Fusing data mining, machine learning and traditional statistics to detect biomarkers associated with depression

BACKGROUND: Atheoretical large-scale data mining techniques using machine learning algorithms have promise in the analysis of large epidemiological datasets. This study illustrates the use of a hybrid methodology for variable selection that took account of missing data and complex survey design to identify key biomarkers associated with depression from a large epidemiological study.

METHODS: The study used a three-step methodology amalgamating multiple imputation, a machine learning boosted regression algorithm and logistic regression, to identify key biomarkers associated with depression in the National Health and Nutrition Examination Study (2009-2010). Depression was measured using the Patient Health Questionnaire-9 and 67 biomarkers were analysed. Covariates in this study included gender, age, race, smoking, food security, Poverty Income Ratio, Body Mass Index, physical activity, alcohol use, medical conditions and medications. The final imputed weighted multiple logistic regression model included possible confounders and moderators.

RESULTS: After the creation of 20 imputation data sets from multiple chained regression sequences, machine learning boosted regression initially identified 21 biomarkers associated with depression. Using traditional logistic regression methods, including controlling for possible confounders and moderators, a final set of three biomarkers were selected. The final three biomarkers from the novel hybrid variable selection methodology were red cell distribution width (OR 1.15; 95% CI 1.01, 1.30), serum glucose (OR 1.01; 95% CI 1.00, 1.01) and total bilirubin (OR 0.12; 95% CI 0.05, 0.28). Significant interactions were found between total bilirubin with Mexican American/Hispanic group (p = 0.016), and current smokers (p<0.001).

CONCLUSION: The systematic use of a hybrid methodology for variable selection, fusing data mining techniques using a machine learning algorithm with traditional statistical modelling, accounted for missing data and complex survey sampling methodology and was demonstrated to be a useful tool for detecting three biomarkers associated with depression for future hypothesis generation: red cell distribution width, serum glucose and total bilirubin.

Does student engagement improve when 1:1 device technologies are used and adapted to cater for individual learning styles during online delivery of engineering courses?

BACKGROUND OR CONTEXT: A developing international engineering industry is dependent on competition and innovation, creating a market for highly skilled graduates from respected overseas and Australian Engineering universities. The delivery of engineering teaching and learning via blended faceto-face, problem based, research focused and online collaborative learning will continue to be the foundation of future engineering education, however, it will be those institutions who can reshape its learning spaces within a culture of innovation using 1:1 devices that will continue to attract the brightest minds. Investing in educational research that explores the preferred learning styles of learners and matching this to specifically designed 1:1 personalized web applications may be the ‘value add’ to improve student engagement. In this paper, a survey of Australian engineering education is presented and contrasted against a backdrop of internationally recognised educational pedagogy to demonstrate how engineering teaching and learning has changed over time. This paper draws on research and identifies a gap where a necessity to question the validity of 1:1 devices as the next step in the evolution of engineering education needs to be undertaken. How will teaching and learning look using 1:1 devices and will it drive student demand into engineering higher
education courses. Will this lead to improving professional standards within a dynamic engineering education context? How will current and future teaching and learning be influenced by constructivism using 1:1 device technologies? How will the engineering industry benefit from higher education investment in individualised engineering education
using 1:1 devices for teaching and learning?
PURPOSE OR GOAL: To review the current academic thinking around the topic of 1:1 devices within higher education engineering teaching and learning context in Australia. To identify any gaps in the current understandings and use of 1:1 devices within engineering courses in Australia. To generate discussion and better understanding about how the use of 1:1 devices may hinder and/or improve teaching and learning and student engagement.
APPROACH: A review covering the development of engineering education in Australia and a broader international review of engineering teaching methodology. To identify the extent of research into the use and effectiveness of online strategies within engineering education utilising 1:1 devices for teaching and learning. i.e. “Students must feel that they are part of a learning community and derive motivation to engage in the study material from the lecturer.’ (Lloyd et al., 2001) It is proposed to add to the current body of understandings and explore the effectiveness of a constructiveness teaching approach using course material specifically designed to cater for individual learning styles and delivered via the use of 1:1 devices in the classroom. It is anticipated the research will contrast current engineering teaching and learning practices and identify factors that will facilitate a greater understanding about student connectedness and engagement with the teaching and learning experience; where a constructiveness environment is supported with the use of 1:1 devices. Also, it is anticipated that the constructed learning environment will foster a culture of innovation and students will be empowered to take control of their own learning and be encouraged to contribute back to the discussion initiated by the lecture and/or course material with the aid of 1:1 device technologies. A gap has been identified in the academic literature that show there is a need to understand the relationship between engineering teaching, learning, students engagement and the use of 1:1 devices.
DISCUSSION: A review covering the development of engineering education in Australia and a broader international review of engineering teaching methodology. To identify the extent of research into the use and effectiveness of online strategies within engineering education utilising 1:1 devices for teaching and learning. i.e. “Students must feel that they are part of a learning community and derive motivation to engage in the study material from the lecturer.’ (Lloyd et al., 2001) It is proposed to add to the current body of understandings and explore the effectiveness of a constructiveness teaching approach using course material specifically designed to cater for individual learning styles and delivered via the use of 1:1 devices in the classroom.
ANTICIPATED OUTCOMES: It is anticipated the research will contrast current engineering teaching and learning practices and identify factors that will facilitate a greater understanding about student connectedness and engagement with the teaching and learning experience; where a constructiveness environment is supported with the use of 1:1 devices. Also, it is anticipated that the constructed learning environment will foster a culture of innovation and students will be empowered to take control of their own learning and be encouraged to contribute back to the discussion initiated by the lecture and/or course material with the aid of 1:1 device technologies. A gap has been identified in the academic literature that show there is a need to understand the relationship between engineering teaching, learning, students engagement and the use of 1:1 devices.
RECOMMENDATIONS/IMPLICATIONS/CONCLUSION: A gap exists in the current research about the effectiveness and use of 1:1 devices in engineering education; therefore, it is necessary to undertake further research in the area. It is proposed to hypothesize and conduct field research to identify any shortcomings and possible benefits for engineering educators and learners within a constructivist-teaching
context that explores the relationship between the use of personalized 1:1 devices for teaching and learning, adapting for individual learning styles, and identification and application of appropriate teaching and learning strategies within a constructiveness engineering course approach. Research is required to clarify the following research questions;
• What education teaching and learning strategies best facilitate the use of 1:1 devices for online teaching and learning?
• Does student engagement improve when 1:1 device technologies are used and adapted to cater for individual learning styles during online delivery of engineering courses?
• What are the factors within a university engineering faculty that may hinder and/or support the use of 1:1 devices for online teaching and learning?
• To what extent do 1:1 devices assist engineering educators and students to foster a culture of innovation? The study results will offer engineering educators and students an opportunity to reflect on
their current teaching and learning practice, and contextualise the use of 1:1 devices as a tool to improve student engagement. It is expected the learning benefits will outweigh the implementation costs and derive a unique learning experience that will empower engineering educators and students to inspire a culture of innovation.

A performance evaluation of machine learning-based streaming spam tweets detection

The popularity of Twitter attracts more and more spammers. Spammers send unwanted tweets to Twitter users to promote websites or services, which are harmful to normal users. In order to stop spammers, researchers have proposed a number of mechanisms. The focus of recent works is on the application of machine learning techniques into Twitter spam detection. However, tweets are retrieved in a streaming way, and Twitter provides the Streaming API for developers and researchers to access public tweets in real time. There lacks a performance evaluation of existing machine learning-based streaming spam detection methods. In this paper, we bridged the gap by carrying out a performance evaluation, which was from three different aspects of data, feature, and model. A big ground-truth of over 600 million public tweets was created by using a commercial URL-based security tool. For real-time spam detection, we further extracted 12 lightweight features for tweet representation. Spam detection was then transformed to a binary classification problem in the feature space and can be solved by conventional machine learning algorithms. We evaluated the impact of different factors to the spam detection performance, which included spam to nonspam ratio, feature discretization, training data size, data sampling, time-related data, and machine learning algorithms. The results show the streaming spam tweet detection is still a big challenge and a robust detection technique should take into account the three aspects of data, feature, and model.

Simplified framework for managing team learning in engineering subjects

Teamwork has been included as a major component of graduate attributes in all engineering programs at universities. In spite of enormous research advances in theoretical aspects of learning and working in teams, anecdotal evidence suggests that most engineering academic staff are inundated by student complaints of not being able to learn and work in teams due to numerous reasons. In order to facilitate engineering academic staff and engineering schools, this study develops a simplified framework for managing learning teams in engineering subjects that integrates theoretical conceptions, empirical evidences and anecdotal practices by reviewing a substantial body of existing literature. The framework identifies that in addition to managing student complaints about learning and working in teams more effectively and efficiently, engineering academic staff and engineering schools need to focus on specifying learning outcomes of teamwork, identifying appropriate approaches to achieve these learning outcomes, judging the suitability of teamwork-based learning in a particular educational context, developing a clear plan for implementing teamwork, implementing and monitoring teamwork, and reflecting and re-evaluating teamwork. The developed framework can be a useful tool to help understand these essential components and complexities of team learning.

Rock slope stability analyses using extreme learning neural network and terminal steepest descent algorithm

The analysis of rock slope stability is a classical problem for geotechnical engineers. However, for practicing engineers, proper software is not usually user friendly, and additional resources capable of providing information useful for decision-making are required. This study developed a convenient tool that can provide a prompt assessment of rock slope stability. A nonlinear input-output mapping of the rock slope system was constructed using a neural network trained by an extreme learning algorithm. The training data was obtained by using finite element upper and lower bound limit analysis methods. The newly developed techniques in this study can either estimate the factor of safety for a rock slope or obtain the implicit parameters through back analyses. Back analysis parameter identification was performed using a terminal steepest descent algorithm based on the finite-time stability theory. This algorithm not only guarantees finite-time error convergence but also achieves exact zero convergence, unlike the conventional steepest descent algorithm in which the training error never reaches zero.

Informal networked learning as teamwork in design studio Cmyview: using mobile digital technologies to connect with student's everyday experiences

CmyView is a research project that investigates how mobile technologies have the potential to facilitate new ways to share, experience and understand the connections that people have with places. The aim of the project is to theorise and develop a tool and a methodology that addresses the reception of architecture and the built environment using mobile digital technologies that harness ubiquitous everyday practices, such as photography and walking. While CmyView is primarily focused on evidencing the reception of places, this chapter argues that these activities can also make a contribution to the core pedagogy of architectural education, the design studio. This chapter presents findings of an initial pilot study with four students at an Australian university that demonstrates how CmyView offers a valuable contribution to the educational experience in the design studio.

Purposeful exploratory learning with video using analysis categories

There is still much to be learnt about best practices in leveraging digital resources for learning in higher education. Research on student interactions with online video indicates such practices are as minimal as setting passive-receptive viewing through to teacher-structured purposeful engagement. This position paper focuses on teacher-set analysis categories to guide student exploration of digital video content and to help novices to scaffold their thinking. Various uses of analysis categories within one Australian university in conjunction with a video annotation tool are reviewed. Then practice examples from other universities are reviewed to demonstrate the use of analysis categories in higher education settings without an annotation tool. The literature indicates that the use of categories to inform the design of digital video analysis needs to ensure that the learning challenge is retained. Analysis guided by teacher-set categories tends to be beneficial for performance evaluation in particular. Further research on university teacher practices with digital video is required.

Student perspectives on the value of experiential learning in history

This paper is a reflective study of experiential learning as an American history teaching-tool. It is based on a survey of students who took a University of Melbourne study tour to the United States in the years from 2001–2011. This survey asked students to identify the tour’s long-term outcomes. The responses showed that students believed the study tour was beneficial academically, and that it also opened up employment opportunities. However, the most significant benefit identified by the students was positive social outcomes—in other words, the friends they made on the tour and the professional networks they formed. The conclusion we drew from these results was that students believe that experiential learning has a legitimate place in history curriculums, and that it is an antidote to the loneliness they feel in traditional classroom settings.