You are here

ISCRAM 2013 – Best Papers

Best Paper Award

Muhammed Imran, Shady Elbassuoni, Carlos Castillo, Fernando  Diaz and Patrick Meier
Extracting Information Nuggets from Disaster-Related Messages in Social Media

The paper is about social media, which play an important role during a crisis. Many people post a huge number of tweets via Twitter, some of them are interesting to get a full awareness of what is going on, but most of the tweets are about the same information or about no information. The authors designed and implemented an automatic system to extract key information from a huge set of tweets which is very important for first responders and decision makers to get real time, up to date information about the ongoing crisis.

As a first step the authors annotated a huge dataset of tweets about a hurricane disaster using many annotators in a crowd sourcing design. Next they extracted linguistic features from the tweet text and used a classifier to find the classes automatically given the set of extracted features.

Strong points of the paper are: 1. the design is very complex 2. crowd sourcing requires a lot of managing power 3. the extraction of linguistic features using natural language processing technology is impressing. It is ongoing work and it can be expected that this paper is a starting point of many other papers on this topic.

Best Student Paper Award

Satria Hutomo Jihan and Aviv Segev
Context Ontology for Humanitarian Assistance in Crisis Response

The paper was presented at the humanitarian challenges track of ISCRAM 2013 and concerned a method that merges ontologies and logic rules to represent the humanitarian needs and recommend appropriate humanitarian responses. The work addresses a critical issue that humanitarian response practitioners have to deal with on a regular basis. Furthermore the ideas in the paper are timely as information explosion in the new digital age is affecting information management during crisis.

More precisely, the paper proposed a method for the automatic extraction of ontological information from unstructured crisis data and demonstrated by way of a case study how this contextual information can be mapped to pre-existing crisis response contexts. Both the reviewers and the best student paper evaluation committee felt that the work represents a notable contribution to the field and that it demonstrates the possibility of more 'intelligent' information systems assisting crisis responders.

The Best Paper Committeee

Tim, Zeno, Ionassis and Leon

Institutional Members



Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer