Parallel user profiling based on folksonomy for Large Scaled Recommender Systems : an implimentation of Cascading MapReduce

The Large scaled emerging user created information in web 2.0 such as tags, reviews, comments and blogs can be used to profile users’ interests and preferences to make personalized recommendations. To solve the scalability problem of the current user profiling and recommender systems, this paper proposes a parallel user profiling approach and a scalable recommender system. The current advanced cloud computing techniques including Hadoop, MapReduce and Cascading are employed to implement the proposed approaches. The experiments were conducted on Amazon EC2 Elastic MapReduce and S3 with a real world large scaled dataset from Del.icio.us website.

Learning the kernel matrix with semidefinite programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive semidefinite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space - classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semidefinite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -using the labeled part of the data one can learn an embedding also for the unlabeled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method for learning the 2-norm soft margin parameter in support vector machines, solving an important open problem.

Learning the Kernel Matrix with Semi-Definite Programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive definite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space -- classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semi-definite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -- using the labelled part of the data one can learn an embedding also for the unlabelled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method to learn the 2-norm soft margin parameter in support vector machines, solving another important open problem. Finally, the novel approach presented in the paper is supported by positive empirical results.

Frequent activating FGFR2 mutations in endometrial carcinomas parallel germline mutations associated with craniosynostosis and skeletal dysplasia syndromes

Endometrial carcinoma is the most common gynecological malignancy in the United States. Although most women present with early disease confined to the uterus, the majority of persistent or recurrent tumors are refractory to current chemotherapies. We have identified a total of 11 different FGFR2 mutations in 3/10 (30%) of endometrial cell lines and 19/187 (10%) of primary uterine tumors. Mutations were seen primarily in tumors of the endometrioid histologic subtype (18/115 cases investigated, 16%). The majority of the somatic mutations identified were identical to germline activating mutations in FGFR2 and FGFR3 that cause Apert Syndrome, Beare-Stevenson Syndrome, hypochondroplasia, achondroplasia and SADDAN syndrome. The two most common somatic mutations identified were S252W (in eight tumors) and N550K (in five samples). Four novel mutations were identified, three of which are also likely to result in receptor gain-of-function. Extensive functional analyses have already been performed on many of these mutations, demonstrating they result in receptor activation through a variety of mechanisms. The discovery of activating FGFR2 mutations in endometrial carcinoma raises the possibility of employing anti-FGFR molecularly targeted therapies in patients with advanced or recurrent endometrial carcinoma.

Massively parallel sequencing and analysis of expressed sequence tags in a successful invasive plant

Background Invasive species pose a significant threat to global economies, agriculture and biodiversity. Despite progress towards understanding the ecological factors associated with plant invasions, limited genomic resources have made it difficult to elucidate the evolutionary and genetic factors responsible for invasiveness. This study presents the first expressed sequence tag (EST) collection for Senecio madagascariensis, a globally invasive plant species. Methods We used pyrosequencing of one normalized and two subtractive libraries, derived from one native and one invasive population, to generate an EST collection. ESTs were assembled into contigs, annotated by BLAST comparison with the NCBI non-redundant protein database and assigned gene ontology (GO) terms from the Plant GO Slim ontologies. Key Results Assembly of the 221 746 sequence reads resulted in 12 442 contigs. Over 50 % (6183) of 12 442 contigs showed significant homology to proteins in the NCBI database, representing approx. 4800 independent transcripts. The molecular transducer GO term was significantly over-represented in the native (South African) subtractive library compared with the invasive (Australian) library. Based on NCBI BLAST hits and literature searches, 40 % of the molecular transducer genes identified in the South African subtractive library are likely to be involved in response to biotic stimuli, such as fungal, bacterial and viral pathogens. Conclusions This EST collection is the first representation of the S. madagascariensis transcriptome and provides an important resource for the discovery of candidate genes associated with plant invasiveness. The over-representation of molecular transducer genes associated with defence responses in the native subtractive library provides preliminary support for aspects of the enemy release and evolution of increased competitive ability hypotheses in this successful invasive. This study highlights the contribution of next-generation sequencing to better understanding the molecular mechanisms underlying ecological hypotheses that are important in successful plant invasions.

Optimistic linear programming gives logarithmic regret for irreducible MDPs

We present an algorithm called Optimistic Linear Programming (OLP) for learning to optimize average reward in an irreducible but otherwise unknown Markov decision process (MDP). OLP uses its experience so far to estimate the MDP. It chooses actions by optimistically maximizing estimated future rewards over a set of next-state transition probabilities that are close to the estimates, a computation that corresponds to solving linear programs. We show that the total expected reward obtained by OLP up to time T is within C(P) log T of the reward obtained by the optimal policy, where C(P) is an explicit, MDP-dependent constant. OLP is closely related to an algorithm proposed by Burnetas and Katehakis with four key differences: OLP is simpler, it does not require knowledge of the supports of transition probabilities, the proof of the regret bound is simpler, but our regret bound is a constant factor larger than the regret of their algorithm. OLP is also similar in flavor to an algorithm recently proposed by Auer and Ortner. But OLP is simpler and its regret bound has a better dependence on the size of the MDP.

Parallel implementation of stochastic simulation for large scale cellular processes

Experimental and theoretical studies have shown the importance of stochastic processes in genetic regulatory networks and cellular processes. Cellular networks and genetic circuits often involve small numbers of key proteins such as transcriptional factors and signaling proteins. In recent years stochastic models have been used successfully for studying noise in biological pathways, and stochastic modelling of biological systems has become a very important research field in computational biology. One of the challenge problems in this field is the reduction of the huge computing time in stochastic simulations. Based on the system of the mitogen-activated protein kinase cascade that is activated by epidermal growth factor, this work give a parallel implementation by using OpenMP and parallelism across the simulation. Special attention is paid to the independence of the generated random numbers in parallel computing, that is a key criterion for the success of stochastic simulations. Numerical results indicate that parallel computers can be used as an efficient tool for simulating the dynamics of large-scale genetic regulatory networks and cellular processes

Investigation of methods for increasing the energy efficiency on unmanned aerial vehicles (UAVS)

In recent years, development of Unmanned Aerial Vehicles (UAV) has become a significant growing segment of the global aviation industry. These vehicles are developed with the intention of operating in regions where the presence of onboard human pilots is either too risky or unnecessary. Their popularity with both the military and civilian sectors have seen the use of UAVs in a diverse range of applications, from reconnaissance and surveillance tasks for the military, to civilian uses such as aid relief and monitoring tasks. Efficient energy utilisation on an UAV is essential to its functioning, often to achieve the operational goals of range, endurance and other specific mission requirements. Due to the limitations of the space available and the mass budget on the UAV, it is often a delicate balance between the onboard energy available (i.e. fuel) and achieving the operational goals. This thesis presents an investigation of methods for increasing the energy efficiency on UAVs. One method is via the development of a Mission Waypoint Optimisation (MWO) procedure for a small fixed-wing UAV, focusing on improving the onboard fuel economy. MWO deals with a pre-specified set of waypoints by modifying the given waypoints within certain limits to achieve its optimisation objectives of minimising/maximising specific parameters. A simulation model of a UAV was developed in the MATLAB Simulink environment, utilising the AeroSim Blockset and the in-built Aerosonde UAV block and its parameters. This simulation model was separately integrated with a multi-objective Evolutionary Algorithm (MOEA) optimiser and a Sequential Quadratic Programming (SQP) solver to perform single-objective and multi-objective optimisation procedures of a set of real-world waypoints in order to minimise the onboard fuel consumption. The results of both procedures show potential in reducing fuel consumption on a UAV in a ight mission. Additionally, a parallel Hybrid-Electric Propulsion System (HEPS) on a small fixedwing UAV incorporating an Ideal Operating Line (IOL) control strategy was developed. An IOL analysis of an Aerosonde engine was performed, and the most efficient (i.e. provides greatest torque output at the least fuel consumption) points of operation for this engine was determined. Simulation models of the components in a HEPS were designed and constructed in the MATLAB Simulink environment. It was demonstrated through simulation that an UAV with the current HEPS configuration was capable of achieving a fuel saving of 6.5%, compared to the ICE-only configuration. These components form the basis for the development of a complete simulation model of a Hybrid-Electric UAV (HEUAV).