A new approach for finding an appropriate combination of texture parameters for classification

This paper suggests an approach for finding an appropriate combination of various parameters for extracting texture features (e.g. choice of spectral band for extracting texture feature, size of the moving window, quantization level of the image, and choice of texture feature etc.) to be used in the classification process. Gray level co-occurrence matrix (GLCM) method has been used for extracting texture from remotely sensed satellite image. Results of the classification of an Indian urban environment using spatial property (texture), derived from spectral and multi-resolution wavelet decomposed images have also been reported. A multivariate data analysis technique called ‘conjoint analysis’ has been used in the study to analyze the relative importance of these parameters. Results indicate that the choice of texture feature and window size have higher relative importance in the classification process than quantization level or the choice of image band for extracting texture feature. In case of texture features derived using wavelet decomposed image, the parameter ‘decomposition level’ has almost equal relative importance as the size of moving window and the decomposition of images up to level one is sufficient and there is no need to go for further decomposition. It was also observed that the classification incorporating texture features improves the overall classification accuracy in a statistically significant manner in comparison to pure spectral classification.

Quality of child abuse information for measurement and surveillance : linking hospital and child welfare data

National estimates of the prevalence of child abuse-related injuries are obtained from a variety of sectors including welfare, justice, and health resulting in inconsistent estimates across sectors. The International Classification of Diseases (ICD) is used as the international standard for categorising health data and aggregating data for statistical purposes, though there has been limited validation of the quality, completeness or concordance of these data with other sectors. This research study examined the quality of documentation and coding of child abuse recorded in hospital records in Queensland and the concordance of these data with child welfare records. A retrospective medical record review was used to examine the clinical documentation of over 1000 hospitalised injured children from 20 hospitals in Queensland. A data linkage methodology was used to link these records with records in the child welfare database. Cases were sampled from three sub-groups according to the presence of target ICD codes: Definite abuse, Possible abuse, unintentional injury. Less than 2% of cases coded as being unintentional were recoded after review as being possible abuse, and only 5% of cases coded as possible abuse cases were reclassified as unintentional, though there was greater variation in the classification of cases as definite abuse compared to possible abuse. Concordance of health data with child welfare data varied across patient subgroups. This study will inform the development of strategies to improve the quality, consistency and concordance of information between health and welfare agencies to ensure adequate system responses to children at risk of abuse.

Beyond statistical procedures for predictive modelling: Data mining algorithms and support for university research at QUT

In a seminal data mining article, Leo Breiman [1] argued that to develop effective predictive classification and regression models, we need to move away from the sole dependency on statistical algorithms and embrace a wider toolkit of modeling algorithms that include data mining procedures. Nevertheless, many researchers still rely solely on statistical procedures when undertaking data modeling tasks; the sole reliance on these procedures has lead to the development of irrelevant theory and questionable research conclusions ([1], p.199). We will outline initiatives that the HPC & Research Support group is undertaking to engage researchers with data mining tools and techniques; including a new range of seminars, workshops, and one-on-one consultations covering data mining algorithms, the relationship between data mining and the research cycle, and limitations and problems with these new algorithms. Organisational limitations and restrictions to these initiatives are also discussed.

Recognising audio-visual speech in vehicles using the AVICAR database

Interacting with technology within a vehicle environment using a voice interface can greatly reduce the effects of driver distraction. Most current approaches to this problem only utilise the audio signal, making them susceptible to acoustic noise. An obvious approach to circumvent this is to use the visual modality in addition. However, capturing, storing and distributing audio-visual data in a vehicle environment is very costly and difficult. One current dataset available for such research is the AVICAR [1] database. Unfortunately this database is largely unusable due to timing mismatch between the two streams and in addition, no protocol is available. We have overcome this problem by re-synchronising the streams on the phone-number portion of the dataset and established a protocol for further research. This paper presents the first audio-visual results on this dataset for speaker-independent speech recognition. We hope this will serve as a catalyst for future research in this area.

Using routinely collected hospital data for child maltreatment surveillance : issues, methods and patterns

Background: International data on child maltreatment are largely derived from child protection agencies, and predominantly report only substantiated cases of child maltreatment. This approach underestimates the incidence of maltreatment and makes inter-jurisdictional comparisons difficult. There has been a growing recognition of the importance of health professionals in identifying, documenting and reporting suspected child maltreatment. This study aimed to describe the issues around case identification using coded morbidity data, outline methods for selecting and grouping relevant codes, and illustrate patterns of maltreatment identified. Methods: A comprehensive review of the ICD-10-AM classification system was undertaken, including review of index terms, a free text search of tabular volumes, and a review of coding standards pertaining to child maltreatment coding. Identified codes were further categorised into maltreatment types including physical abuse, sexual abuse, emotional or psychological abuse, and neglect. Using these code groupings, one year of Australian hospitalisation data for children under 18 years of age was examined to quantify the proportion of patients identified and to explore the characteristics of cases assigned maltreatment-related codes. Results: Less than 0.5% of children hospitalised in Australia between 2005 and 2006 had a maltreatment code assigned, almost 4% of children with a principal diagnosis of a mental and behavioural disorder and over 1% of children with an injury or poisoning as the principal diagnosis had a maltreatment code assigned. The patterns of children assigned with definitive T74 codes varied by sex and age group. For males selected as having a maltreatment-related presentation, physical abuse was most commonly coded (62.6% of maltreatment cases) while for females selected as having a maltreatment-related presentation, sexual abuse was the most commonly assigned form of maltreatment (52.9% of maltreatment cases). Conclusion: This study has demonstrated that hospital data could provide valuable information for routine monitoring and surveillance of child maltreatment, even in the absence of population-based linked data sources. With national and international calls for a public health response to child maltreatment, better understanding of, investment in and utilisation of our core national routinely collected data sources will enhance the evidence-base needed to support an appropriate response to children at risk.

Reliability of routinely collected hospital data for child maltreatment surveillance

Background: Internationally, research on child maltreatment-related injuries has been hampered by a lack of available routinely collected health data to identify cases, examine causes, identify risk factors and explore health outcomes. Routinely collected hospital separation data coded using the International Classification of Diseases and Related Health Problems (ICD) system provide an internationally standardised data source for classifying and aggregating diseases, injuries, causes of injuries and related health conditions for statistical purposes. However, there has been limited research to examine the reliability of these data for child maltreatment surveillance purposes. This study examined the reliability of coding of child maltreatment in Queensland, Australia. Methods: A retrospective medical record review and recoding methodology was used to assess the reliability of coding of child maltreatment. A stratified sample of hospitals across Queensland was selected for this study, and a stratified random sample of cases was selected from within those hospitals. Results: In 3.6% of cases the coders disagreed on whether any maltreatment code could be assigned (definite or possible) versus no maltreatment being assigned (unintentional injury), giving a sensitivity of 0.982 and specificity of 0.948. The review of these cases where discrepancies existed revealed that all cases had some indications of risk documented in the records. 15.5% of cases originally assigned a definite or possible maltreatment code, were recoded to a more or less definite strata. In terms of the number and type of maltreatment codes assigned, the auditor assigned a greater number of maltreatment types based on the medical documentation than the original coder assigned (22% of the auditor coded cases had more than one maltreatment type assigned compared to only 6% of the original coded data). The maltreatment types which were the most ‘under-coded’ by the original coder were psychological abuse and neglect. Cases coded with a sexual abuse code showed the highest level of reliability. Conclusion: Given the increasing international attention being given to improving the uniformity of reporting of child-maltreatment related injuries and the emphasis on the better utilisation of routinely collected health data, this study provides an estimate of the reliability of maltreatment-specific ICD-10-AM codes assigned in an inpatient setting.

Assigning metabolic equivalent (MET) values to the 2002 Census Occupational Classification System

Background: The Current Population Survey (CPS) and the American Time Use Survey (ATUS) use the 2002 census occupation system to classify workers into 509 separate occupations arranged into 22 major occupational categories. Methods: We describe the methods and rationale for assigning detailed MET estimates to occupations and present population estimates (comparing outputs generated by analysis of previously published summary MET estimates to the detailed MET estimates) of intensities of occupational activity using the 2003 ATUS data comprised of 20,720 respondents, 5,323 (2,917 males and 2,406 females) of whom reported working 6+ hours at their primary occupation on their assigned reporting day. Results: Analysis using the summary MET estimates resulted in 4% more workers in sedentary occupations, 6% more in light, 7% less in moderate, and 3% less in vigorous compared to using the detailed MET estimates. The detailed estimates are more sensitive to identifying individuals who do any occupational activity that is moderate or vigorous in intensity resulting in fewer workers in sedentary and light intensity occupations. Conclusions: Since CPS/ATUS regularly captures occupation data it will be possible to track prevalence of the different intensity levels of occupations. Updates will be required with inevitable adjustments to future occupational classification systems.

Overview of the INEX 2010 XML mining track : clustering and classification of XML documents

The XML Document Mining track was launched for exploring two main ideas: (1) identifying key problems and new challenges of the emerging field of mining semi-structured documents, and (2) studying and assessing the potential of Machine Learning (ML) techniques for dealing with generic ML tasks in the structured domain, i.e., classification and clustering of semi-structured documents. This track has run for six editions during INEX 2005, 2006, 2007, 2008, 2009 and 2010. The first five editions have been summarized in previous editions and we focus here on the 2010 edition. INEX 2010 included two tasks in the XML Mining track: (1) unsupervised clustering task and (2) semi-supervised classification task where documents are organized in a graph. The clustering task requires the participants to group the documents into clusters without any knowledge of category labels using an unsupervised learning algorithm. On the other hand, the classification task requires the participants to label the documents in the dataset into known categories using a supervised learning algorithm and a training set. This report gives the details of clustering and classification tasks.

Towards a classification of criterion referenced assessment models in mathematics courses : student and academic perspectives

We present the findings of a study into the implementation of explicitly criterion- referenced assessment in undergraduate courses in mathematics. We discuss students' concepts of criterion referencing and also the various interpretations that this concept has among mathematics educators. Our primary goal was to move towards a classification of criterion referencing models in quantitative courses. A secondary goal was to investigate whether explicitly presenting assessment criteria to students was useful to them and guided them in responding to assessment tasks. The data and feedback from students indicates that while students found the criteria easy to understand and useful in informing them as to how they would be graded, it did not alter the way the actually approached the assessment activity.

Spectral–texture feature extraction using statistical moments with application to object-based vegetation species classification

The use of appropriate features to characterise an output class or object is critical for all classification problems. In order to find optimal feature descriptors for vegetation species classification in a power line corridor monitoring application, this article evaluates the capability of several spectral and texture features. A new idea of spectral–texture feature descriptor is proposed by incorporating spectral vegetation indices in statistical moment features. The proposed method is evaluated against several classic texture feature descriptors. Object-based classification method is used and a support vector machine is employed as the benchmark classifier. Individual tree crowns are first detected and segmented from aerial images and different feature vectors are extracted to represent each tree crown. The experimental results showed that the proposed spectral moment features outperform or can at least compare with the state-of-the-art texture descriptors in terms of classification accuracy. A comprehensive quantitative evaluation using receiver operating characteristic space analysis further demonstrates the strength of the proposed feature descriptors.

“What’s going on here?” The pedagogy of a data analysis session

Data analysis sessions are a common feature of discourse analytic communities, often involving participants with varying levels of expertise to those with significant expertise. Learning how to do data analysis and working with transcripts, however, are often new experiences for doctoral candidates within the social sciences. While many guides to doctoral education focus on procedures associated with data analysis (Heath, Hindmarsh, & Luff, 2010; McHoul & Rapley, 2001; Silverman, 2011; Wetherall, Taylor, & Yates, 2001), the in situ practices of doing data analysis are relatively undocumented. This chapter has been collaboratively written by members of a special interest research group, the Transcript Analysis Group (TAG), who meet regularly to examine transcripts representing audio- and video-recorded interactional data. Here, we investigate our own actual interactional practices and participation in this group where each member is both analyst and participant. We particularly focus on the pedagogic practices enacted in the group through investigating how members engage in the scholarly practice of data analysis. A key feature of talk within the data sessions is that members work collaboratively to identify and discuss ‘noticings’ from the audio-recorded and transcribed talk being examined, produce candidate analytic observations based on these discussions, and evaluate these observations. Our investigation of how talk constructs social practices in these sessions shows that participants move fluidly between actions that demonstrate pedagogic practices and expertise. Within any one session, members can display their expertise as analysts and, at the same time, display that they have gained an understanding that they did not have before. We take an ethnomethodological position that asks, ‘what’s going on here?’ in the data analysis session. By observing the in situ practices in fine-grained detail, we show how members participate in the data analysis sessions and make sense of a transcript.

The sample complexity of pattern classification with neural networks: the size of the weights is more important than the size of the network

Sample complexity results from computational learning theory, when applied to neural network learning for pattern classification problems, suggest that for good generalization performance the number of training examples should grow at least linearly with the number of adjustable parameters in the network. Results in this paper show that if a large neural network is used for a pattern classification problem and the learning algorithm finds a network with small weights that has small squared error on the training patterns, then the generalization performance depends on the size of the weights rather than the number of weights. For example, consider a two-layer feedforward network of sigmoid units, in which the sum of the magnitudes of the weights associated with each unit is bounded by A and the input dimension is n. We show that the misclassification probability is no more than a certain error estimate (that is related to squared error on the training set) plus A3 √((log n)/m) (ignoring log A and log m factors), where m is the number of training patterns. This may explain the generalization performance of neural networks, particularly when the number of training examples is considerably smaller than the number of weights. It also supports heuristics (such as weight decay and early stopping) that attempt to keep the weights small during training. The proof techniques appear to be useful for the analysis of other pattern classifiers: when the input domain is a totally bounded metric space, we use the same approach to give upper bounds on misclassification probability for classifiers with decision boundaries that are far from the training examples.

Utility of routine data sources for feedback on the quality of cancer care: an assessment based on clinical practice guidelines

Background Not all cancer patients receive state-of-the-art care and providing regular feedback to clinicians might reduce this problem. The purpose of this study was to assess the utility of various data sources in providing feedback on the quality of cancer care. Methods Published clinical practice guidelines were used to obtain a list of processes-of-care of interest to clinicians. These were assigned to one of four data categories according to their availability and the marginal cost of using them for feedback. Results Only 8 (3%) of 243 processes-of-care could be measured using population-based registry or administrative inpatient data (lowest cost). A further 119 (49%) could be measured using a core clinical registry, which contains information on important prognostic factors (e.g., clinical stage, physiological reserve, hormone-receptor status). Another 88 (36%) required an expanded clinical registry or medical record review; mainly because they concerned long-term management of disease progression (recurrences and metastases) and 28 (11.5%) required patient interview or audio-taping of consultations because they involved information sharing between clinician and patient. Conclusion The advantages of population-based cancer registries and administrative inpatient data are wide coverage and low cost. The disadvantage is that they currently contain information on only a few processes-of-care. In most jurisdictions, clinical cancer registries, which can be used to report on many more processes-of-care, do not cover smaller hospitals. If we are to provide feedback about all patients, not just those in larger academic hospitals with the most developed data systems, then we need to develop sustainable population-based data systems that capture information on prognostic factors at the time of initial diagnosis and information on management of disease progression.

Rural and Urban Differentials in Primary Care Management of Chronic Heart Failure: New Data from the CASE Study

Objective: To determine whether primary care management of chronic heart failure (CHF) differed between rural and urban areas in Australia. Design: A cross-sectional survey stratified by Rural, Remote and Metropolitan Areas (RRMA) classification. The primary source of data was the Cardiac Awareness Survey and Evaluation (CASE) study. Setting: Secondary analysis of data obtained from 341 Australian general practitioners and 23 845 adults aged 60 years or more in 1998. Main outcome measures: CHF determined by criteria recommended by the World Health Organization, diagnostic practices, use of pharmacotherapy, and CHF-related hospital admissions in the 12 months before the study. Results: There was a significantly higher prevalence of CHF among general practice patients in large and small rural towns (16.1%) compared with capital city and metropolitan areas (12.4%) (P < 0.001). Echocardiography was used less often for diagnosis in rural towns compared with metropolitan areas (52.0% v 67.3%, P < 0.001). Rates of specialist referral were also significantly lower in rural towns than in metropolitan areas (59.1% v 69.6%, P < 0.001), as were prescribing rates of angiotensin-converting enzyme inhibitors (51.4% v 60.1%, P < 0.001). There was no geographical variation in prescribing rates of β-blockers (12.6% [rural] v 11.8% [metropolitan], P = 0.32). Overall, few survey participants received recommended “evidence-based practice” diagnosis and management for CHF (metropolitan, 4.6%; rural, 3.9%; and remote areas, 3.7%). Conclusions: This study found a higher prevalence of CHF, and significantly lower use of recommended diagnostic methods and pharmacological treatment among patients in rural areas.

On a Syntactic Characterization of Classification with a Mind Change Bound

Most learning paradigms impose a particular syntax on the class of concepts to be learned; the chosen syntax can dramatically affect whether the class is learnable or not. For classification paradigms, where the task is to determine whether the underlying world does or does not have a particular property, how that property is represented has no implication on the power of a classifier that just outputs 1’s or 0’s. But is it possible to give a canonical syntactic representation of the class of concepts that are classifiable according to the particular criteria of a given paradigm? We provide a positive answer to this question for classification in the limit paradigms in a logical setting, with ordinal mind change bounds as a measure of complexity. The syntactic characterization that emerges enables to derive that if a possibly noncomputable classifier can perform the task assigned to it by the paradigm, then a computable classifier can also perform the same task. The syntactic characterization is strongly related to the difference hierarchy over the class of open sets of some topological space; this space is naturally defined from the class of possible worlds and possible data of the learning paradigm.