Articles

SMS Spam Filtering: Methods and Data

Sarah Jane Delany et al. — Mon, 01 Oct 2012 08:45:19 PDT

Mobile or SMS spam is a real and growing problem primarily due to the availability of very cheap bulk pre-pay SMS packages and the fact that SMS engenders higher response rates as it is a trusted and personal service. SMS spam filtering is a relatively new task which inherits many issues and solu- tions from email spam filtering. However it poses its own specific challenges. This paper motivates work on filtering SMS spam and reviews recent devel- opments in SMS spam filtering. The paper also discusses the issues with data collection and availability for furthering research in this area, analyses a large corpus of SMS spam, and provides some initial benchmark results.

Profiling Instances in Noise Reduction

Sarah Jane Delany et al. — Mon, 01 Oct 2012 02:20:24 PDT

The dependency on the quality of the training data has led to significant work in noise reduction for instance-based learning algorithms. This paper presents an empirical evaluation of current noise reduction techniques, not just from the perspective of their comparative performance, but from the perspective of investigating the types of instances that they focus on for re- moval. A novel instance profiling technique known as RDCL profiling allows the structure of a training set to be analysed at the instance level cate- gorising each instance based on modelling their local competence properties. This profiling approach o↵ers the opportunity of investigating the types of instances removed by the noise reduction techniques that are currently in use in instance-based learning. The paper also considers the e↵ect of removing instances with specific profiles from a dataset and shows that a very simple approach of removing instances that are misclassified by the training set and cause other instances in the dataset to be misclassified is an e↵ective noise reduction technique.

Systematically Evaluating the Effectiveness of Quality Assurance Programmes in Leading to Improvements in Institutional Performance

Deirdre Lillis — Mon, 17 Sep 2012 03:00:01 PDT

Higher education institutions worldwide invest significant resources in their quality assurance systems however little empirical research exists which demonstrates their effectiveness (or otherwise). Methodological approaches for determining effectiveness are also underdeveloped. Self study with peer review is a widely used model for ensuring the quality of the core teaching, research and engagement activities of higher education institutions. This paper illustrates how an established social programme evaluation methodology can be used to determine its effectiveness in leading to improvements in institutional performance. The concept of effectiveness and the particular challenges posed by the higher education organisational culture are considered. An example of the systematic evaluation of three self study programmes is provided to illustrate the concept. It is concluded that social programme evaluation has significant potential in evaluating the effectiveness of quality assurance initiatives in higher education.

Programme Oriented and Institutional Oriented Approaches to Quality Assurance: New Developments and Mixed Approaches

Deirdre Lillis et al. — Mon, 17 Sep 2012 03:00:00 PDT

This paper considers the programme validation arrangements in place in one half of the Irish higher education sector. It outlines how responsibility for programme validation can be safely delegated to Institutions within a robust overarching framework for quality assurance. It compares programme validation in Institutions with self awarding status with Institutions that have their programmes validated by a national Awarding agency. The paper concludes that when programme validation in Ireland and (potentially) across Europe is examined more closely, processes that appear to be very different on the surface can be quite similar in reality. From a philosophical perspective it appears that the degree to which providers can be trusted to manage their own quality assurance is a key consideration, however difficult it may be to measure trust.

Irish Education and the Financial Crisis

Deirdre Lillis et al. — Mon, 17 Sep 2012 02:59:59 PDT

Regardless of economic fortunes, investment in education remains a constant in terms of its perceived value to Ireland and it is seen as critical to the economic recovery of the country. Ireland has one of the highest educational participation rates in the world, is considered to have the ‘most employable’ graduates in Europe and produces more graduates per 1000 inhabitants than any other European country. The degree to which the financial crisis has impacted Irish education is explored here.

Using Semi-supervised Classifiers for Credit Scoring

Kenneth Kennedy et al. — Thu, 08 Dec 2011 08:59:02 PST

In credit scoring, low-default portfolios are those for which very little default history exists. This makes it problematic for financial institutions to estimate a reliable probability of a customer defaulting on a loan. Banking regulation (Basel II Capital Accord), and best practice, however, necessitate an accurate and valid estimate of the probability of default. In this article the suitability of semi-supervised one-class classification algorithms as a solution to the lowdefault portfolio problem are evaluated. The performance of one-class classification algorithms is compared with the performance of supervised two-class classification algorithms. This study also investigates the suitability of oversampling, which is a common approach to dealing with low-default portfolios. Assessment of the performance of oneand two-class classification algorithms using nine real-world banking data sets, which have been modified to replicate low-default portfolios, is provided. Our results demonstrate that only in the near or complete absence of defaulters should semi-supervised one-class classification algorithms be used instead of supervised two-class classification algorithms. Furthermore, we demonstrate for data sets whose class labels are unevenly distributed that optimising the threshold value on classifier output yields, in many cases, an improvement in classification performance. Finally, our results suggest that oversampling produces no overall improvement to the best performing two-class classification algorithms.

Forty Years of Movie Hacking: Considering the Potential Implications of the Popular Media Representation of Computer Hackers from 1968 to 2008

Damian Gordon — Tue, 22 Feb 2011 07:00:40 PST

Increasingly movies are being produced which feature plots that incorporate elements of computer security and hacking, and cumulatively these movies are creating a public perception as to the nature of computer security. This research examines movies that feature hackers (and hacking) to identify if any common themes emerge from these movies in their representation of these issues. To achieve this, first a corpus of hacking movies is created, and then using a qualitative data analysis technique, guidelines are developed which distinguish those movies that actually have the potential to create a perception with the general public. The resultant dataset is analysed and the salient details are compared to the reality of hacking. This research has implications in a range of fields, including in the education of computer students, organisation computer security and in the behaviour of the general public when using computers.

Activity Recognition using Temporal Evidence Theory

Susan McKeever et al. — Fri, 04 Feb 2011 06:37:02 PST

The ability to identify the behavior of people in a home is at the core of Smart Home functionality. Such environments are equipped with sensors that unobtrusively capture information about the occupants. Reasoning mechanisms transform the technical, frequently noisy data of sensors into meaningful interpretations of occupant activities. Time is a natural human way to reason about activities. Peoples‟ activities in the home often have an identifiable routine; activities take place at distinct times throughout the day and last for predicable lengths of time. However, the inclusion of temporal information is still limited in the domain of activity recognition. Evidence theory is gaining increasing interest in the field of activity recognition, and is suited to the incorporation of time related domain knowledge into the reasoning process. In this paper, an evidential reasoning framework that incorporates temporal knowledge is presented. We evaluate the effectiveness of the framework using a third party published smart home dataset. An improvement in activity recognition of 70% is achieved when time patterns and activity durations are included in activity recognition. We also compare our approach with Naïve Bayes classifier and J48 Decision Tree, with temporal evidence theory achieving higher accuracies than both classifiers.

Applying Computational Models of Spatial Prepositions to Visually Situated Dialog

John Kelleher et al. — Wed, 26 Jan 2011 14:25:15 PST

This article describes the application of computational models of spatial prepositions to visually situated dialog systems. In these dialogs, spatial prepositions are important because people often use them to refer to entities in the visual context of a dialog. We first describe a generic architecture for a visually situated dialog system and highlight the interactions between the spatial cognition module, which provides the interface to the models of prepositional semantics, and the other components in the architecture. Following this, we present two new computational models of topological and projective spatial prepositions. The main novelty within these models is the fact that they account for the contextual effect which other distractor objects in a visual scene can have on the region described by a given preposition. We next present psycholinguistic tests evaluating our approach to distractor interference on prepositional semantics, and illustrate how these models are used for both interpretation and generation of prepositional expressions.

Motion in Augmented Reality Games: an Engine for Creating Plausible Physical Interactions in Augmented Reality Games

Brian Mac Namee et al. — Tue, 18 Jan 2011 10:21:30 PST

The next generation of Augmented Reality (AR) games will require real and virtual objects to coexist in motion in immersive game environments. This will require the illusion that real and virtual objects interact physically together in a plausible way. The Motion in Augmented Reality Games (MARG) engine described in this paper has been developed to allow these kinds of game environments. The paper describes the design and implementation of the MARG engine and presents two proof-of-concept AR games that have been developed using it. Evaluations of these games have been performed and are presented to show that the MARG engine takes an important step in developing the next generation of motion-rich AR games.

Knowledge Management in a Project Environment: Organisational CT and Project Influences

Taya Polyaninova — Tue, 21 Dec 2010 01:54:26 PST

During a project implementation various forms of information and experience are generated within the organization. If this accumulated knowledge is not recorded and shared amongst other projects, this knowledge will be lost and no longer be available to assist future projects. This may lead to increased future projects costs as resources, time and money will be wasted on redefining the knowledge that once existed within the company. By not capturing and redeploying this knowledge, the quality of a project’s deliverables may adversely suffer. First the publication reviews the concept of project knowledge management. It defines the reasons for managing project knowledge, the benefits it brings to organizations and challenges that exist within the organization that prevent the process from being successfully deployed. Secondly, the sources of knowledge within projects and strategies that can be used by organizations to manage this knowledge are reviewed. Further more the paper examines a number of studies to identify the practices of project knowledge management within different organizations. Finally, the paper draws conclusions on the influences and implementation of project knowledge management by project-based organizations. From the findings it is evident that the nature of the projects and organizational culture has a significant impact on the ability to manage project knowledge which in turn aids to the projects success or failure. The paper assists practitioners in understanding the complexities of establishing organizational and project culture to enable knowledge management activities within projects in order to improve project performance and enhance chances of project success.

Attention Driven Reference Resolution in Multimodal Contexts

John Kelleher — Tue, 03 Feb 2009 08:33:03 PST

In recent years a a number of psycholinguistic experiments have pointed to the interaction between language and vision. In particular, the interaction between visual attention and linguistic reference. In parallel with this, several theories of discourse have attempted to provide an account of the relationship between types of referential expressions on the one hand and the degree of mental activation on the other. Building on both of these traditions, this paper describes an attention based approach to visually situated reference resolution. The framework uses the relationship between referential form and preferred mode of interpretation as a basis for a weighted integration of linguistic and visual attention scores for each entity in the multimodal context. The resulting integrated attention scores are then used to rank the candidate referents during the resolution process, with the candidate scoring the highest selected as the referent. One advantage of this approach is that the resolution process occurs within the full multimodal context, in so far as the referent is selected from a full list of the objects in the multimodal context. As a result situations where the intended target of the reference is erroneously excluded, due to an individual assumption within the resolution process, are avoided. Moreover, the system can recognise situations where attention cues from different modalities make a reference potentially ambiguous.

Strategies for Evaluating Software Usability

Ronan Fitzpatrick — Thu, 29 Jan 2009 06:25:15 PST

This paper presents a usage analysis and taxonomy of methods which are used to evaluate the usability of computer systems. To accommodate the analysis and taxonomy, a matrix of strategies which can be used for effective usability evaluation is presented. Such an analysis, taxonomy and strategies support human-computer interaction (HCI) professionals who have the responsibility for ensuring computer system usability. The strategies outlined are named Virtual Engineering, Soft Modelling, Hard Review and Real World. This paper also uses a composite set of existing popular generic evaluation methods which can be used as part of these strategies. The methods used are observation, questionnaire, interview, empirical methods, user groups, cognitive walkthroughs, heuristic methods, review methods and model methods. The paper continues by presenting a Usage Analysis Table of these methods and concludes by grouping them into a Taxonomy of Usability Evaluation Methods. A key emphasis of this paper is the appropriateness of individual methods to lifecycle timing.