Extending WHIRL with background knowledge for improved text classification

Intelligent use of the many diverse forms of data available on the Internet requires new tools for managing and manipulating heterogeneous forms of information. This paper uses WHIRL, an extension of relational databases that can manipulate textual data using statistical similarity measures developed by the information retrieval community. We show that although WHIRL is designed for more general similarity-based reasoning tasks, it is competitive with mature systems designed explicitly for inductive classification. In particular, WHIRL is well suited for combining different sources of knowledge in the classification process. We show on a diverse set of tasks that the use of appropriate sets of unlabeled background knowledge often decreases error rates, particularly if the number of examples or the size of the strings in the training set is small. This is especially useful when labeling text is a labor-intensive job and when there is a large amount of information available about a particular problem on the World Wide Web.

The influence of game location and outcome on behaviour and mood states among professional rugby league players

In this study, we examined the relationship between home and away matches on mood. In addition, the relationships between game location, game outcome, behavioural factors, and mood were investigated among 12 professional English rugby league players competing in the Super League. Participants completed daily diaries for 27 days. The diary consisted of six analogue scales measuring mood (relaxed-tense, energetic-weary, depressed-elated, tired-alert, anxious-calm, cheerful-miserable) as well as behavioural factors and self-rated performance. There were no significant differences in self-reported mood states leading up to home or away matches except for players feeling more tired when playing away. Significant relationships between mood and behaviours (e.g. sleep and eating) and subjective performance were observed. The outcome of the match was found to influence mood, with a defeat resulting in decreased mood. Our results show that game location did not influence mood and therefore does not provide an explanation for the home advantage.     

Interpreting the differences: A reply to Kaufman

Kaufman's critique of our article was based on a misunderstanding of our purpose and procedures. This reply clarifies those points and suggests further areas of concern regarding the McCarthy Screening Test.     

Identifying Attention Deficit Hyperactive Disorder with the WISC-R and the Stroop Color and Word Test

Twenty-nine children with a diagnosis of Attention Deficit Hyperactive Disorder (ADHD) were compared to a group of 21 emotionally disturbed (ED) children and a control group (CO) of 20 nonproblem children. The meaures used in the comparison were the 12 subtests of the Wechsler Intelligence Scale for Children-Revised (WISC-R) and the Stroop Color and Word Test (SCWT). The results showed that the CO group was superior to the ADHD and ED groups and that the ED group performed better than the ADHD group on most of the variables. A stepwise discriminant function analysis showed that similar WISC-R subtests differentiated between the ADHD group and the ED and CO groups. Two measures of the SCWT helped in discriminating between the ADHD and ED groups, but not between the ADHD and CO groups. The authors discuss the meaning of these findings and their contribution to better understanding of the ADHD group, its specific cognitive processing, and its problems.