PSOM<Superscript>2</Superscript>—partitioning-based scalable ontology matching using MapReduce

The growth and use of semantic web has led to a drastic increase in the size, heterogeneity and number of ontologies that are available on the web. Correspondingly, scalable ontology matching algorithms that will eliminate the heterogeneity among large ontologies have become a necessity. Ontology matching algorithms generally do not scale well due to the massive number of complex computations required to achieve matching. One of the methods used to address this problem is the use of partition-based systems to reduce the matching space. In this paper, we propose a new partitioning-based scalable ontology matching system called PSOM². We have designed a new neighbour-based intra-similarity measure to increase the quality of the cluster set formation for the partition-based ontology matching process. These sets of clusters or sub-ontologies are matched across the input ontologies to identify matchable cluster pairs, based on anchors that are efficiently discovered through a new light-weight linguistic matcher (EI-sub). However, in order to further increase the efficiency of the time-consuming anchor discovery process we have designed a MapReduce-based EI-sub process where anchors are discovered in distributed and parallel fashion. Experiments on benchmark OAEI (Ontology Alignment Evaluation Initiative) large scale ontologies demonstrate that the new PSOM² system achieves, on an average, 31% decrease in entropy of the clusters and 54.5% reduction in overall run time. Based on the experimental results, it is evident that the new PSOM² achieves better quality clusters and a major reduction in execution time, leading to an effective and scalable ontology matching system.     

Parameter optimization of friction stir welding of cryorolled AA2219 alloy using artificial neural network modeling with genetic algorithm

In this paper, parameter optimization of FSW of cryorolled AA2219 alloy was carried out to obtain defect free weld joint with maximum weld strength. To achieve this, artificial neural network (ANN) was used to model the relationship between the input parameters and the mechanical and corrosion properties (output) of the weld joints. The optimal FSW parameters were determined by genetic algorithm (GA). The feasible solution of the GA was tool rotational speed of 1005 rpm, tool travel speed of 20 mm/min and tool tilt angle of 3°. The feasible parameter was used to weld and check the ability of the parameter to produce better weld joint than the L₉ orthogonal array parameters. The weld, subjected to the confirmation test, was investigated by means of metallurgical, mechanical, and corrosion testing. This process reduces the costs associated with trial runs to obtain optimal parameters and also the production cost of the cryorolled (CR) plate which is high.     

Implication of unequal rivet load distribution in the failures and damage tolerant design of metal and composite civil aircraft riveted lap joints

Riveted lap joints are being used widely in civil aircraft structures. Conventional design procedures assume that the joint can be designed as if all rivets carry load equally. As found in literature associated with fatigue and fracture, forensic studies on structural failures, this assumption is not entirely valid. In this paper, the regulatory codes for civil aircraft as applicable to riveted joints in the form of FAR 25 regulations are briefly reviewed. The regulatory code discusses safety factors in an implied way, but has little specific recommendations for riveted joints. However, studies on the failures of specific aircraft illustrated in this paper add to the argument that both static strength and life are affected by the initial design procedures for the riveted joints. In this paper, finite element models for metal–metal, composite–metal, composite–composite lap joints are studied. A three row lap joint used in commercial aircraft and which was part of failure studies is also examined. Unequal rivet loads and in cases, nearly 50% more than conventional design has been seen in linear finite element analysis. Elasto-plastic analysis using rivet flexibility shows re-distribution of loads. Based on these observations, the effect of rivet loads on life estimation including the use of concepts such as by-pass stresses is discussed. These results have implications for static strength at ultimate load, damage tolerance and fail safety and are discussed in this paper. Next, in a composite–composite lap joint, the influence of ply-angle on the rivet loads is studied. Also, a composite–metal lap joint is studied for the rivet load distribution and life estimation. It is found that the load shared by the rivet rows in a composite–metal lap joint are not symmetric and therefore are more susceptible to cracking and subsequent failure as the unequal distribution can cause some of the rivet loads to be high. In conclusion, the issue of fail safe and damage tolerant design of civil aircraft structures with riveted joints are addressed, especially the implication of unequal load distribution on the failures of such joints and it is suggested that these unequal rivet load distributions be catered for at the early design stage itself via finite element analysis and the possibility of an over-arching safety factor could be considered that incorporates both ultimate load and damage tolerance conditions.     

Biopreservative technologies of food: an alternative to chemical preservation and recent developments

Food Science and Biotechnology - Despite centuries of developing strategies to prevent food-associated illnesses, food safety remains a significant concern, even with multiple technological...     

Ensemble Learning Based Collaborative Filtering with Instance Selection and Enhanced Clustering

Recommender system is a tool to suggest items to the users from the extensive history of the user's feedback. Though, it is an emerging research area concerning academics and industries, where it suffers from sparsity, scalability, and cold start problems. This paper addresses sparsity, and scalability problems of model-based collaborative recommender system based on ensemble learning approach and enhanced clustering algorithm for movie recommendations. In this paper, an effective movie recommendation system is proposed by Classification and Regression Tree (CART) algorithm, enhanced Balanced Iterative Reducing and Clustering using Hierarchies (BIRCH) algorithm and truncation method. In this research paper, a new hyper parameters tuning is added in BIRCH algorithm to enhance the cluster formation process, where the proposed algorithm is named as enhanced BIRCH. The proposed model yields quality movie recommendation to the new user using Gradient boost classification with broad coverage. In this paper, the proposed model is tested on Movielens dataset, and the performance is evaluated by means of Mean Absolute Error (MAE), precision, recall and f-measure. The experimental results showed the superiority of proposed model in movie recommendation compared to the existing models. The proposed model obtained 0.52 and 0.57 MAE value on Movielens 100k and 1M datasets. Further, the proposed model obtained 0.83 of precision, 0.86 of recall and 0.86 of f-measure on Movielens 100k dataset, which are effective compared to the existing models in movie recommendation.     

Optimization of process parameters of friction welding of UNS S31803 duplex stainless steels joints

This article presents an approach based on Taguchi method and Grey relational analysis to optimize process parameters of friction welding of UNS31803 duplex stainless steel. The main objective is to maximize mechanical properties like tensile strength, hardness and impact toughness and to minimize corrosion rate. Heating pressure, heating time, upsetting pressure and upsetting time were the four process parameters taken each at three levels. According to Taguchi quality design concept, an L9 orthogonal array was selected for experiments. The best combination of process parameters was found by both Taguchi method and Grey relational analysis. The influence of the process parameters on overall quality characteristics of the friction welding process was evaluated by the analysis of variance (ANOVA) method. The confirmation test results with optimal parameters confirmed the effectiveness of the proposed method in this study. Later, comparison was done between Taguchi method and Grey relational analysis on the basis of improvement in multiresponse signal to noise (S/N) ratio over initial process parameters. Grey relational analysis was proved to be a better technique than Taguchi method for optimization of multiple responses.     

Modelling the correlation between the geometrical features and the forming limit strains of perforated Al 8011 sheets using artificial neural network

In the recent years, sheet metals are produced with perforations in various shapes and patterns to improve the appearance of sheet and to save weight of components. As in conventional metal sheets, it is important to form the perforated sheet metals also within their safe strain regions to avoid the forming failures like necking, fracture and wrinkling. The Forming Limit Diagram (FLD) is an appropriate tool to determine the forming limit strains. The limiting strains of perforated sheet metals mainly depend on the geometry of the perforations and forming variables. This leads to large increase in number of test to be conducted with various geometry and forming variables for determining the forming limit strain for perforated sheets. Aiming to reduce the number of experiments needed, in this work, an Artificial Neural Network (ANN) model has been developed for forming limit diagram of perforated Al 8011 sheets based on experimental results and correlated with the geometrical features of the perforated sheets. This model is a feed forward back propagation neural network (BPNN) with a set of geometrical variables as its inputs and the safe true strains as its output. This ANN model can be applied for prediction of FLD of perforated sheet having any geometry.     

Asymptotic behaviors of support vector machines with Gaussian kernel

Support vector machines (SVMs) with the gaussian (RBF) kernel have been popular for practical use. Model selection in this class of SVMs involves two hyperparameters: the penalty parameter C and the kernel width sigma. This letter analyzes the behavior of the SVM classifier when these hyperparameters take very small or very large values. Our results help in understanding the hyperparameter space that leads to an efficient heuristic method of searching for hyperparameter values with small generalization errors. The analysis also indicates that if complete model selection using the gaussian kernel has been conducted, there is no need to consider linear SVM.