Falls are one of the greatest risks for older adults living alone at home. This paper presents a novel visual-based fall detection approach to support independent living for older adults through analysing the motion and shape of the human body. The proposed approach employs a new set of features to detect a fall. Motion information of a segmented silhouette when extracted can provide a useful cue for classifying different behaviours, while variation in shape and the projection histogram can be used to describe human body postures and subsequent fall events. The proposed approach presented here extracts motion information using best-fit approximated ellipse and bounding box around the human body, produces projection histograms and determines the head position over time, to generate 10 features to identify falls. These features are fed into a multilayer perceptron neural network for fall classification. Experimental results show the reliability of the proposed approach with a high fall detection rate of 99.60% and a low false alarm rate of 2.62% when tested with the UR Fall Detection dataset. Comparisons with state of the art fall detection techniques show the robustness of the proposed approach.

Falls are one of the greatest risks for the older adults living alone at home. This paper presents a novel visual-based fall detection approach to support independent living for older adults. The proposed approach employs three unique features; motion information, human shape variation and projection histogram to detect a fall. Motion information of a segmented silhouette, which when extracted can provide a useful cue for classifying different behaviours. Also, the projection histogram and variation in human shape can be used to describe human body postures and subsequently fall events. The proposed approach presented here extracts motion information, using best-fit approximated ellipse around the human body and in addition projection histogram features to further improve the accuracy of fall detection. Experimental results are presented and show high fall detection rate of 99.81% with partially occluded video data.

Computer vision systems offer a new promising solution which can help older people stay at home by providing a secure environment and improve their quality of life. One application area of video surveillance is to analyse human behaviour and detect unusual behaviour. Falls are one of the greatest risks for the elderly living at home. This paper presents a novel approach for detecting falls, based on a combination of motion information and human shape variation. The motion information of a segmented silhouette, when extracted can provide a useful cue for classifying different behaviours. Also, the variation in human shape can used to establish the pose and hence fall events. The approach presented here extracts motion information, use variation in shape and in addition use best-fit approximated ellipse around the human body to further improved the accuracy of falls detection. Result of our approach demonstrates a 20% improvement over motion information only implementations.

The vanishing gradients problem inherent in Simple Recurrent Networks (SRN) trained with back-propagation, has led to a significant shift towards the use of Long Short-Term Memory (LSTM) and Echo State Networks (ESN), which overcome this problem through either second order error-carousel schemes or different learning algorithms, respectively. This paper re-opens the case for SRN-based approaches, by considering a variant, the Multi-recurrent Network (MRN). We show that memory units embedded within its architecture can ameliorate against the vanishing gradient problem, by providing variable sensitivity to recent and more historic information through layer- and self-recurrent links with varied weights, to form a so-called sluggish state-based memory. We demonstrate that an MRN, optimised with noise injection, is able to learn the long term dependency within a complex grammar induction task, significantly outperforming the SRN, NARX and ESN. Analysis of the internal representations of the networks, reveals that sluggish state-based representations of the MRN are best able to latch on to critical temporal dependencies spanning variable time delays, to maintain distinct and stable representations of all underlying grammar states. Surprisingly, the ESN was unable to fully learn the dependency problem, suggesting the major shift towards this class of models may be premature.

Understanding human behaviour and activities is a challenging problem in computer vision. In application areas like health care and ambient intelligence, the use of a camera feed might be seen as too invasive and may be resented. Human behaviour understanding can combine images, signals, feature extraction and other machine learning techniques. This paper presents an overview of our technique that aims to investigate low cost and acceptable visual camera monitoring systems for the elderly. The main idea is to limit the amount of information that needs to be transmitted from the visual sensor unit. The proposed technique will use only filtered images, without saving or transmitting any visual information and thus maintaining privacy.

This article illustrates the utility of mixed methods research (i.e., combining quantitative and qualitative techniques) to the field of school psychology. First, the use of mixed methods approaches in school psychology practice is discussed. Second, the mixed methods research process is described in terms of school psychology research. Third, the current state of affairs with respect to mixed methods designs in school psychology research is illustrated through a mixed methods analysis of the types of empirical studies published in the four leading school psychology journals between 2001 and 2005. Only 13.7% of these studies were classified as representing mixed methods research. We conclude that this relatively small proportion likely reflects the fact that only 3.5% of graduate-level school psychology programs appear to require that students enroll in one or more qualitative and/or mixed methods research courses, and only 19.3% appear to offer one or more qualitative courses as an elective. Finally, the utility of mixed methods research is illustrated by critiquing select monomethod (i.e., qualitative or quantitative) and mixed methods studies conducted on the increasingly important topic of bullying. We demonstrate how using mixed methods techniques results in richer data being collected, leading to a greater understanding of underlying phenomena. (c) 2008 Wiley Periodicals, Inc.

The current study examined important predictors of substance use during early adolescence. The authors hypothesized that adolescents' relationships with key adults (i.e., teachers and parents) influence their choices to use substances indirectly through links with their decisions regarding peer groups. A total of 461 middle school students from an affluent suburban community completed self-report measures of authoritative parenting, perceived social support from teachers, affiliation with rule-breaking and substance-using peers, and frequency of alcohol, cigarette, and drug use. Results of structural equation modeling supported the hypothesized model. Authoritative parenting and teacher support accounted for 31% of the variance in affiliation with deviant peers which, in turn, accounted for 27% of the variance in adolescent substance use; direct paths from parenting and teacher support to substance use were not indicated. Implications for school psychologists' involvement in substance use prevention and intervention are discussed.

As family and peers are primary socializing agents in the lives of young adults, a social learning based model of communication about HIV/AIDS among dating partners was developed and tested, examining the role of interactions with family and peers in this type of communication. Specifically, the model describes relationships between general communication, communication about sexuality, and communication about HIV/AIDS with parents, peers, and dating partners. Participants were 153 young adult couples who completed measures of their communication practices, as well as their communication with family and peers. Communication practices in the family of origin appear to influence both general communication and communication about HIV/AIDS with dating partners. Communication practices with peers influenced general communication, communication about sexuality, and communication about HIV/AIDS with dating partners. Participants and their dating partners exhibited relative agreement about their general communication practices and their communication about HIV/AIDS, but showed less agreement in reports of their communication about sexuality. Implications for understanding the role of family and peer interactions in communication about HIV/AIDS with dating partners are discussed.

This study tested the hypothesis that symptoms of depression are negatively related to relational quality, which in turn is negatively related to feelings of loneliness among members of dating couples. Potential sex differences in the magnitude of association between depressive symptoms and relational quality, and potential emotional contagion of depressive symptoms within dyads, were also explored. One hundred and one dating couples completed the Oral History Interview along with other measures of relational quality, depressive symptoms, and loneliness. Results for both males and females indicated that depressive symptoms were negatively associated with relational quality and that relational quality was negatively associated with loneliness. The association between symptoms of depression and poor relational quality was similar for females and males. There was no evidence suggestive of emotional contagion in these dating couples. Implications of these findings and their potential limitations are discussed.

The key developments of two decades of connectionist parsing are reviewed. Connectionist parsers are assessed according to their ability to learn to represent syntactic structures from examples automatically, without being presented with symbolic grammar rules. This review also considers the extent to which connectionist parsers offer computational models of human sentence processing and provide plausible accounts of psycholinguistic data. In considering these issues, special attention is paid to the level of realism, the nature of the modularity, and the type of processing that is to be found in a wide range of parsers.

Teaching of initial programming is a significant pedagogical problem for computing departments. It is shown that by understanding the changing characteristics of computing students helps to identify their learning approaches and requirements. These findings are used to explain the rationale for the development and use of a virtual learning environment to support the learning of introductory programming.

The Sixth Natural Language Processing Pacific Rim Symposium was held on 27--30 November 2001, at the National Center of Science, Tokyo, Japan.

We describe a deterministic shift-reduce parsing model that combines the advantages of connectionism with those of traditional symbolic models for parsing realistic sub-domains of natural language. It is a modular system that learns to annotate natural language texts with syntactic structure. The parser acquires its linguistic knowledge directly from pre-parsed sentence examples extracted from an annotated corpus. The connectionist modules enable the automatic learning of linguistic constraints and provide a distributed representation of linguistic information that exhibits tolerance to grammatical variation. The inputs and outputs of the connectionist modules represent symbolic information which can be easily manipulated and interpreted and provide the basis for organizing the parse. Performance is evaluated using labelled precision and recall. (For a test set of 4128 words, precision and recall of 75% and 69%, respectively, were achieved.) The work presented represents a significant step towards demonstrating that broad coverage parsing of natural language can be achieved with simple hybrid connectionist architectures which approximate shift-reduce parsing behaviours. Crucially, the model is adaptable to the grammatical framework of the training corpus used and so is not predisposed to a particular grammatical formalism.

The objective of this study was to determine the accuracy of administrative data (by use of hospital discharge codes) for measuring comorbidity in patients with heart disease. One thousand seven hundred and sixty-five medical records of subjects admitted to hospital for AMI, unstable angina, angina pectoris, chronic IHD or heart failure were reviewed. The number and types of comorbidities were determined from the medical records (regarded as the "gold standard"). These were compared with the 10 discharge codes obtained from the hospital administrative records (referred to as the "administrative data"). The rate of false-negative and false-positive comorbidity diagnoses were determined. Twenty of the 21 comorbidities studied were underreported in the administrative data. For these 20 comorbidities, the median false-negative rate was 49.5% and ranged from 11% for diabetes to 100% for dementia. False-positive rates were low, less than 1.5%, except for chronic arrythmia (4.8%) and hypertension (4.2%). Mean percent agreement was high, ranging from 88% for hypertension to 100% for AIDS/HIV. Administrative data based on hospital discharge codes consistently underestimate the presence of comorbid conditions in our population. This has implications for administrators when estimating mortality, length of stay and disability. Researchers also need to be aware when using administrative data based on hospital discharge codes to assess subject's comorbidities that they may be widely underreported.

OBJECTIVE To report on the nature, incidence and severity of problems commonly experienced by cardiac patients in the early months of recovery, and to test the hypotheses that there exist differences in the incidences of these problems depending on age and sex. METHODS 1124 emergency cardiac patients discharged from hospital with acute myocardial infarction, unstable angina, stable angina pectoris, chronic ischaemic heart disease or heart failure were surveyed 4 months after discharge. They were asked to indicate how often during the previous 2 weeks they had experienced each of a range of feelings and problems common to cardiac patients. RESULTS A large proportion of patients reported experiencing problems in the areas of emotional reactions (70%), physical condition (79%), convalescence (67%) and relating to family and friends (63%). Severe problems were experienced especially in the physical and convalescence areas (43% and 44%, respectively). A greater proportion of patients diagnosed with heart failure experienced problems than those with other diagnoses, and these problems were more severe. Amongst myocardial infarction patients, a greater proportion of females than males reported severe problems in the emotional and physical areas, and patients 65 years and over were more likely than younger patients to report experiencing severe problems with physical condition. CONCLUSIONS Many cardiac patients are experiencing psychosocial problems 4 months after hospital discharge, especially with physical activities and convalescence. A knowledge of the incidence and nature of these problems may help nurses to assist patients to validate their experiences.

Examines the pedagogical effectiveness of a natural language exploratory tool developed to supplement a hypermedia learning environment. The learning environment includes an authoring tool with which the relationships between entities can be easily expressed. These relationships are utilised directly by a natural language interface to engage the learner. By closely analysing the activities of learners using the tool, this study compares natural language exploration with the basic interaction provided by hypermedia environments. To facilitate this study, extensive data was compiled using both qualitative and quantitative techniques. Analysis of learner interactions revealed differences in their interactive behaviour. The group engaged in a discourse was more likely to be examining a coherent set of entities at any one time, and was less likely to experience feelings of disorientation. It is concluded that the natural language tool supports a wide range of learning styles.

Connectionism is a relatively new approach to language processing and has comparatively few standard methods for syntax analysis and parsing relative to classical symbolic methods. The interest in connectionism has arisen due to its learning capability, tolerance to noisy input, and ability to generalize from previous examples. Classical rule-based techniques are well understood but tend to be intolerant of minor variations that do not strictly adhere to predefined rules.