Abstract: The goal of motor image brain computer interface research is to create a link between the central nervous system and a computer or device. The most important signal for brain-computer interface is the electroencephalogram. The aim of this research is to explore a set of effective features from EEG signals, separated into frequency bands, using wave atom sub-bands to discriminate right and left-hand motor imagery signals. Over the transform coefficients, feature vectors are constructed for each frequency range and each transform sub-band, and their classification performances are tested. The method is validated using EEG signals from the BCI competition III dataset IIIa and classifiers such as support vector machine and k-nearest neighbors.
Abstract: Aflatoxins are highly poisonous and carcinogenic compounds produced by species of the genus Aspergillus spp. that can infect a variety of agricultural foods, including dried figs. Biological and environmental factors, such as population, pathogenicity and aflatoxinogenic capacity of the strains, topography, soil and climate parameters of the fig orchards are believed to have a strong effect on aflatoxin levels. Existing methods for aflatoxin detection and measurement, such as high-performance liquid chromatography (HPLC), and enzyme-linked immunosorbent assay (ELISA), can provide accurate results, but the procedures are usually time-consuming, sample-destructive and expensive. Predicting aflatoxin levels prior to crop harvest is useful for minimizing the health and financial impact of a contaminated crop. Consequently, there is interest in developing a tool that predicts aflatoxin levels based on topography and soil analysis data of fig orchards. This paper describes the development of a risk assessment tool for the contamination of aflatoxin on dried figs, based on the location and altitude of the fig orchards, the population of the fungus Aspergillus spp. in the soil, and soil parameters such as pH, saturation percentage (SP), electrical conductivity (EC), organic matter, particle size analysis (sand, silt, clay), concentration of the exchangeable cations (Ca, Mg, K, Na), extractable P and trace of elements (B, Fe, Mn, Zn and Cu), by employing machine learning methods. In particular, our proposed method integrates three machine learning techniques i.e., dimensionality reduction on the original dataset (Principal Component Analysis), metric learning (Mahalanobis Metric for Clustering) and K-nearest Neighbors learning algorithm (KNN), into an enhanced model, with mean performance equal to 85% by terms of the Pearson Correlation Coefficient (PCC) between observed and predicted values.
Abstract: Engagement is one of the most important factors in determining successful outcomes and deep learning in students. Existing approaches to detect student engagement involve periodic human observations that are subject to inter-rater reliability. Our solution uses real-time multimodal multisensor data labeled by objective performance outcomes to infer the engagement of students. The study involves four students with a combined diagnosis of cerebral palsy and a learning disability who took part in a 3-month trial over 59 sessions. Multimodal multisensor data were collected while they participated in a continuous performance test. Eye gaze, electroencephalogram, body pose, and interaction data were used to create a model of student engagement through objective labeling from the continuous performance test outcomes. In order to achieve this, a type of continuous performance test is introduced, the Seek-X type. Nine features were extracted including high-level handpicked compound features. Using leave-one-out cross-validation, a series of different machine learning approaches were evaluated. Overall, the random forest classification approach achieved the best classification results. Using random forest, 93.3% classification for engagement and 42.9% accuracy for disengagement were achieved. We compared these results to outcomes from different models: AdaBoost, decision tree, k-Nearest Neighbor, naïve Bayes, neural network, and support vector machine. We showed that using a multisensor approach achieved higher accuracy than using features from any reduced set of sensors. We found that using high-level handpicked features can improve the classification accuracy in every sensor mode. Our approach is robust to both sensor fallout and occlusions. The single most important sensor feature to the classification of engagement and distraction was shown to be eye gaze. It has been shown that we can accurately predict the level of engagement of students with learning disabilities in a real-time approach that is not subject to inter-rater reliability, human observation or reliant on a single mode of sensor input. This will help teachers design interventions for a heterogeneous group of students, where teachers cannot possibly attend to each of their individual needs. Our approach can be used to identify those with the greatest learning challenges so that all students are supported to reach their full potential.
Abstract: Uncontrolled growth of abnormal cells in the lung in the form of tumor can be either benign (non-cancerous) or malignant (cancerous). Patients with Lung Cancer (LC) have an average of five years life span expectancy provided diagnosis, detection and prediction, which reduces many treatment options to risk of invasive surgery increasing survival rate. Computed Tomography (CT), Positron Emission Tomography (PET), and Magnetic Resonance Imaging (MRI) for earlier detection of cancer are common. Gaussian filter along with median filter used for smoothing and noise removal, Histogram Equalization (HE) for image enhancement gives the best results without inviting further opinions. Lung cavities are extracted and the background portion other than two lung cavities is completely removed with right and left lungs segmented separately. Region properties measurements area, perimeter, diameter, centroid and eccentricity measured for the tumor segmented image, while texture is characterized by Gray-Level Co-occurrence Matrix (GLCM) functions, feature extraction provides Region of Interest (ROI) given as input to classifier. Two levels of classifications, K-Nearest Neighbor (KNN) is used for determining patient condition as normal or abnormal, while Artificial Neural Networks (ANN) is used for identifying the cancer stage is employed. Discrete Wavelet Transform (DWT) algorithm is used for the main feature extraction leading to best efficiency. The developed technology finds encouraging results for real time information and on line detection for future research.
Abstract: This paper presents a Machine Learning (ML) approach to support Meningitis diagnosis in patients at a children’s hospital in Sao Paulo, Brazil. The aim is to use ML techniques to reduce the use of invasive procedures, such as cerebrospinal fluid (CSF) collection, as much as possible. In this study, we focus on predicting the probability of Meningitis given the results of a blood and urine laboratory tests, together with the analysis of pain or other complaints from the patient. We tested a number of different ML algorithms, including: Adaptative Boosting (AdaBoost), Decision Tree, Gradient Boosting, K-Nearest Neighbors (KNN), Logistic Regression, Random Forest and Support Vector Machines (SVM). Decision Tree algorithm performed best, with 94.56% and 96.18% accuracy for training and testing data, respectively. These results represent a significant aid to doctors in diagnosing Meningitis as early as possible and in preventing expensive and painful procedures on some children.
Abstract: Goal: Purpose of the project was to make a plastic surgery prediction by using pre-operative images for the plastic surgeries’ patients and to show this prediction on a screen to compare between the current case and the appearance after the surgery. Methods: To this aim, we implemented a software which used data from the internet for facial skin diseases, skin burns, pre-and post-images for plastic surgeries then the post- surgical prediction is done by using K-nearest neighbor (KNN). So we designed and fabricated a smart mirror divided into two parts a screen and a reflective mirror so patient's pre- and post-appearance will be showed at the same time. Results: We worked on some skin diseases like vitiligo, skin burns and wrinkles. We classified the three degrees of burns using KNN classifier with accuracy 60%. We also succeeded in segmenting the area of vitiligo. Our future work will include working on more skin diseases, classify them and give a prediction for the look after the surgery. Also we will go deeper into facial deformities and plastic surgeries like nose reshaping and face slim down. Conclusion: Our project will give a prediction relates strongly to the real look after surgery and decrease different diagnoses among doctors. Significance: The mirror may have broad societal appeal as it will make the distance between patient's satisfaction and the medical standards smaller.
Abstract: Both Lidars and Radars are sensors for obstacle
detection. While Lidars are very accurate on obstacles positions
and less accurate on their velocities, Radars are more precise on
obstacles velocities and less precise on their positions. Sensor
fusion between Lidar and Radar aims at improving obstacle
detection using advantages of the two sensors. The present
paper proposes a real-time Lidar/Radar data fusion algorithm
for obstacle detection and tracking based on the global nearest
neighbour standard filter (GNN). This algorithm is implemented
and embedded in an automative vehicle as a component generated
by a real-time multisensor software. The benefits of data fusion
comparing with the use of a single sensor are illustrated through
several tracking scenarios (on a highway and on a bend) and
using real-time kinematic sensors mounted on the ego and tracked
vehicles as a ground truth.
Abstract: Missing values in real-world datasets are a common
problem. Many algorithms were developed to deal with this
problem, most of them replace the missing values with a fixed
value that was computed based on the observed values. In
our work, we used a distance function based on Bhattacharyya
distance to measure the distance between objects with missing
values. Bhattacharyya distance, which measures the similarity of
two probability distributions. The proposed distance distinguishes
between known and unknown values. Where the distance between
two known values is the Mahalanobis distance. When, on the other
hand, one of them is missing the distance is computed based on the
distribution of the known values, for the coordinate that contains
the missing value. This method was integrated with Wikaya, a
digital health company developing a platform that helps to improve
prevention of chronic diseases such as diabetes and cancer. In order
for Wikaya’s recommendation system to work distance between users
need to be measured. Since there are missing values in the collected
data, there is a need to develop a distance function distances between
incomplete users profiles. To evaluate the accuracy of the proposed
distance function in reflecting the actual similarity between different
objects, when some of them contain missing values, we integrated it
within the framework of k nearest neighbors (kNN) classifier, since
its computation is based only on the similarity between objects. To
validate this, we ran the algorithm over diabetes and breast cancer
datasets, standard benchmark datasets from the UCI repository. Our
experiments show that kNN classifier using our proposed distance
function outperforms the kNN using other existing methods.
Abstract: Datasets or collections are becoming important assets by themselves and now they can be accepted as a primary intellectual output of a research. The quality and usage of the datasets depend mainly on the context under which they have been collected, processed, analyzed, validated, and interpreted. This paper aims to present a collection of program educational objectives mapped to student’s outcomes collected from self-study reports prepared by 32 engineering programs accredited by ABET. The manual mapping (classification) of this data is a notoriously tedious, time consuming process. In addition, it requires experts in the area, which are mostly not available. It has been shown the operational settings under which the collection has been produced. The collection has been cleansed, preprocessed, some features have been selected and preliminary exploratory data analysis has been performed so as to illustrate the properties and usefulness of the collection. At the end, the collection has been benchmarked using nine of the most widely used supervised multiclass classification techniques (Binary Relevance, Label Powerset, Classifier Chains, Pruned Sets, Random k-label sets, Ensemble of Classifier Chains, Ensemble of Pruned Sets, Multi-Label k-Nearest Neighbors and Back-Propagation Multi-Label Learning). The techniques have been compared to each other using five well-known measurements (Accuracy, Hamming Loss, Micro-F, Macro-F, and Macro-F). The Ensemble of Classifier Chains and Ensemble of Pruned Sets have achieved encouraging performance compared to other experimented multi-label classification methods. The Classifier Chains method has shown the worst performance. To recap, the benchmark has achieved promising results by utilizing preliminary exploratory data analysis performed on the collection, proposing new trends for research and providing a baseline for future studies.
Abstract: Crop yield prediction is a paramount issue in
agriculture. The main idea of this paper is to find out efficient
way to predict the yield of corn based meteorological records.
The prediction models used in this paper can be classified into
model-driven approaches and data-driven approaches, according to
the different modeling methodologies. The model-driven approaches are based on crop mechanistic
modeling. They describe crop growth in interaction with their
environment as dynamical systems. But the calibration process of
the dynamic system comes up with much difficulty, because it
turns out to be a multidimensional non-convex optimization problem.
An original contribution of this paper is to propose a statistical
methodology, Multi-Scenarios Parameters Estimation (MSPE), for the
parametrization of potentially complex mechanistic models from a
new type of datasets (climatic data, final yield in many situations).
It is tested with CORNFLO, a crop model for maize growth. On the other hand, the data-driven approach for yield prediction
is free of the complex biophysical process. But it has some strict
requirements about the dataset.
A second contribution of the paper is the comparison of these
model-driven methods with classical data-driven methods. For this
purpose, we consider two classes of regression methods, methods
derived from linear regression (Ridge and Lasso Regression, Principal
Components Regression or Partial Least Squares Regression) and
machine learning methods (Random Forest, k-Nearest Neighbor,
Artificial Neural Network and SVM regression).
The dataset consists of 720 records of corn yield at county scale
provided by the United States Department of Agriculture (USDA) and
the associated climatic data. A 5-folds cross-validation process and
two accuracy metrics: root mean square error of prediction(RMSEP),
mean absolute error of prediction(MAEP) were used to evaluate the
crop prediction capacity.
The results show that among the data-driven approaches, Random
Forest is the most robust and generally achieves the best prediction
error (MAEP 4.27%). It also outperforms our model-driven approach
(MAEP 6.11%). However, the method to calibrate the mechanistic
model from dataset easy to access offers several side-perspectives.
The mechanistic model can potentially help to underline the stresses
suffered by the crop or to identify the biological parameters of interest
for breeding purposes. For this reason, an interesting perspective is
to combine these two types of approaches.
Abstract: This paper investigates successful sub-bands of wave atom transform via classification of mammograms, when the coefficients of sub-bands are used as features. A computer-aided diagnosis system is constructed by using wave atom transform, support vector machine and k-nearest neighbor classifiers. Two-class classification is studied in detail using two data sets, separately. The successful sub-bands are determined according to the accuracy rates, coefficient numbers, and sensitivity rates.
Abstract: The success or failure of students is a concern for every academic institution, college, university, governments and students themselves. Several approaches have been researched to address this concern. In this paper, a view is held that when a student enters a university or college or an academic institution, he or she enters an academic environment. The academic environment is unique concept used to develop the solution for making predictions effectively. This paper presents a model to determine the propensity of a student to succeed or fail in the French South African Schneider Electric Education Center (FSASEC) at the Vaal University of Technology (VUT). The Decision Tree algorithm is used to implement the model at FSASEC.
Abstract: DNA data have been used in forensics for decades. However, current research looks at using the DNA as a biometric identity verification modality. The goal is to improve the speed of identification. We aim at using gene data that was initially used for autism detection to find if and how accurate is this data for identification applications. Mainly our goal is to find if our data preprocessing technique yields data useful as a biometric identification tool. We experiment with using the nearest neighbor classifier to identify subjects. Results show that optimal classification rate is achieved when the test set is corrupted by normally distributed noise with zero mean and standard deviation of 1. The classification rate is close to optimal at higher noise standard deviation reaching 3. This shows that the data can be used for identity verification with high accuracy using a simple classifier such as the k-nearest neighbor (k-NN).
Abstract: Instance selection (IS) technique is used to reduce
the data size to improve the performance of data mining methods.
Recently, to process very large data set, several proposed methods
divide the training set into some disjoint subsets and apply IS
algorithms independently to each subset. In this paper, we analyze
the limitation of these methods and give our viewpoint about how to
divide and conquer in IS procedure. Then, based on fast condensed
nearest neighbor (FCNN) rule, we propose a large data sets instance
selection method with MapReduce framework. Besides ensuring the
prediction accuracy and reduction rate, it has two desirable properties:
First, it reduces the work load in the aggregation node; Second
and most important, it produces the same result with the sequential
version, which other parallel methods cannot achieve. We evaluate the
performance of FCNN-MR on one small data set and two large data
sets. The experimental results show that it is effective and practical.
Abstract: In this study, our goal was to perform tumor staging and subtype determination automatically using different texture analysis approaches for a very common cancer type, i.e., non-small cell lung carcinoma (NSCLC). Especially, we introduced a texture analysis approach, called Law’s texture filter, to be used in this context for the first time. The 18F-FDG PET images of 42 patients with NSCLC were evaluated. The number of patients for each tumor stage, i.e., I-II, III or IV, was 14. The patients had ~45% adenocarcinoma (ADC) and ~55% squamous cell carcinoma (SqCCs). MATLAB technical computing language was employed in the extraction of 51 features by using first order statistics (FOS), gray-level co-occurrence matrix (GLCM), gray-level run-length matrix (GLRLM), and Laws’ texture filters. The feature selection method employed was the sequential forward selection (SFS). Selected textural features were used in the automatic classification by k-nearest neighbors (k-NN) and support vector machines (SVM). In the automatic classification of tumor stage, the accuracy was approximately 59.5% with k-NN classifier (k=3) and 69% with SVM (with one versus one paradigm), using 5 features. In the automatic classification of tumor subtype, the accuracy was around 92.7% with SVM one vs. one. Texture analysis of FDG-PET images might be used, in addition to metabolic parameters as an objective tool to assess tumor histopathological characteristics and in automatic classification of tumor stage and subtype.
Abstract: A database that records average traffic speeds measured at five-minute intervals for all the links in the traffic network of a metropolitan city. While learning from this data the models that can predict future traffic speed would be beneficial for the applications such as the car navigation system, building predictive models for every link becomes a nontrivial job if the number of links in a given network is huge. An advantage of adopting k-nearest neighbor (k-NN) as predictive models is that it does not require any explicit model building. Instead, k-NN takes a long time to make a prediction because it needs to search for the k-nearest neighbors in the database at prediction time. In this paper, we investigate how much we can speed up k-NN in making traffic speed predictions by reducing the amount of data to be searched for without a significant sacrifice of prediction accuracy. The rationale behind this is that we had a better look at only the recent data because the traffic patterns not only repeat daily or weekly but also change over time. In our experiments, we build several different k-NN models employing different sets of features which are the current and past traffic speeds of the target link and the neighbor links in its up/down-stream. The performances of these models are compared by measuring the average prediction accuracy and the average time taken to make a prediction using various amounts of data.
Abstract: Developing complete mechanistic models for polymerization reactors is not easy, because complex reactions occur simultaneously; there is a large number of kinetic parameters involved and sometimes the chemical and physical phenomena for mixtures involving polymers are poorly understood. To overcome these difficulties, empirical models based on sampled data can be used instead, namely regression methods typical of machine learning field. They have the ability to learn the trends of a process without any knowledge about its particular physical and chemical laws. Therefore, they are useful for modeling complex processes, such as the free radical polymerization of methyl methacrylate achieved in a batch bulk process. The goal is to generate accurate predictions of monomer conversion, numerical average molecular weight and gravimetrical average molecular weight. This process is associated with non-linear gel and glass effects. For this purpose, an adaptive sampling technique is presented, which can select more samples around the regions where the values have a higher variation. Several machine learning methods are used for the modeling and their performance is compared: support vector machines, k-nearest neighbor, k-nearest neighbor and random forest, as well as an original algorithm, large margin nearest neighbor regression. The suggested method provides very good results compared to the other well-known regression algorithms.
Abstract: Data mining is the process of extracting useful or hidden information from a large database. Extracted information can be used to discover relationships among features, where data objects are grouped according to logical relationships; or to predict unseen objects to one of the predefined groups. In this paper, we aim to investigate four well-known data mining algorithms in order to predict groundwater areas in Jordan. These algorithms are Support Vector Machines (SVMs), Naïve Bayes (NB), K-Nearest Neighbor (kNN) and Classification Based on Association Rule (CBA). The experimental results indicate that the SVMs algorithm outperformed other algorithms in terms of classification accuracy, precision and F1 evaluation measures using the datasets of groundwater areas that were collected from Jordanian Ministry of Water and Irrigation.
Abstract: Self-driving vehicle require a high level of situational
awareness in order to maneuver safely when driving in real world
condition. This paper presents a LiDAR based real time perception
system that is able to process sensor raw data for multiple target
detection and tracking in dynamic environment. The proposed
algorithm is nonparametric and deterministic that is no assumptions
and priori knowledge are needed from the input data and no
initializations are required. Additionally, the proposed method is
working on the three-dimensional data directly generated by LiDAR
while not scarifying the rich information contained in the domain of
3D. Moreover, a fast and efficient for real time clustering algorithm
is applied based on a radially bounded nearest neighbor (RBNN).
Hungarian algorithm procedure and adaptive Kalman filtering are
used for data association and tracking algorithm. The proposed
algorithm is able to run in real time with average run time of 70ms
per frame.
Abstract: In present scenario, cardiovascular problems are growing challenge for researchers and physiologists. As heart disease have no geographic, gender or socioeconomic specific reasons; detecting cardiac irregularities at early stage followed by quick and correct treatment is very important. Electrocardiogram is the finest tool for continuous monitoring of heart activity. Heart rate variability (HRV) is used to measure naturally occurring oscillations between consecutive cardiac cycles. Analysis of this variability is carried out using time domain, frequency domain and non-linear parameters. This paper presents HRV analysis of the online dataset for normal sinus rhythm (taken as healthy subject) and sudden cardiac death (SCD subject) using all three methods computing values for parameters like standard deviation of node to node intervals (SDNN), square root of mean of the sequences of difference between adjacent RR intervals (RMSSD), mean of R to R intervals (mean RR) in time domain, very low-frequency (VLF), low-frequency (LF), high frequency (HF) and ratio of low to high frequency (LF/HF ratio) in frequency domain and Poincare plot for non linear analysis. To differentiate HRV of healthy subject from subject died with SCD, k –nearest neighbor (k-NN) classifier has been used because of its high accuracy. Results show highly reduced values for all stated parameters for SCD subjects as compared to healthy ones. As the dataset used for SCD patients is recording of their ECG signal one hour prior to their death, it is therefore, verified with an accuracy of 95% that proposed algorithm can identify mortality risk of a patient one hour before its death. The identification of a patient’s mortality risk at such an early stage may prevent him/her meeting sudden death if in-time and right treatment is given by the doctor.