Joint Analysis of Multiple Algorithms and Performance Measures

There has been an increasing interest in the development of new methods using Pareto optimality to deal with multi-objective criteria (for example, accuracy and time complexity). Once one has developed an approach to a problem of interest, the problem is then how to compare it with the state of art. In machine learning, algorithms are typically evaluated by comparing their performance on different data sets by means of statistical tests. Standard tests used for this purpose are able to consider jointly neither performance measures nor multiple competitors at once. The aim of this paper is to resolve these issues by developing statistical procedures that are able to account for multiple competing measures at the same time and to compare multiple algorithms altogether. In particular, we develop two tests: a frequentist procedure based on the generalized likelihood-ratio test and a Bayesian procedure based on a multinomial-Dirichlet conjugate model. We further extend them by discovering conditional independences among measures to reduce the number of parameters of such models, as usually the number of studied cases is very reduced in such comparisons. Data from a comparison among general purpose classifiers is used to show a practical application of our tests.

Abnormal Pedestrian Trajectory analysis based on arbitrary-length clustering

This papers examines the use of trajectory distance measures and clustering techniques to define normal
and abnormal trajectories in the context of pedestrian tracking in public spaces. In order to detect abnormal
trajectories, what is meant by a normal trajectory in a given scene is firstly defined. Then every trajectory
that deviates from this normality is classified as abnormal. By combining Dynamic Time Warping and a
modified K-Means algorithms for arbitrary-length data series, we have developed an algorithm for trajectory
clustering and abnormality detection. The final system performs with an overall accuracy of 83% and 75%
when tested in two different standard datasets.

Informal caregiving and mental ill health – differential relationships by workload, gender, age and area-remoteness in a UK region

Informal caregiving can be a demanding role which has been shown to impact on physical, psychological and social wellbeing. Methodological weaknesses including small sample sizes and subjective measures of mental health have led to inconclusive evidence about the relationship between informal caregiving and mental health. This paper reports on a study carried out in a UK region which investigated the relationship between informal caregiving and mental ill health. The analysis was conducted by linking three datasets, the Northern Ireland Longitudinal Study, the Northern Ireland Enhanced Prescribing Database and the Proximity to Service Index from the Northern Ireland Statistics and Research Agency. Our analysis used both a subjective measure of mental ill health, i.e. a question asked in the 2011 Census, and an objective measure, whether the respondents had been prescribed antidepressants by a General Practitioner between 2010 and 2012. We applied binary logistic multilevel modelling to these two responses to test whether, and for what sub-groups of the population, informal caregiving was related to mental ill health. The results showed that informal caregiving per se was not related to mental ill health although there was a strong relationship between the intensity of the caregiving role and mental ill health. Females under 50, who provided over 19 hours of care, were not employed or worked part-time and who provided care in both 2001 and 2011 were at a statistically significantly elevated risk of mental ill health. Caregivers in remote areas with limited access to shops and services were also at a significantly increased risk as evidenced by prescription rates for antidepressants. With community care policies aimed at supporting people to remain at home, the paper highlights the need for further research in order to target resources appropriately.