Fuzzy clustering with learnable cluster-dependent kernels

We propose a new relational clustering approach, called Fuzzy clustering with Learnable Cluster-dependent Kernels (FLeCK), that learns the underlying cluster-dependent dissimilarity measure while seeking compact clusters. The learned dissimilarity is based on a Gaussian kernel function with cluster-dependent parameters. Each cluster’s parameter learned by FLeCK reflects the relative intra-cluster and inter-cluster characteristics. These parameters are learned by optimizing both the intra-cluster and the inter-cluster distances. This optimization is achieved iteratively by dynamically updating the partition and the local kernel. This makes the kernel learning task takes advantages of the available unlabeled data and reciprocally, the categorization task takes advantages of the learned local kernels. Another key advantage of FLeCK is that it is formulated to work on relational data. This makes it applicable to data where objects cannot be represented by vectors or when clusters of similar objects cannot be represented efficiently by a single prototype. Using synthetic and real data sets, we show that FLeCK learns meaningful parameters and outperforms several other algorithms. In particular, we show that when data include clusters with various inter- and intra-cluster distances, learning cluster-dependent parameters is crucial in obtaining a good partition.     

Exploring and exploiting a historical corpus for Arabic

This paper presents a historical Arabic corpus named HAC. At this early embryonic stage of the project, we report about the design, the architecture and some of the experiments which we have conducted on HAC. The corpus, and accordingly the search results, will be represented using a primary XML exchange format. This will serve as an intermediate exchange tool within the project and will allow the user to process the results offline using some external tools. HAC is made up of Classical Arabic texts that cover 1600 years of language use; the Quranic text, Modern Standard Arabic texts, as well as a variety of monolingual Arabic dictionaries. The development of this historical corpus assists linguists and Arabic language learners to effectively explore, understand, and discover interesting knowledge hidden in millions of instances of language use. We used techniques from the field of natural language processing to process the data and a graph-based representation for the corpus. We provided researchers with an export facility to render further linguistic analysis possible.     

Emotional Loneliness and Coping Strategies: A Reference to Older Malaysians at Nursing Homes

This article presents an in-depth qualitative study using a phenomenological approach to understand loneliness among elderly individuals in Malaysia. The objective of the study was to understand how the Malaysian elderly perceive and understand social isolation as well as loneliness, with the aim of identifying the factors that cause emotional loneliness among the elderly in nursing homes. In addition, this study also explored their coping strategies when dealing with loneliness. Semi-structured interviews were conducted with ten elderly participants from two different nursing homes in Kuala Lumpur and Selangor with representatives from the three major ethnic groups of Malaysia. Based on the results, there are several factors that cause the elderly to feel lonely – health factors, lack of family ties, and the lack of communication and cognitive factors, such as memory and perception. It was also found that internal (expectations and optimism) and external (work and activities) coping strategies play major roles in overcoming loneliness. In conclusion, some recommendations are made to respective party families and the government to consider when developing plans to help the elderly overcome loneliness, which could strengthen the family and social support system in Malaysia.     

Employing both gender and emotion cues to enhance speaker identification performance in emotional talking environments

Speaker recognition performance in emotional talking environments is not as high as it is in neutral talking environments. This work focuses on proposing, implementing, and evaluating a new approach to enhance the performance in emotional talking environments. The new proposed approach is based on identifying the unknown speaker using both his/her gender and emotion cues. Both Hidden Markov Models (HMMs) and Suprasegmental Hidden Markov Models (SPHMMs) have been used as classifiers in this work. This approach has been tested on our collected emotional speech database which is composed of six emotions. The results of this work show that speaker identification performance based on using both gender and emotion cues is higher than that based on using gender cues only, emotion cues only, and neither gender nor emotion cues by 7.22 %, 4.45 %, and 19.56 %, respectively. This work also shows that the optimum speaker identification performance takes place when the classifiers are completely biased towards suprasegmental models and no impact of acoustic models in the emotional talking environments. The achieved average speaker identification performance based on the new proposed approach falls within 2.35 % of that obtained in subjective evaluation by human judges.     

Collaborative learning in the clouds

Cloud computing has emerged as one of the most highly discussed topics both in the academic community and in the computing industry. While most of the work that has been conducted to explore this field focuses either on establishing the basis for cloud computing or almost exclusively on the issues surrounding security and data privacy, this paper takes the first exploratory step into exploring the actual internal working of cloud computing and demonstrates its viability for organizations, more specifically educational establishments . The paper starts by introducing the most important key clouds computing concepts, including virtualization technologies, Web services and Service Oriented Architectures (SOA), and distributed computing. Light will be then shed on the impact and potential benefits of cloud computing on teaching and learning in educational institutions. The paper closes by describing building a private cloud inside educational institution and highlights its offerings for students, staff and lecturers.     

Enhanced fuzzy-filtered neural networks for material fatigue prognosis

Although fuzzy-filtered neural networks (FFNN) have been used in pattern classification because of their unique characteristics in feature extraction, they usually have poor performance in forecasting applications due to their structure complexities especially in their consequent reasoning part. In this paper, an enhanced FFNN, EFFNN, is proposed for time series forecasting and material fatigue prognosis. A novel neural network scheme is developed to facilitate computation implementation. A new conjugate technique is proposed to improve training efficiency. The effectiveness of the developed EFFNN scheme and the related training technique is demonstrated by a series of simulation tests. The EFFNN is also implemented for material fatigue prognosis. Test results show that the developed EFFNN predictor is an effective forecasting tool; it can capture system dynamics effectively and track system characteristics accurately.     

Human immunodeficiency virus 1 (HIV-1)-specific reverse transcriptase (RT) inhibitors may suppress the replication of specific drug-resistant (E138K)RT HIV-1 mutants or select for highly resistant (Y181C-->C181I)RT HIV-1 mutants

Mutant HIV-1 that expresses a Glu138-->Lys substitution in its RT [(E138K)RT] is resistant to the HIV-1-specific RT inhibitor 2',5'-bis-O-(tert-butyldimethylsilyl)-3'-spiro-5"-(4"-amino-1",2"- oxathiole-2",2"-dioxide)pyrimidine (TSAO). However, cell cultures infected with this mutant were completely protected against virus-mediated destruction by micromolar concentrations of the HIV-1-specific RT inhibitors tetrahydroimidazo[4,5,1-jk][1,4]benzodiazepin-2(1H)-one and -thione (TIBO), nevirapine, and bis(heteroaryl)piperazine (BHAP). In contrast, cells infected with a virus mutant that expresses a Tyr181-->Cys substitution in its RT [(Y181C)RT] were not protected by nevirapine and TIBO and were only temporarily protected by BHAP. HIV-1 mutant that emerged under the latter conditions contained a Cys181-->Ile substitution in their RT [(LC181I)RT]. This mutant proved highly resistant to all HIV-1-specific RT inhibitors tested, except for several 1-(2-hydroxyethoxymethyl)-6-(phenylthio)thymine (HEPT) derivatives. When recombinant (C181I)RT was evaluated for susceptibility to the HIV-1-specific RT inhibitors, it was resistant to all inhibitors except the HEPT compounds. Since a (Y181F)RT HIV mutant strain was isolated from cells infected with (Y181C)RT HIV-1 and treated with BHAP, we postulate that the Ile codon was derived from a Cys-->Phe transversion mutation (TGT-->TTT), followed by a Phe-->Ile transversion mutation (TTT-->ATT).     

Development and characterization of a purified diet to identify obesity-susceptible and resistant rat populations

A purified moderately high fat diet has been developed to examine diet-induced obesity in rats. Male Sprague-Dawley rats were fed this or an AIN-76A diet for 15 wk and energy metabolism indices were monitored. Food intake, body weight and water balance indices were recorded on a weekly or daily basis. Over the 15-wk period, rats fed the experimental diet diverged into two groups differing in the rate of body weight gain. Animals were labeled as "gainers" or "resisters" depending on their susceptibility to obesity. Following the dietary period, rats were decapitated and trunk blood was collected for glucose and insulin measurements. Gainers consumed slightly more energy than resisters over the experimental period (P < 0.05), but due to greater fecal energy loss, absorbed energy did not differ. Hence gainers became obese without significantly altered energy retention. Urinary creatinine, urea nitrogen and water balance were not different between the groups and consequently could not explain body weight differences. Further, gainers had significantly greater plasma glucose concentration than controls, indicating a potential for these animals to become diabetic. Results suggest metabolic differences must account for the divergence in weight gain observed in the two groups. The dietary model characterized in this study should provide a useful tool to study diet-induced obesity and to determine its underlying mechanism.     

Mammary gland expression of transgenes and the potential for altering the properties of milk

Transgenic animals are a useful in vivo experimental model for assessing the ability and impact of foreign gene expression in a biological system. Transgenic mice are most commonly used, while transgenic sheep, goats, pigs and cows have also been developed for specific, "applied" purposes. Most of the work directed at targeting expression of transgenes to the mammary gland of an animal, by using a milk gene promoter, has been with the intent of either studying promoter function or recovering the desired protein from the milk. Transgenic technology can also be used to alter the functional and physical properties of milk resulting in novel manufacturing properties. The properties of milk have been altered by adding a new protein with the aim of improving the milk, not of recovering the protein for other uses.     

Semi-Supervised Deep Learning for Multi-Tissue Segmentation from Multi-Contrast MRI

Journal of Signal Processing Systems - Segmentation of thigh tissues (muscle, fat, inter-muscular adipose tissue (IMAT), bone, and bone marrow) from magnetic resonance imaging (MRI) scans is useful...