Massachusetts in 1990. He is Director of the Collaborative Software Development Laboratory, which pursues research in software engineering, the smart grid, gamification, educational technologies, human-computer interaction, and computer supported cooperative work. Johnson is active in the Hawaii technology community, has co-founded two software startups, and has served on the Board of Directors of several technology companies. More details are available at: http://philipmjohnson.org.

 

Events Index