Clustering based multi-label classification for image annotation and retrieval

This paper presents a novel multi-label classification framework for domains with large numbers of labels. Automatic image annotation is such a domain, as the available semantic concepts are typically hundreds. The proposed framework comprises an initial clustering phase that breaks the original training set into several disjoint clusters of data. It then trains a multi-label classifier from the data of each cluster. Given a new test instance, the framework first finds the nearest cluster and then applies the corresponding model. Empirical results using two clustering algorithms, four multi-label classification algorithms and three image annotation data sets suggest that the proposed approach can improve the performance and reduce the training time of standard multi-label classification algorithms, particularly in the case of large number of labels.

Dual-random ensemble method for multi-label classification of biological data

This paper presents a dual-random ensemble multi-label classification method for classification of multi-label data. The method is formed by integrating and extending the concepts of feature subspace method and random k-label set ensemble multi-label classification method. Experiemental results show that the developed method outperforms the exisiting multi-lable classification methods on three different multi-lable datasets including the biological yeast and genbase datasets.

Image to text translation by multi-label classification

This paper presents an image to text translation platform consisting of image segmentation, region features extraction, region blobs clustering, and translation components. A multi-label learning method is suggested for realizing the translation component. Empirical studies show that the predictive performance of the translation component is better than its counterparts when employed a dual-random ensemble multi-label classification algorithm.

Image to text translation by multi-label classification

This thesis includes the development of an architectural framework for the proposed image to text translation system containing four components. Selection of appropriate algorithms for the first three components developed three effective multi-label classification algorithms for the fourth component, i.e. the translation component, for different problem settings.

Empirical study of multi-label classification methods for image annotation and retrieval

This paper presents an empirical study of multi-label classification methods, and gives suggestions for multi-label classification that are effective for automatic image annotation applications. The study shows that triple random ensemble multi-label classification algorithm (TREMLC) outperforms among its counterparts, especially on scene image dataset. Multi-label k-nearest neighbor (ML-kNN) and binary relevance (BR) learning algorithms perform well on Corel image dataset. Based on the overall evaluation results, examples are given to show label prediction performance for the algorithms using selected image examples. This provides an indication of the suitability of different multi-label classification methods for automatic image annotation under different problem settings.

A triple-random ensemble classification method for mining multi-label data

This paper presents a triple-random ensemble learning method for handling multi-label classification problems. The proposed method integrates and develops the concepts of random subspace, bagging and random k-label sets ensemble learning methods to form an approach to classify multi-label data. It applies the random subspace method to feature space, label space as well as instance space. The devised subsets selection procedure is executed iteratively. Each multi-label classifier is trained using the randomly selected subsets. At the end of the iteration, optimal parameters are selected and the ensemble MLC classifiers are constructed. The proposed method is implemented and its performance compared against that of popular multi-label classification methods. The experimental results reveal that the proposed method outperforms the examined counterparts in most occasions when tested on six small to larger multi-label datasets from different domains. This demonstrates that the developed method possesses general applicability for various multi-label classification problems.

Osteo-cise : Strong bones for life': protocol for a community-based randomised controlled trial of a multi-modal exercise and osteoporosis education program for older adults at risk of falls and fractures

Background : Osteoporosis affects over 220 million people worldwide, and currently there is no 'cure' for the disease. Thus, there is a need to develop evidence-based, safe and acceptable prevention strategies at the population level that target multiple risk factors for fragility fractures to reduce the health and economic burden of the condition.

Methods : The 'Osteo-cise: Strong Bones for Life' study will investigate the effectiveness and feasibility of a multi-component targeted exercise, osteoporosis education/awareness and behavioural change program for improving bone health and muscle function, and reducing falls risk in community-dwelling older adults at an increased risk of fracture. Men and women aged 60 years or above will participate in an 18-month randomised controlled trial comprising a 12-month structured and supervised community-based program and a 6-month 'research to practise' translational phase. Participants will be randomly assigned to either the 'Osteo-cise' intervention or a self-management control group. The intervention will comprise a multi-modal exercise program incorporating high velocity progressive resistance training, moderate impact weight-bearing exercise and high challenging balance exercises performed three times weekly at local community-based fitness centres. A behavioural change program will be used to enhance exercise adoption and adherence to the program. Community-based osteoporosis education seminars will be conducted to improve participant knowledge and understanding of the risk factors and preventative measures for osteoporosis, falls and fractures. The primary outcomes measures, to be collected at baseline, 6, 12, and 18 months, will include DXA-derived hip and spine bone mineral density measurements and functional muscle power (timed stair-climb test). Secondary outcomes measures include: MRI-assessed distal femur and proximal tibia trabecular bone micro-architecture, lower limb and back maximal muscle strength, balance and function (four square step test, functional reach test, timed up-and-go test and 30-second sit-to-stand), falls incidence and health-related quality of life. Cost-effectiveness will also be assessed.

Discussion : The findings from the Osteo-cise: Strong Bones for Life study will provide new information on the efficacy of a targeted multi-modal community-based exercise program incorporating high velocity resistance training, together with an osteoporosis education and behavioural change program for improving multiple risk factors for falls and fracture in older adults at risk of fragility fracture. Trial Registration: Australian New Zealand Clinical Trials Registry reference ACTRN12609000100291

M-Dimension : multi-characteristics based routing protocol in human associated delay-tolerant networks with improved performance over one dimensional classic models

Human associated delay-tolerant network (HDTN) is a new delay-tolerant network where mobile devices are associated with humans. It can be viewed from both their geographic and social dimensions. The combination of these different dimensions can enable us to more accurately comprehend a delay-tolerant network and consequently use this multi-dimensional information to improve overall network efficiency. Alongside the geographic dimension of the network which is concerned with geographic topology of routing, social dimensions such as social hierarchy can be used to guide the routing message to improve not only the routing efficiency for individual nodes, but also efficiency for the entire network.

We propose a multi-dimensional routing protocol (M-Dimension) for the human associated delay-tolerant network which uses the local information derived from multiple dimensions to identify a mobile node more accurately. Each dimension has a weight factor and is organized by the Distance Function to select an intermediary and applies multi-cast routing. We compare M-Dimension to existing benchmark routing protocols using the MIT Reality Dataset, a well-known benchmark dataset based on a human associated mobile network trace file. The results of our simulations show that M-Dimension has a significant increase in the average success ratio and is very competitive when End-to-End Delay of packet delivery is used in comparison to other multi-cast DTN routing protocols.

Comparative evaluation of multi-label classification methods

This paper presents a comparative evaluation of popular multi-label classification methods on several multi-label problems from different domains. The methods include multi-label k-nearest neighbor, binary relevance, label power set, random k-label set ensemble learning, calibrated label ranking, hierarchy of multi-label classifiers and triple random ensemble multi-label classification algorithms. These multi-label learning algorithms are evaluated using several widely used MLC evaluation metrics. The evaluation results show that for each multi-label classification problem a particular MLC method can be recommended. The multi-label evaluation datasets used in this study are related to scene images, multimedia video frames, diagnostic medical report, email messages, emotional music data, biological genes and multi-structural proteins categorization.

Multi-label classification with clustering for image and text categorization

This paper explores effective multi-label classification methods for multi-semantic image and text categorization. We perform an experimental study of clustering based multi-label classification (CBMLC) for the target problem. Experimental evaluation is conducted for identifying the impact of different clustering algorithms and base classifiers on the predictive performance and efficiency of CBMLC. In the experimental setting, three widely used clustering algorithms and six popular multi-label classification algorithms are used and evaluated on multi-label image and text datasets. A multi-label classification evaluation metrics, micro F1-measure, is used for presenting predictive performances of the classifications. Experimental evaluation results reveal that clustering based multi-label learning algorithms are more effective compared to their non-clustering counterparts.

Out & online; effectiveness of a tailored online multi-symptom mental health and wellbeing program for same-sex attracted young adults: study protocol for a randomised controlled trial

Same-sex attracted young adults have been found to experience higher rates of mental health problems and greater difficulties in accessing specialist mental health care services compared to their heterosexual peers. Internet-based mental health interventions have the potential to be more engaging and accessible to young adults compared to those delivered face-to-face. However, they are rarely inclusive of lesbian women and gay men. Thus, the current study aims to evaluate the effectiveness of an online mental health and wellbeing program, Out & Online (http://www.outandonline.org.au), in comparison to a wait-list control group, for reducing anxiety and depressive symptoms in same-sex attracted young adults aged between 18 and 25 years.

The effects of a protein enriched diet with lean red meat combined with a multi-modal exercise program on muscle and cognitive health and function in older adults: study protocol for a randomised controlled trial

BACKGROUND: Age-related muscle wasting has been strongly implicated with falls and fractures in the elderly, but it has also been associated with cognitive decline and dementia. Progressive resistance training (PRT) and adequate dietary protein are recognised as important contributors to the maintenance of muscle health and function in older adults. However, both factors also have the potential to improve brain function and prevent cognitive decline via several pathways, including the regulation of various growth and neurotrophic factors [insulin-like growth factor-1 (IGF-1)]; brain-derived growth factor (BDNF)] and/or the modulation of systemic inflammation. The primary aim of this study is to investigate whether a modest increase in dietary protein achieved through the consumption of lean red meat three days per week, when combined with PRT, can enhance muscle mass, size and strength and cognitive function in community-dwelling older people. METHODS/DESIGN: The study design is a 48-week randomised controlled trial consisting of a 24-week intervention with a 24-week follow-up. Men and women (n=152) aged 65 years and over residing in the community will be randomly allocated to: 1) PRT and provided with 220 g (raw weight) of lean red meat to be cooked and divided into two 80 g servings on each of the three days that they complete their exercise session, or 2) control PRT in which participants will be provided with and advised to consume ≥1 serving (~1/2 cup) of rice and/or pasta or 1 medium potato on each of the three training days. The primary outcome measures will be muscle mass, size and strength and cognitive function. Secondary outcomes will include changes in: muscle function, neural health (corticospinal excitability and inhibition and voluntary activation), serum IGF-1 and BDNF, adipokines and inflammatory markers, fat mass and inter-/intra-muscular fat, blood pressure, lipids and health-related quality of life. All outcome measures will be assessed at baseline and 24 weeks, with the exception of cognitive function and the various neurobiological and inflammatory markers which will also be assessed at week 12. DISCUSSION: The findings from this study will provide important new information on whether a modest increase in dietary protein achieved through the ingestion of lean red meat can enhance the effects of PRT on muscle mass, size and strength as well as cognitive function in community-dwelling older adults. If successful, the findings will form the basis for more precise exercise and nutrition guidelines for the management and prevention of age-related changes in muscle and neural health and cognitive function in the elderly. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry: ACTRN12613001153707 . Date registered 16(th) October, 2013.