Simple Machine Learning Algorithms Research Articles

A machine learning approach to detect changes in gait parameters following a fatiguing occupational task

The purpose of this study is to provide a method for classifying non-fatigued vs. fatigued states following manual material handling. A method of template matching pattern recognition for feature extraction ($1 Recognizer) along with the support vector machine model for classification were applied on the kinematics of gait cycles segmented by our stepwise search-based segmentation algorithm. A single inertial measurement unit on the ankle was used, providing a minimally intrusive and inexpensive tool for monitoring. The classifier distinguished between states using distance-based scores from the recogniser and the step duration. The results of fatigue detection showed an accuracy of 90% across data from 20 recruited subjects. This method utilises the minimum amount of data and features from only one low-cost sensor to reliably classify the state of fatigue induced by a realistic manufacturing task using a simple machine learning algorithm that can be extended to real-time fatigue monitoring as a future technology to be employed in the manufacturing facilities.Practitioner Summary: We examined the use of a wearable sensor for the detection of fatigue-related changes in gait based on a simulated manual material handling task. Classification based on foot acceleration and position trajectories resulted in 90% accuracy. This method provides a practical framework for predicting realistic levels of fatigue.

Information Token Driven Machine Learning For Electronic Markets: Performance Effects In Behavioral Financial Big Data Analytics.

Conjunct with the universal acceleration in information growth, financial services have been immersed in an evolution of information dynamics. It is not just the dramatic increase in volumes of data, but the speed, the complexity and the unpredictability of big-data phenomena that have compounded the challenges faced by researchers and practitioners in financial services. Math, statistics and technology have been leveraged creatively to create analytical solutions. Given the many unique characteristics of financial bid data (FBD) it is necessary to gain insights into strategies and models that can be used to create FBD specific solutions. Behavioral finance data, a subset of FBD, is seeing exponential growth and this presents an unprecedented opportunity to study behavioral finance employing big data analytics methodologies. The present study maps machine learning (ML) techniques and behavioral finance categories to explore the potential for using ML techniques to address behavioral aspects in FBD. The ontological feasibility of such an approach is presented and the primary purpose of this study is propositioned- ML based behavioral models can effectively estimate performance in FBD. A simple machine learning algorithm is successfully employed to study behavioral performance in an artificial stock market to validate the propositions. Keywords: Information; Big Data; Electronic Markets; Analytics; Behavior

Direct tomography imaging for inelastic X-ray scattering experiments at high pressure.

A method to separate the non-resonant inelastic X-ray scattering signal of a micro-metric sample contained inside a diamond anvil cell (DAC) from the signal originating from the high-pressure sample environment is described. Especially for high-pressure experiments, the parasitic signal originating from the diamond anvils, the gasket and/or the pressure medium can easily obscure the sample signal or even render the experiment impossible. Another severe complication for high-pressure non-resonant inelastic X-ray measurements, such as X-ray Raman scattering spectroscopy, can be the proximity of the desired sample edge energy to an absorption edge energy of elements constituting the DAC. It is shown that recording the scattered signal in a spatially resolved manner allows these problems to be overcome by separating the sample signal from the spurious scattering of the DAC without constraints on the solid angle of detection. Furthermore, simple machine learning algorithms facilitate finding the corresponding detector pixels that record the sample signal. The outlined experimental technique and data analysis approach are demonstrated by presenting spectra of the Si L2,3-edge and O K-edge of compressed α-quartz. The spectra are of unprecedented quality and both the O K-edge and the Si L2,3-edge clearly show the existence of a pressure-induced phase transition between 10 and 24 GPa.

QSAR Modeling of Tox21 Challenge Stress Response and Nuclear Receptor Signaling Toxicity Assays

The ability to determine which environmental chemicals pose the greatest potential threats to human health remains one of the major concerns in regulatory toxicology. Computation methods that can accurately predict the chemicals’ toxic potential in silico are increasingly sought-after to replace in vitro high-throughput screening (HTS) as well as controversial and costly in vivo animal studies. To this end, we have built Quantitative Structure-Activity Relationship (QSAR) models of twelve (12) stress response and nuclear receptor signaling pathways toxicity assays as part of the 2014 Tox21 Challenge. Our models were built using the Random Forest, Deep Neural Networks and various combinations of descriptors and balancing protocols. All of our models were statistically significant for each of the 12 assays with the balanced accuracy in the range between 0.58 and 0.82. Our results also show that models built with Deep Neural Networks had high accuracy than those developed with simple machine learning algorithms and that dataset balancing led to a significant accuracy decrease.

Regression and classification using extreme learning machine based on L1-norm and L2-norm

Extreme learning machine (ELM) is a very simple machine learning algorithm and it can achieve a good generalization performance with extremely fast speed. Therefore it has practical significance for data analysis in real-world applications. However, it is implemented normally under the empirical risk minimization scheme and it may tend to generate a large-scale and over-fitting model. In this paper, an ELM model based on L1-norm and L2-norm regularizations is proposed to handle regression and multiple-class classification problems in a unified framework. The proposed model called L1–L2-ELM combines the grouping effect benefits of L2 penalty and the tendency towards sparse solution of L1 penalty, thus it can control the complexity of the network and prevent over-fitting. To solve the mixed penalty problem, the separate elastic net algorithm and Bayesian information criterion (BIC) are adopted to find the optimal model for each response variable. We test the L1–L2-ELM algorithm on one artificial case and nine benchmark data sets to evaluate its performance. Simulation results have shown that the proposed algorithms outperform the original ELM as well as other advanced ELM algorithms in terms of prediction accuracy, and it is more robust in both regression and classification applications.

Assessing Intervention Timing in Computer-Based Education Using Machine Learning Algorithms

The use of computer-based and online education systems has made new data available that can describe the temporal and process-level progression of learning. To date, machine learning research has not considered the impacts of these properties on the machine learning prediction task in educational settings. Machine learning algorithms may have applications in supporting targeted intervention approaches. The goals of this paper are to: 1) determine the impact of process-level information on machine learning prediction results and 2) establish the effect of type of machine learning algorithm used on prediction results. Data were collected from a university level course in human factors engineering (n=35), which included both traditional classroom assessment and computer-based assessment methods. A set of common regression and classification algorithms were applied to the data to predict final course score. The overall prediction accuracy as well as the chronological progression of prediction accuracy was analyzed for each algorithm. Simple machine learning algorithms (linear regression, logistic regression) had comparable performance with more complex methods (support vector machines, artificial neural networks). Process-level information was not useful in post-hoc predictions, but contributed significantly to allowing for accurate predictions to be made earlier in the course. Process level information provides useful prediction features for development of targeted intervention techniques, as it allows more accurate predictions to be made earlier in the course. For small course data sets, the prediction accuracy and simplicity of linear regression and logistic regression make these methods preferable to more complex algorithms.

MARTT: Automatic Markup of Taxonomic Descriptions with XML

Despite the sub-language nature of taxonomic descriptions of animals and plants, researchers have warned about the existence of large variations among different description collections in terms of information content and its representation. These variations impose a serious threat to the development of automatic tools to structure large volumes of text-based descriptions. This paper presents a general approach to mark up different collections of taxonomic descriptions with XML, using two large-scale floras as examples. The markup system, MARTT, is based on machine learning methods and enhanced by machine learned domain rules and conventions. Experiments show that our simple and efficient machine learning algorithms outperform significantly general purpose algorithms and that rules learned from one flora can be used when marking up a second flora and help to improve the markup performance, especially for elements that have sparse training examples.Malgré la nature de sous-langage des descriptions taxinomiques des animaux et des plantes, les chercheurs reconnaissent l’existence de vastes variations parmi différentes collections de descriptions, en termes de contenu informationnel et de leur représentation. Ces variations présentent une menace sérieuse pour le développement d’outils automatiques pour la structuration de larges… 

Ecosystem mapping at the African continent scale using a hybrid clustering approach based on 1-km resolution multi-annual data from SPOT/VEGETATION

The goal of this study is to propose a new classification of African ecosystems based on an 8-year analysis of Normalized Difference Vegetation Index (NDVI) data sets from SPOT/VEGETATION. We develop two methods of classification. The first method is obtained from a k-nearest neighbour (k-NN) classifier, which represents a simple machine learning algorithm in pattern recognition. The second method is hybrid in that it combines k-NN clustering, hierarchical principles and the Fast Fourier Transform (FFT). The nomenclature of the two classifications relies on three levels of vegetation structural categories based on the Land Cover Classification System (LCCS). The two main outcomes are: (i) The delineation of the spatial distribution of ecosystems into five bioclimatic ecoregions at the African continental scale; (ii) Two ecosystem maps were made sequentially: an initial map with 92 ecosystems from the k-NN, plus a deduced hybrid classification with 73 classes, which better reflects the bio-geographical patterns. The inclusion of bioclimatic information and successive k-NN clustering elements helps to enhance the discrimination of ecosystems. Adopting this hybrid approach makes the ecosystem identification and labelling more flexible and more accurate in comparison to straightforward methods of classification. The validation of the hybrid classification, conducted by crossing-comparisons with validated continental maps, displayed a mapping accuracy of 54% to 61%.

How to build the cognitive network?

Event Abstract Back to Event How to build the cognitive network? Plewczynski Dariusz1* 1 University of Warsaw, Interdisciplinary Centre for Mathematical and Comp, Poland How to build the cognitive network? What is the influence of the learning network on the final quality of information processing in cognitive systems? How a node of such network has to be designed in order to effectively process incoming information that is both time dependent and spatially heterogenous? We address those questions by benchmarking different machine learning algorithms using real life examples (mostly from bioinformatics and chemoinformatics) that mimics the different pattern recognition tasks typically used in various machine learning applications. In the framework of Agent Based Modelling (ABM) the most promising cognitive network appears to be constructed from an sparse ensemble of interacting agents. The cognitive networks are able to successfully combine outcomes of individual learning agents in order to determine the final decision with the higher accuracy than any single learning method used in the construction of a network. We discuss here also the mean-field approximation of the cognitive networks in the limit of the infinite number of learning agents. The ensemble of interacting learning agents, acquires and process incoming information using various types, or different versions of machine learning and clustering algorithms. The abstract learning space, where all agents are located, is constructed here using a randomly connected, sparse model. It couples locally the certain small percentage of similar learning agents, yet also connect remotely about 10% of the other agents with random strength values. Such network simulates the higher level integration of information acquired from the independent learning trials. The final classification of incoming input data is therefore defined as the stationary state of the cognitive system using simple majority rule, yet the minority clusters that share opposite classification outcome can be observed in the system. The cognitive network is able to couple different scales of both space and time patterns by assigning them to different subsets of learning nodes. The probability of selecting proper class for a given input data, can be estimated even without the prior knowledge of its affiliation. The fuzzy logic can be easily introduced into the system, even if learning agents are build from simple binary classification machine learning algorithms by calculating the percentage of agreeing agents. The model is inspired by the large scale cognitive computing models of the brain cortex areas (D. Modha et al., Izhikevich et al.).

Mean-field theory of meta-learning

We discuss here the mean-field theory for a cellular automata model of meta-learning.Meta-learning is the process of combining outcomes of individual learning procedures inorder to determine the final decision with higher accuracy than any single learning method.Our method is constructed from an ensemble of interacting, learning agents that acquireand process incoming information using various types, or different versions, of machinelearning algorithms. The abstract learning space, where all agents are located,is constructed here using a fully connected model that couples all agents withrandom strength values. The cellular automata network simulates the higher levelintegration of information acquired from the independent learning trials. The finalclassification of incoming input data is therefore defined as the stationary state of themeta-learning system using simple majority rule, yet the minority clusters that share theopposite classification outcome can be observed in the system. Therefore, theprobability of selecting a proper class for a given input data, can be estimatedeven without the prior knowledge of its affiliation. The fuzzy logic can be easilyintroduced into the system, even if learning agents are built from simple binaryclassification machine learning algorithms by calculating the percentage of agreeingagents.