Joel Scanlan

DynamicWEB: Adapting to Concept Drift and Object Drift in COBWEB

Examining concepts that change over time has been an active area of research within data mining. This paper presents a new method that functions in contexts where concept drift is present, while also allowing for modification of the instances themselves as they change over time. This method is well suited to domains where subjects of interest are sampled multiple times, and where they may migrate from one resultant concept to another due to Object Drift. The method presented here is an extensive modification to the conceptual clustering algorithm COBWEB, and is titled DynamicWEB.

Increasing knowledge of mental illness through secondary research of electronic health records: opportunities and challenges

Aim/Purpose: The primary use of electronic health records (EHRs) is in the care of the individual patient. Secondary research uses employ information in EHRs for purposes beyond that of care of the individual. Secondary research uses may broadly be divided into studies which focus on improving care and treatment of individuals and those which aim to increase knowledge about disease causes, associations and prevalence at a population level. This paper provides a review of studies that have used EHRs to increase knowledge at a population level. It examines the methods used, types of research conducted, difficulties and challenges faced and implications for future research and mental health research in particular. Method: A review was undertaken based on a search for peer-reviewed and recently (i.e. since 2005) published articles with full-text available online. Findings/Results: The studies which have used EHRs to increase knowledge have predominantly involved; (1) data mining to identify biomarkers and gene–disease associations, (2) epidemiological research using linked/merged health records and (3) surveillance, prediction and alerts for diseases/illnesses. The principal methodological challenges identified were data quality, discrepancies/inconsistencies in data and interoperability of EHRs. Conclusions: Despite the challenges faced in secondary usage of EHRs, significant research has been undertaken and researchers have proposed and tested various approaches to address methodological issues. The study methods employed in other fields of medical research can be extrapolated to study issues of significance to mental health using EHRs.

DynamicWEB: profile correlation using COBWEB

Establishing relationships within a dataset is one of the core objectives of data mining. In this paper a method of correlating behaviour profiles in a continuous dataset is presented. The profiling problem which motivated the research is intrusion detection. The profiles are dynamic in nature, changing frequently, and are made up of many attributes. The paper describes a modified version of the COBWEB hierarchical conceptual clustering algorithm called DynamicWEB. DynamicWEB operates at runtime, keeping the profiles up to date, and in the correct location within the clustering tree. Further, as there are a number of attributes within the domain of interest, the tree also extends multi-dimensionally. This allows for multiple correlations to occur simultaneously, focusing on different attributes within the one profile.

Personally controlled electronic health records in Australia: Challenges in communication of mental health information

Abstract The purpose of this paper was to explore the implications of mental health literacy for uptake, use and benefits of Australia’s Personally Controlled Electronic Health Record (PCEHR). A narrative review was undertaken using literature gained through university search engines, Google Scholar, and government and other reputable websites. Documents retrieved were predominantly recent (i.e., since 2005) and Australian-based. Key findings were that low levels of mental health literacy can adversely affect: Interpretations of health-related information, help-seeking behaviours, use of health services, and peer support for those living with mental illness. Consumers with low levels of mental health literacy, as observed in many disadvantaged groups in Australia, may benefit from additional support in order to use and derive the benefits envisaged for the PCEHR. It was concluded that low levels of mental health literacy may limit the uptake, use and benefits of Australia’s PCEHR. A number of possible strategies to assist consumers with low mental health literacy were discussed. It was noted that targeted approaches to addressing mental health literacy in disadvantaged groups are warranted to minimise disparity in health care access and long-term health outcomes. It was also suggested that healthcare providers could play an important role in encouraging uptake of the PCEHR by placing greater emphasis on patient education and support.

A performance profile of learner completion and retention in Australian VET MOOCs

Abstract The uptake of Massive Open Online Courses (MOOCs) has been substantial and they continue to flourish as an educational delivery model. Much of the available literature regarding student retention and completions in MOOCs comes from university-designed courses. The Vocational Education and Training (VET) sector has lagged behind its university counterparts as VET pedagogy relies heavily on skills acquisition and does not lend itself as easily to MOOC delivery. This paper analysed the attributes of 366,099 enrolled learners and 132,867 learners that were designated as ‘starters’ from 263 Australian universities, 27 transnational universities and 73 VET MOOCs to compare the differences in learning environments and to build a VET MOOC learner profile. The findings indicated that higher levels of VET MOOC learners were retained when compared to university MOOCs. It also revealed eight course design factors that may contribute to increased learner retention and completions for VET MOOCs.

DynamicWEB: A Method for Reconnaissance Activity Profiling

Port scan correlation aims to differentiate between benign and malicious scans. In this paper we will examine a new method of profiling port scan activity in an attempt to link different source IP addresses to being the same end user. A data mining approach DynamicWEB based upon the COBWEB conceptual clustering algorithm is shown along with some preliminary results of it functioning within the context of scan correlation.

Mobile Application Based Heavy Vehicle Fatigue Compliance in Australian Operations

The Australian National Heavy Vehicle Regulator (NHVR) defines the rules for fatigue management in heavy haulage trucking operations. The rules place restrictions on total work and minimum rest hours, and are aimed at regulating the potential for fatigue risk amongst drivers. This paper presents a performance-based fatigue management system based on driver fatigue data stored in simple mobile databases and deployed via Android smart phones. The system funded by WorkSafe Tasmania and entitled, Logistics Fatigue Manager (LFM), was evaluated with a cohort of heavy haulage drivers in Australian forestry. The correlation between driver fatigue estimates and actual sleep hours (recorded using FitBits) is confirmed, and is also supported through driver interviews. The benefit is that management of fatigue risk could be more tailored to individual drivers opening up efficiency gains across supply chains.

Risk prediction using natural language processing of electronic mental health records in an Inpatient forensic psychiatry setting

OBJECTIVE Instruments rating risk of harm to self and others are widely used in inpatient forensic psychiatry settings. A potential alternate or supplementary means of risk prediction is from the automated analysis of case notes in Electronic Health Records (EHRs) using Natural Language Processing (NLP). This exploratory study rated presence or absence and frequency of words in a forensic EHR dataset, comparing four reference dictionaries. Seven machine learning algorithms and different time periods of EHR analysis were used to probe which dictionary and which time period were most predictive of risk assessment scores on validated instruments. MATERIALS AND METHODS The EHR dataset comprised de-identified forensic inpatient notes from the Wilfred Lopes Centre in Tasmania. The data comprised unstructured free-text case note entries and serial ratings of three risk assessment scales: Historical Clinical Risk Management-20 (HCR-20), Short-Term Assessment of Risk and Treatability (START) and Dynamic Appraisal of Situational Aggression (DASA). Four NLP dictionary word lists were selected: 6865 mental health symptom words from the Unified Medical Language System (UMLS), 455 DSM-IV diagnoses from UMLS repository, 6790 English positive and negative sentiment words, and 1837 high frequency words from the Corpus of Contemporary American English (COCA). Seven machine learning methods Bagging, J48, Jrip, Logistic Model Trees (LMT), Logistic Regression, Linear Regression and Support Vector Machine (SVM) were used to identify the combination of dictionaries and algorithms that best predicted risk assessment scores. RESULTS The most accurate prediction was attained on the DASA dataset using the sentiment dictionary and the LMT and SVM algorithms. CONCLUSIONS NLP, used in conjunction with NLP dictionaries and machine learning, predicted risk ratings on the HCR-20, START, and DASA, based on EHR content. Further research is required to ascertain the utility of NLP approaches in predicting endpoints of actual self-harm, harm to others or victimisation.

Engagement and retention in VET MOOCs and online courses: A systematic review of literature from 2013 to 2017

Building stronger structures that encourage deeper levels of learner engagement and retention in Massive Open Online Courses (MOOCs) is of significant interest to teachers of Vocational Education and Training (VET). Previous literature on MOOCs is predominately occupied with university sector developments and alternative educational contexts such as VET are neglected. This systematic review of literature from 2013 to 2017 evaluated learner engagement and retention in university MOOCs and VET online courses to identify functional approaches that could be implemented into VET MOOCs. Ten databases were searched, eliciting 1950 papers, which were then screened. Data from 30 university MOOCs and 8 VET online delivery articles that met the inclusion and quality assurance criteria were analysed. Four key themes and eleven component categories emerged repeatedly across the literature. Analysis revealed six functional approaches relevant to VET MOOCs. The findings suggested that coupling these functional approaches into VET MOOCs can improve learner retention and promote engagement. The implications for practice and further research are presented.