Statistical Machine Learning Techniques for the Prediction of Learning Disabilities in School-Age Children

Learning Disability (LD) is a general term that describes specific kinds of learning problems. It is a neurological condition that affects a child's brain and impairs his ability to carry out one or many specific tasks. The learning disabled children are neither slow nor mentally retarded. This disorder can make it problematic for a child to learn as quickly or in the same way as some child who isn't affected by a learning disability. An affected child can have normal or above average intelligence. They may have difficulty paying attention, with reading or letter recognition, or with mathematics. It does not mean that children who have learning disabilities are less intelligent. In fact, many children who have learning disabilities are more intelligent than an average child. Learning disabilities vary from child to child. One child with LD may not have the same kind of learning problems as another child with LD. There is no cure for learning disabilities and they are life-long. However, children with LD can be high achievers and can be taught ways to get around the learning disability. In this research work, data mining using machine learning techniques are used to analyze the symptoms of LD, establish interrelationships between them and evaluate the relative importance of these symptoms. To increase the diagnostic accuracy of learning disability prediction, a knowledge based tool based on statistical machine learning or data mining techniques, with high accuracy,according to the knowledge obtained from the clinical information, is proposed. The basic idea of the developed knowledge based tool is to increase the accuracy of the learning disability assessment and reduce the time used for the same. Different statistical machine learning techniques in data mining are used in the study. Identifying the important parameters of LD prediction using the data mining techniques, identifying the hidden relationship between the symptoms of LD and estimating the relative significance of each symptoms of LD are also the parts of the objectives of this research work. The developed tool has many advantages compared to the traditional methods of using check lists in determination of learning disabilities. For improving the performance of various classifiers, we developed some preprocessing methods for the LD prediction system. A new system based on fuzzy and rough set models are also developed for LD prediction. Here also the importance of pre-processing is studied. A Graphical User Interface (GUI) is designed for developing an integrated knowledge based tool for prediction of LD as well as its degree. The designed tool stores the details of the children in the student database and retrieves their LD report as and when required. The present study undoubtedly proves the effectiveness of the tool developed based on various machine learning techniques. It also identifies the important parameters of LD and accurately predicts the learning disability in school age children. This thesis makes several major contributions in technical, general and social areas. The results are found very beneficial to the parents, teachers and the institutions. They are able to diagnose the child’s problem at an early stage and can go for the proper treatments/counseling at the correct time so as to avoid the academic and social losses.

Design and Performance Evaluation of EXMAN: An EXtended MANchester Data Flow Computer

Data flow computers are high-speed machines in which an instruction is executed as soon as all its operands are available. This paper describes the EXtended MANchester (EXMAN) data flow computer which incorporates three major extensions to the basic Manchester machine. As extensions we provide a multiple matching units scheme, an efficient, implementation of array data structure, and a facility to concurrently execute reentrant routines. A simulator for the EXMAN computer has been coded in the discrete event simulation language, SIMULA 67, on the DEC 1090 system. Performance analysis studies have been conducted on the simulated EXMAN computer to study the effectiveness of the proposed extensions. The performance experiments have been carried out using three sample problems: matrix multiplication, Bresenham's line drawing algorithm, and the polygon scan-conversion algorithm.

Fuzzy data envelopment analysis for performance evaluation of an irrigation system

Sixteen irrigation subsystems of the Mahi Bajaj Sagar Project, Rajasthan, India, are evaluated and selection of the most suitable/best is made using data envelopment analysis (DEA) in both deterministic and fuzzy environments. Seven performance-related indicators, namely, land development works (LDW), timely supply of inputs (TSI), conjunctive use of water resources (CUW), participation of farmers (PF), environmental conservation (EC), economic impact (EI) and crop productivity (CPR) are considered. Of the seven, LDW, TSI, CUW, PF and EC are considered inputs, whereas CPR and EI are considered outputs for DEA modelling purposes. Spearman rank correlation coefficient values are also computed for various scenarios. It is concluded that DEA in both deterministic and fuzzy environments is useful for the present problem. However, the outcome of fuzzy DEA may be explored for further analysis due to its simple, effective data and discrimination handling procedure. It is inferred that the present study can be explored for similar situations with suitable modifications.

Inter-comparison and performance evaluation of chemistry transport models over Indian region

Aerosol loading over the South Asian region has the potential to affect the monsoon rainfall, Himalayan glaciers and regional air-quality, with implications for the billions in this region. While field campaigns and network observations provide primary data, they tend to be location/season specific. Numerical models are useful to regionalize such location-specific data. Studies have shown that numerical models underestimate the aerosol scenario over the Indian region, mainly due to shortcomings related to meteorology and the emission inventories used. In this context, we have evaluated the performance of two such chemistry-transport models: WRF-Chem and SPRINTARS over an India-centric domain. The models differ in many aspects including physical domain, horizontal resolution, meteorological forcing and so on etc. Despite these differences, both the models simulated similar spatial patterns of Black Carbon (BC) mass concentration, (with a spatial correlation of 0.9 with each other), and a reasonable estimates of its concentration, though both of them under-estimated vis-a-vis the observations. While the emissions are lower (higher) in SPRINTARS (WRF-Chem), overestimation of wind parameters in WRF-Chem caused the concentration to be similar in both models. Additionally, we quantified the under-estimations of anthropogenic BC emissions in the inventories used these two models and three other widely used emission inventories. Our analysis indicates that all these emission inventories underestimate the emissions of BC over India by a factor that ranges from 1.5 to 2.9. We have also studied the model simulations of aerosol optical depth over the Indian region. The models differ significantly in simulations of AOD, with WRF-Chem having a better agreement with satellite observations of AOD as far as the spatial pattern is concerned. It is important to note that in addition to BC, dust can also contribute significantly to AOD. The models differ in simulations of the spatial pattern of mineral dust over the Indian region. We find that both meteorological forcing and emission formulation contribute to these differences. Since AOD is column integrated parameter, description of vertical profiles in both models, especially since elevated aerosol layers are often observed over Indian region, could be also a contributing factor. Additionally, differences in the prescription of the optical properties of BC between the models appear to affect the AOD simulations. We also compared simulation of sea-salt concentration in the two models and found that WRF-Chem underestimated its concentration vis-a-vis SPRINTARS. The differences in near-surface oceanic wind speeds appear to be the main source of this difference. In-spite of these differences, we note that there are similarities in their simulation of spatial patterns of various aerosol species (with each other and with observations) and hence models could be valuable tools for aerosol-related studies over the Indian region. Better estimation of emission inventories could improve aerosol-related simulations. (C) 2015 Elsevier Ltd. All rights reserved.

Development of a model-based transient calibration process for diesel engine electronic control module tables – Part 1: data requirements, processing, and analysis

This is the first part of a study investigating a model-based transient calibration process for diesel engines. The motivation is to populate hundreds of parameters (which can be calibrated) in a methodical and optimum manner by using model-based optimization in conjunction with the manual process so that, relative to the manual process used by itself, a significant improvement in transient emissions and fuel consumption and a sizable reduction in calibration time and test cell requirements is achieved. Empirical transient modelling and optimization has been addressed in the second part of this work, while the required data for model training and generalization are the focus of the current work. Transient and steady-state data from a turbocharged multicylinder diesel engine have been examined from a model training perspective. A single-cylinder engine with external air-handling has been used to expand the steady-state data to encompass transient parameter space. Based on comparative model performance and differences in the non-parametric space, primarily driven by a high engine difference between exhaust and intake manifold pressures (ΔP) during transients, it has been recommended that transient emission models should be trained with transient training data. It has been shown that electronic control module (ECM) estimates of transient charge flow and the exhaust gas recirculation (EGR) fraction cannot be accurate at the high engine ΔP frequently encountered during transient operation, and that such estimates do not account for cylinder-to-cylinder variation. The effects of high engine ΔP must therefore be incorporated empirically by using transient data generated from a spectrum of transient calibrations. Specific recommendations on how to choose such calibrations, how many data to acquire, and how to specify transient segments for data acquisition have been made. Methods to process transient data to account for transport delays and sensor lags have been developed. The processed data have then been visualized using statistical means to understand transient emission formation. Two modes of transient opacity formation have been observed and described. The first mode is driven by high engine ΔP and low fresh air flowrates, while the second mode is driven by high engine ΔP and high EGR flowrates. The EGR fraction is inaccurately estimated at both modes, while EGR distribution has been shown to be present but unaccounted for by the ECM. The two modes and associated phenomena are essential to understanding why transient emission models are calibration dependent and furthermore how to choose training data that will result in good model generalization.

Customer information system data pre-processing with feature selection techniques for non-technical losses prediction in an electricity market

Non-technical losses (NTL) identification and prediction are important tasks for many utilities. Data from customer information system (CIS) can be used for NTL analysis. However, in order to accurately and efficiently perform NTL analysis, the original data from CIS need to be pre-processed before any detailed NTL analysis can be carried out. In this paper, we propose a feature selection based method for CIS data pre-processing in order to extract the most relevant information for further analysis such as clustering and classifications. By removing irrelevant and redundant features, feature selection is an essential step in data mining process in finding optimal subset of features to improve the quality of result by giving faster time processing, higher accuracy and simpler results with fewer features. Detailed feature selection analysis is presented in the paper. Both time-domain and load shape data are compared based on the accuracy, consistency and statistical dependencies between features.

Pre-formulation and systematic evaluation of amino acid assisted permeability of insulin across in vitro buccal cell layers

The aim of this work was to investigate alternative safe and effective permeation enhancers for buccal peptide delivery. Basic amino acids improved insulin solubility in water while 200 and 400 µg/mL lysine significantly increased insulin solubility in HBSS. Permeability data showed a significant improvement in insulin permeation especially for 10 µg/mL of lysine (p < 0.05) and 10 µg/mL histidine (p < 0.001), 100 µg/mL of glutamic acid (p < 0.05) and 200 µg/mL of glutamic acid and aspartic acid (p < 0.001) without affecting cell integrity; in contrast to sodium deoxycholate which enhanced insulin permeability but was toxic to the cells. It was hypothesized that both amino acids and insulin were ionised at buccal cavity pH and able to form stable ion pairs which penetrated the cells as one entity; while possibly triggering amino acid nutrient transporters on cell surfaces. Evidence of these transport mechanisms was seen with reduction of insulin transport at suboptimal temperatures as well as with basal-to-apical vectoral transport, and confocal imaging of transcellular insulin transport. These results obtained for insulin is the first indication of a possible amino acid mediated transport of insulin via formation of insulin-amino acid neutral complexes by the ion pairing mechanism.

Modelling and performance evaluation of the IEEE 802.11 DCF for real-time control

Popular wireless networks, such as IEEE 802.11/15/16, are not designed for real-time applications. Thus, supporting real-time quality of service (QoS) in wireless real-time control is challenging. This paper adopts the widely used IEEE 802.11, with the focus on its distributed coordination function (DCF), for soft-real-time control systems. The concept of the critical real-time traffic condition is introduced to characterize the marginal satisfaction of real-time requirements. Then, mathematical models are developed to describe the dynamics of DCF based real-time control networks with periodic traffic, a unique feature of control systems. Performance indices such as throughput and packet delay are evaluated using the developed models, particularly under the critical real-time traffic condition. Finally, the proposed modelling is applied to traffic rate control for cross-layer networked control system design.

Design and performance evaluation of a systolic architecture for hidden-surface removal

In this paper, we propose a systolic architecture for hidden-surface removal. Systolic architecture is a kind of parallel architecture best known for its easy VLSI implementability. After discussing the design details of the architecture, we present the results of the simulation experiments conducted in order to evaluate the performance of the architecture.

Experimental characterisation and performance evaluation of a coaxial current transformer for measurement of insulated gate bipolar transistor switching current

Measurement of device current during switching characterisation of an insulated gate bipolar transistor (IGBT) requires a current sensor with low insertion impedance and high bandwidth. This study presents an experimental procedure for evaluating the performance of a coaxial current transformer (CCT), designed for the above purpose. A prototype CCT, which can be mounted directly on a power terminal of a 1200 V/50 A half-bridge IGBT module, is characterised experimentally. The measured characteristics include insertion impedance, gain and phase of the CCT at different frequencies. The bounds of linearity within which the CCT can operate without saturation are determined theoretically, and are also verified experimentally. The experimental study on linearity of the CCT requires a high-amplitude current source. A proportional-resonant (PR) controller-based current-controlled half-bridge inverter is developed for this purpose. A systematic procedure for selection of PR controller parameters is also reported in this study. This set-up is helpful to determine the limit of linearity and also to measure the frequency response of the CCT at realistic amplitudes of current in the low-frequency range.