Brian Regan

FUSING TEXT AND IMAGE FOR EVENT DETECTION IN TWITTER

In this contribution, we develop an accurate and effective event detection method to detect events from a Twitter stream, which uses visual and textual information to improve the performance of the mining process. The method monitors a Twitter stream to pick up tweets having texts and images and stores them into a database. This is followed by applying a mining algorithm to detect an event. The procedure starts with detecting events based on text only by using the feature of the bag-of-words which is calculated using the term frequency-inverse document frequency (TF-IDF) method. Then it detects the event based on image only by using visual features including histogram of oriented gradients (HOG) descriptors, grey-level cooccurrence matrix (GLCM), and color histogram. K nearest neighbours (Knn) classification is used in the detection. The final decision of the event detection is made based on the reliabilities of text only detection and image only detection. The experiment result showed that the proposed method achieved high accuracy of 0.94, comparing with 0.89 with texts only, and 0.86 with images only .

Securing an EHR in a health sector digital ecosystem

This paper examines the challenges in maintaining system security in the dynamic environment of eHealth where there are system components scattered among private and public organisations and used by a spectrum of professionals, each with different needs. The need for more interoperability between these systems has strained the ability of system administrators to maintain compliance with security protocols by staff working in these environments.

A multiple kernel learning based fusion for earthquake detection from multimedia twitter data

An efficient way of extracting useful information from multiple sources of data is to use data fusion technology. This paper introduces a data fusion approach in multimedia data for earthquake detection in twitter by using kernel fusion. The fusion method applies to fuse two types of data. The first type is features extracted from text by using bag-of-words method which is based on the calculation of the term frequency-inverse document frequency. The second type is the visual features extracted from images by applying scale-invariant feature transform. A multiple kernel fusion is applied in order to fuse the information from these two sources. Our experiments have indicated that comparing to the approaches using single data source, the proposed approach of using multiple kernel learning algorithm as early fusion increased the accuracy for earthquake detection. Experimental results for the proposed method achieved a high accuracy of 0.94, comparing to accuracy of 0.89 with texts only, and accuracy of 0.83 with images only.

RFID Authentication of Academic Transcripts: In the Context of a RFID Rationale and Deployment Methodology

The integrity and authentication of an academic transcript is vital to the prestige of aneducational organisation as well as its graduates. It is important to the organisation issuing the transcript andany future educational organisation that uses the transcript to ascertain the merit of aproposed enrolment or academic employment. The business case for this research isundertaken within the boundaries of a radio frequency identification rationale and deployment methodology.It is proposed to utilise radio frequency identification technology to enhance the current process of creating andvalidating academic transcripts. This paper will investigate the business case for radio frequency identification technologyand how this correlates with the business environment which is the first phase of the methodology.

Constructing memorable asymmetric information diagrams in three dimensions

This paper presents a model for presenting relational information as a graph in three dimensions, where the emphasis is on the location of nodes by recognition of their locality in the drawing rather than the tracing of paths through the graph. The approach exploits the potential of 3D to present graphs with a high number of nodes specific to an individual user, such as a graph of web-page bookmarks. The key element is the creation of a memorable structure that contains a variety of sub-structures which act as reference points for the viewer as they manipulate their own viewpoint within the three dimensional virtual space of the graph layout. The layout algorithm used here challenges the need for an aesthetic of symmetry in graph drawing within 3D and exploits the user’s ability to recall sub-structures within a drawing to replace edge-tracing as the means for locating nodes in the drawing.

RAISING COMMUNITY AWARENESS ABOUT THE ENVIRONMENT THROUGH DYNAMICALLY GENERATED STORIES

This paper discusses issues raised in the design of a dynamically generated Web-based natural history encyclopedia for children. The novelty of our proposal is the dynamic creation of encyclopedia entries using domain-specific middleware to access biological and ecological readers (users) with tailored reports summarizing information drawn from distributed datasets. A prerequisite, however, is a domain-dependent standardized data labeling system. Our project will piggyback on the host of projects currently developing metalabeling frameworks to dynamically create stories which are peopled by creatures from biological collections, are set in locations of local or other interest to the site visitor, and use scenarios based on relevant ecosystems. The underlying scientific databases provide a much broader range of subject matter (species and locations) that would not otherwise be feasible and alleviate content maintenance problems that beset community Web sites or professional encyclopedias. To incorporate such information in the sorts of format expected in a childrens encyclopedia involves developing sets of “story” templates, rules for selection of a template, and rules for design of the multimedia presentation of the story. It also requires comprehensive labeling systems for data in the scientific databases.