Statistical Data Mining Approach Research Articles

Molecular Weight Controls Interactions between Plastic Deformation and Fracture in Cold Spray of Glassy Polymers.

Polymer cold spray has gained considerable attention as a novel manufacturing process. A promising aspect of this technology involves the ability to deposit uniform polymer coatings without the requirements of solvent and/or high-temperature conditions. The present study investigates the interplay between shear instability, often considered to be the primary mechanism for bond formation, and fracture, as a secondary energy dissipation mechanism, collectively governing the deposition of glassy thermoplastics on similar and dissimilar substrates. A hybrid experimental-computational approach is utilized to explore the simultaneous effects of several interconnected phenomena, namely the particle-substrate relative deformability, molecular weights, and the resultant yielding versus fracture of polystyrene particles, examined herein as a model material system. The computational investigations are based on constitutive plasticity and damage equations determined and calibrated based on a statistical data mining approach applied to a wide collection of previously reported stress-strain and failure data. Results obtained herein demonstrate that the underlying adhesion mechanisms depend strongly on the molecular weight of the sprayed particles. It is also shown that although the plastic deformation and shear instability are still the primary bond formation mechanisms, the molecular-weight-dependent fracture of the sprayed glassy polymers is also a considerable phenomenon capable of significantly affecting the deposition process, especially in cases involving the cold spray of soft thermoplastics on hard substrates. The strong interplay between molecular-weight-dependent plastic yielding and fracture in the examined system emphasizes the importance of molecular weight as a critical variable in the cold spray of glassy polymers, also highlighting the possibility of process optimization by proper feedstock selection.

Modeling of free swelling index based on variable importance measurements of parent coal properties by random forest method

Abstract Coke quality has a critical role in the steelmaking industry. The aim of this study is to examine the complex relationships between various conventional coal analyses using coke making index “free swelling index (FSI)”. Random forest (RF) associated with variable importance measurements (VIMs), which is a new powerful statistical data mining approach, is utilized in this study to analyze a high-dimensional database (3961 samples) to rank variables, and to develop an accurate FSI predictive model based on the most important variables. VIMs was performed on various types of analyses which indicated that volatile matter, carbon, moisture (coal rank parameters) and organic sulfur are the most effective coal properties for the prediction of FSI. These variables have been used as an input set of RF model for the FSI modeling and prediction. Results of FSI model indicated that RF can provide a satisfactory prediction of FSI with the correlation of determination R 2 = 0.96 and mean square error of 0.16 from laboratory FSIs (which is smaller than the interval unit of FSI; 0.5). Based on this result, RF can be used to rank and select effective variables by evaluating nonlinear relationships among parameters. Moreover, it can be further employed as a non-parametric reliable predictive method for modeling, controlling, and optimizing complex variables; which to our knowledge has never been utilized in the fuel and energy sectors.

A Multi-Agent System for Reporting Suspected Adverse Drug Reactions

Adverse drug reactions (ADRs) represent a serious worldwide public health problem. Current post-marketing ADR detection approaches largely rely on spontaneous reports filed by various healthcare professionals such as physicians, pharmacists et.al.. Underreporting is a serious deficiency of these methods - the actually reported adverse events represent less than 10% of all cases. Studies show that two important reasons that cause the underreporting are: 1) healthcare professionals are unaware of encountered ADRs, especially for those unusual ADRs; 2) they are too busy to voluntarily report ADRs since it takes a lot of time to fill out the reporting forms. This paper addresses these two issues by developing a multi-agent ADR reporting system. The system 1) helps healthcare professionals detect the potential causal relationship between a drug and an ADR by analyzing patients’ electronic records via both case-based analysis and statistical data mining approach; 2) allows healthcare professionals to add new rules to signal potential ADRs based on their medical expertise; 3) makes the reporting much easier by automatically linking the patients’ electronic data with the reporting form. A functioning prototype of the system has been developed. The proposed data analysis approaches as well as the performance of the system have been tested. The results indicate that this system has a great potential to improve the spontaneous reporting rate of suspected adverse drug reactions.

A statistical data mining approach in bacteriology for bacterial identification

Statistical data mining is one of the popular research fields of exploring valuable information from the large number of collected data. Applying statistical data mining techniques in several fields like medicine, bioinformatics, business, and bacteriology can be beneficial. Among them, bacteriology is one of the promising fields where statistical data mining is hardly ever used. The main purpose of this paper is to demonstrate the contribution of statistical data mining in the field of bacteriology. The research problem named as bacterial identification from bacteriology that we handle in this paper is a special kind of classification problem of statistical data mining where the only single representation of every class is present in the dataset. After studying this research problem, this paper proposes a novel statistical data mining approach using the decision tree technique in bacterial identification with better performance. The experimental results show significantly less number of biochemical tests are needed in bacterial identification using this proposed approach than the conventional approach that is being followed currently in the biochemical laboratory. Thus, the proposed approach not only benefits microbiologists, but it also improves the traditional approach of bacterial identification by saving time, total cost, and manual labour involvements.

Applying hybrid reasoning to mine for associative features in biological data

We develop the means to mine for associative features in biological data. The hybrid reasoning schema for deterministic machine learning and its implementation via logic programming is presented. The methodology of mining for correlation between features is illustrated by the prediction tasks for protein secondary structure and phylogenetic profiles. The suggested methodology leads to a clearer approach to hierarchical classification of proteins and a novel way to represent evolutionary relationships. Comparative analysis of Jasmine and other statistical and deterministic systems (including Explanation-Based Learning and Inductive Logic Programming) are outlined. Advantages of using deterministic versus statistical data mining approaches for high-level exploration of correlation structure are analyzed.

Objective Speech Quality Measurement Using Statistical Data Mining

Measuring speech quality by machines overcomes two major drawbacks of subjective listening tests, their low speed and high cost. Real-time, accurate, and economical objective measurement of speech quality opens up a wide range of applications that cannot be supported with subjective listening tests. In this paper, we propose a statistical data mining approach to design objective speech quality measurement algorithms. A large pool of perceptual distortion features is extracted from the speech signal. We examine using classification and regression trees (CART) and multivariate adaptive regression splines (MARS), separately and jointly, to select the most salient features from the pool, and to construct good estimators of subjective listening quality based on the selected features. We show designs that use perceptually significant features and outperform the state-of-the-art objective measurement algorithm. The designed algorithms are computationally simple, making them suitable for real-time implementation. The proposed design method is scalable with the amount of learning data; thus, performance can be improved with more offline or online training.