Computing and Information Systems - Research Publications
Permanent URI for this collection
Search Results
1 - 10 of 11
-
ItemEfficient identity-based signatures in the standard model(INST ENGINEERING TECHNOLOGY-IET, 2008-12)
-
ItemBuilding more robust multi-agent systems using a log-based approach(IOS Press, 2009-03-23)
-
ItemDiscovering correlated spatio-temporal changes in evolving graphs(SPRINGER LONDON LTD, 2008-07)
-
Item
-
ItemOn the use of automatically acquired examples for all-nouns Word Sense Disambiguation(AI ACCESS FOUNDATION, 2008)This article focuses on Word Sense Disambiguation (WSD), which is a Natural Language Processing task that is thought to be important for many Language Technology applications, such as Information Retrieval, Information Extraction, or Machine Translation. One of the main issues preventing the deployment of WSD technology is the lack of training examples for Machine Learning systems, also known as the Knowledge Acquisition Bottleneck. A method which has been shown to work for small samples of words is the automatic acquisition of examples. We have previously shown that one of the most promising example acquisition methods scales up and produces a freely available database of 150 million examples from Web snippets for all polysemous nouns in WordNet. This paper focuses on the issues that arise when using those examples, all alone or in addition to manually tagged examples, to train a supervised WSD system for all nouns. The extensive evaluation on both lexical-sample and all-words Senseval benchmarks shows that we are able to improve over commonly used baselines and to achieve top-rank performance. The good use of the prior distributions from the senses proved to be a crucial factor.
-
Item
-
Item
-
ItemExploration of Networks Using Overview plus Detail with Constraint-based Cooperative Layout(IEEE COMPUTER SOC, 2008)A standard approach to large network visualization is to provide an overview of the network and a detailed view of a small component of the graph centred around a focal node. The user explores the network by changing the focal node in the detailed view or by changing the level of detail of a node or cluster. For scalability, fast force-based layout algorithms are used for the overview and the detailed view. However, using the same layout algorithm in both views is problematic since layout for the detailed view has different requirements to that in the overview. Here we present a model in which constrained graph layout algorithms are used for layout in the detailed view. This means the detailed view has high-quality layout including sophisticated edge routing and is customisable by the user who can add placement constraints on the layout. Scalability is still ensured since the slower layout techniques are only applied to the small subgraph shown in the detailed view. The main technical innovations are techniques to ensure that the overview and detailed view remain synchronized, and modifying constrained graph layout algorithms to support smooth, stable layout. The key innovation supporting stability are new dynamic graph layout algorithms that preserve the topology or structure of the network when the user changes the focus node or the level of detail by in situ semantic zooming. We have built a prototype tool and demonstrate its use in two application domains, UML class diagrams and biological networks.
-
ItemAutomatically Determining the Number of Clusters in Unlabeled Data Sets(Institute of Electrical and Electronics Engineers, 2009-03-01)One of the major problems in cluster analysis is the determination of the number of clusters in unlabeled data, which is a basic input for most clustering algorithms. In this paper, we investigate a new method called Dark Block Extraction (DBE) for automatically estimating the number of clusters in unlabeled data sets, which is based on an existing algorithm for Visual Assessment of Cluster Tendency (VAT) of a data set, using several common image and signal processing techniques. Its basic steps include 1) generating a VAT image of an input dissimilarity matrix, 2) performing image segmentation on the VAT image to obtain a binary image, followed by directional morphological filtering, 3) applying a distance transform to the filtered binary image and projecting the pixel values onto the main diagonal axis of the image to form a projection signal, and 4) smoothing the projection signal, computing its first-order derivative, and then detecting major peaks and valleys in the resulting signal to decide the number of clusters. Our DBE method is nearly “automatic,” depending on just one easy-to-set parameter. Several numerical and real-world examples are presented to illustrate the effectiveness of DBE.
-
ItemAn efficient clustering scheme to exploit hierarchical data in network traffic analysis(IEEE COMPUTER SOC, 2008-06)