International Journal on Artificial Intelligence Tools

Feature Selection (FS) is an important preprocessing step in data analytics. It is used to select a subset of the original feature set such that the selected subset does not affect the classification performance significantly. Its objective is to remove irrelevant and redundant features from the original dataset. FS can be done either in offline mode or in online mode. The basic assumption in the former mode is that the entire dataset has been available for the FS algorithm; and the FS algorithm takes multiple epochs to select optimal feature subset that gives good accuracy. In contrast, the FS algorithms in online mode take input data one instance at a time and accumulate knowledge by learning each one of them. In online mode each instance of the original dataset is considered as training and testing sample as well. The offline FS algorithms require long time periods, if the data to be processed is large such as Big data. Whereas online FS algorithms will take only one epoch to learn the entire data and can produce the results swiftly which is highly desirable in the case of Big data. This paper deals with the online FS problem and provides a novel Feature Selection algorithm which uses the Sparse Gradient method to build a sparse classifier. In this proposed method, an online classifier is built and maintained throughout the learning process and feature weights, which are limited to a particular boundary limit, are reduced in a step by step decrement process. This method creates sparsity in the classifier. Effectively, the built classifier is used to select optimal feature subset from the incoming data. As this method reduces the weights in the classifier in step by step manner, only those important features which have value higher than the boundary survive from this repeated decrement process. The resultant optimal feature subset is formed using these non-zero weighted features. Most significantly, this particular method can be used with any learning algorithm. To show its applicability with different learning algorithms, various online feature selection models have been built using Learning Vector Quantization, Radial Basis Function Networks and Adaptive Resonance Theory MAP. In all these models, the proposed Sparse Gradient method is used. The encouraging results shows the effectiveness of the proposed method with different learning algorithms in medium and large sized benchmark datasets.

With a high rise in deaths caused due to novel coronavirus (nCoV), immunocompromised persons are at high risk. Lung cancer is no exception. Classifying lung cancer patients and Covid-19 is the primary aim of the paper. For this, we propose a deep ensemble neural network (VGG16, DenseNet121, ResNet50 and custom CNN) to detect Covid-19 and lung cancer using chest CT images. We validate our model using three different datasets, namely SPIE AAPM Lung CT Challenge (1503 images), Covid CT dataset (349 images), and SARS-CoV-2 CT-scan dataset (1252 images). We utilize a k(= 5) fold cross-validation approach on the individual deep neural networks (DNNs) and a custom designed CNN model architecture, and achieve a benchmark score of 96.30% (accuracy) with a sensitivity and precision value of 96.39% and 98.44%, respectively. The proposed model effectively utilizes diverse models. To the best of our knowledge, using ensemble DNN, this is the first time we analyze chest CT images to separate lung cancer from Covid-19 (and vice-versa).

As our aim is to classify Covid-19 and lung cancer using chest CT images, it helps in prioritizing immunocompromised persons from Covid-19 for a better patient care. Also, mass screening is possible especially in resource-constrained regions since CT scans are cheaper. The long-term goal is to check whether AI-guided tool(s) is(are) able to prioritize patients that are at high risk (e.g., lung disease) from any possible future infectious disease outbreaks.

Early diagnosis of the disease in crop opens the door to potential care and treatment, which in turn improves the yield. Automated detection of rice plant disease from plant images is an emerging research field that is gaining prominence due to the rising interest in it from machine learning researchers. Convolution neural network-based learning techniques are widely used by the research community in accurately handling these types of tasks. In current study, transfer learning models like InceptionV3, ResNet152V2, MobileNetV2, Xception, DenseNet201, InceptionResNetV2 and VGG19 are used for the diagnosis of rice plant diseases on two datasets, which are publicly available in Mendeley and Kaggle. To improve the performance of the diagnosis system, ensembling of Transfer Learning (TL) models has been introduced in this work. All possible combinations of the TL models are designed and experiment is carried out using all of them. It is observed that most of the ensembled models are superior to individual TL models. The work demonstrates that some of the ensemble models are superior in performance than the advanced learning model like Convolution-XGBoost.

Convolutional neural networks (CNN) have shown remarkable performance in enormous computer vision applications over the years, and many works have been done for human activity recognition (HAR) using CNN. However, most of deep learning-based architectures require large training data and plenty of computational resources. Therefore, we proposed a simple and efficient deep learning model for human activity recognition, which works on complex visual data and require less computational resources. In the proposed work, we designed a novel CNN architecture by stacking the repeated components called micro-networks to incorporate multiscale processing in the network layers. We have used a feature fusion strategy to pass the previous layer’s abstract complementary information to the next adjacent layer, believing that each layer encapsulates specific feature maps. Therefore, the hidden complementary information can potentially enhance the feature discrimination capacity of the network and help in learning the network. The proposed architecture is fully trained from scratch with stochastic gradient descent (SGD) optimizer at 0.05 initial learning rate and a softmax classifier is used for activity recognition. The merit of the proposed method over standard deep learning models is its computational efficiency in terms of learnable parameters and computational resources. The proposed model gives good performance on small as well as large size datasets. For authentication of the proposed method several extensive experiments are conducted on publically available datasets, namely UCF-101, HMDB-51, YouTube, and IXMAS datasets. The results have shown the outperformance of the proposed method over the existing state-of-the-art methods.

In the present work, an indirect adaptive neural control method for nonlinear systems having unknown dynamics is proposed. The proposed control architecture is composed by a neural emulator (NE) and a neural controller (NC) where a new decoupled variable learning rates (VLRs) combined with Taylor development (TD) are used to train the NE and the NC. The developed VLRs mixed with the TD (TDVLRs) ensure a quick adaptation of neural networks parameters guaranteeing a faster output convergence and reducing the tracking error. The effectiveness of the proposed TDVLRs is illustrated by simulation with a nonlinear dynamic system. In order to validate simulation results, an application on a transesterification reactors is, also, presented.

The tracking of underwater targets represents a basic unit in underwater surveillance for offering effective search and rescue operations. However, the tracking of the target is a major constraint in finding the underwater target. This paper devises a model for underwater target tracking considering radar signals. Radar signals are subjected to the image reconstruction process to make them suitable for further processing. The gridding is the next step, which is utilized for generating the grids to focus on each region in a precise manner. The features like Ridgelet transform and Local Gabor XOR Pattern (LGXP) are produced. Data augmentation is the next process, which helps to reduce the size of data by performing resizing. At last, the tracking of targets is performed from each grid with a Deep neuro fuzzy network (DNFN). Here, DNFN training is performed with Exponential-Competitive Swarm Optimization (E-CSO), which is derived by integrating Exponential Weighted Moving Average (EWMA) and Competitive Swarm Optimization (CSO). The proposed E-CSO-based DNFN offered high efficiency with a maximal detection rate of 98.8%, minimal Mean square error of 0.012, and Mean Absolute Percentage Error (MAPE) of 17.883.

The complexity and frequent fluctuations of economic data pose a significant challenge to forecasting studies. In order to predict financial data more accurately, we build a fusion model for predicting financial data based on the idea of decomposition-recombination by combining the Sooty Tern Optimization Algorithm (STOA), Variational Mode Decomposition (VMD), Support Vector Machine (SVM) and Back Propagation Neural Network (BPNN). Firstly, after finding the optimal parameters of the VMD by the STOA, adjust the parameters of the VMD and decompose the financial data, remove the residual from the series, and constitute the remaining necessary information within the financial data into new series modeled by applying the SVM. Finally, calculate the error between the predicted and actual values. Notably, the residual is treated specially in the modeling. A Double-layer BPNN model is used to establish a mechanism to increase the sensitivity of the model fluctuations. The influence factor series and the residual are introduced as input variables to the BPNN to establish a mapping relationship between them and the error series. The results show that the model improves data utilization through five experiments, solves the problem of the VMD insensitivity to fluctuations, and improves the prediction accuracy of financial time series effectively.

An integer overflow error occurs when an integer operation in computer software evaluates a value out of the integer range. It can lead to a fatal system failure. The existing approaches to detecting integer overflow errors rely on data/control-flow analysis of the code or execution of the code with test cases. This paper presents a supervised learning approach to predicting whether each method in a given Java program has an integer overflow error by treating the source code as text. Built upon real-world programs, our Java dataset covers all integer data types and operations in Java, the methods for preventing integer overflow errors, and adversarial samples. We have evaluated six classification models, BERT, DistilBERT, codeBERT, Code2Vec, fastText, and NBSVM. They represent different text embedding techniques for dealing with source code. The experiment results show that BERT and its variants have outperformed other models. We have applied the resultant BERT model to eleven real-world projects, including JDK13.0 and ten top-ranked GitHub projects, and revealed 181 integer overflow errors. In addition, we have evaluated the classification models with a C/C++ dataset. The result is similar to that of the Java dataset.

Frequent flyers have been shown to have a significant impact on airline long-term profitability, so it is becoming increasingly crucial to understand their needs. Objective of this paper is to forecast the next flight of frequent flyers, which will not only improve the customer experience but also assist airlines optimize their ticketing service platforms. In the data preparation phase, we use methods including data transformation and genetic algorithms (GA) to directly extract statistical features or excavate new predictors, which is inspired by the characteristics of the frequent flyers of Beijing Daxing International Airport (PKX). We synthesize ensemble models, linear regression models, evolutionary computation models, and fusion models for prediction. The hyperparameters can be optimized by the Tree-structured Parzen Estimator (TPE). After extensive comparison, the model developed with the average fusion strategy obtains the best prediction on the test set with a minimal MSLE of 0.7893.

In this paper, we present a new method for Lane Detection (LD) to reduce the impact of some issues associated with Autonomous Driving Cars (ADCs). It relies on the fact that self-driving will be the getaway of the transportation future. The ADC technology has expanded as it does not require human assistance. For ACD to be fully utilized, Machine Learning (ML) tools, such as Deep Neural Networks (DNNs), are required. However, ACD technology uses tools like a camera, GPS, Ultrasonic Sensor (US), and others, which work together to produce accurate ACDs. While adopting ACD, several issues need to be addressed such as carves, blurry, unclean, shadow, sunlight, pedestrians, and weather conditions. The paper proposes a new method to detect when a portion of a road line is missing due to weather conditions or old marking, protecting pedestrians and cars. For lane detection, the DNN model and Hough transform are used. CULane datasets including training and testing samples were used in the experiments. The results show high accuracy levels of 92%, indicating the ability to detect road lanes.

The research on recognition of single-multiple objects has significantly matured with many techniques and robust results. The probabilistic recognition of incomplete objects, however, remains an active field with challenging issues related to their degree of incompleteness and associated issues due to shadows, illumination and other visual characteristics. Object incompleteness here means that only a small portion of a known object is visible and not low-resolution or deformation of that object. The employment of various single machine-learning methodologies for accurate classification of the incomplete objects did not always provide a robust answer to all challenging issues or requires a very long training list of variations of each object. In this paper, we present the initial results using Local-Global (L-G) graphs and Voronoi (V) diagrams methodologies associated with machine learning to generate probabilistic matches of objects with varying degrees of incomplete views. The first results with limited use of synthesis of views are promising and have triggered an interest to continue the efforts for recognizing incomplete objects in a progressive way with synthesis of views of segmented regions that compose an object.

In the context of machine learning, an imbalanced classification problem states to a dataset in which the classes are not evenly distributed. This problem commonly occurs when attempting to classify data in which the distribution of labels or classes is not uniform. Using resampling methods to accumulate samples or entries from the minority class or to drop those from the majority class can be considered the best solution to this problem. The focus of this study is to propose a framework pattern to handle any imbalance dataset for fraud detection. For this purpose, Undersampling (Random and NearMiss) and oversampling (Random, SMOTE, BorderLine SMOTE) were used as resampling techniques for the concentration of our experiments for balancing an evaluated dataset. For the first time, a large-scale unbalanced dataset collected from the Kaggle website was used to test both methods for detecting fraud in the Tunisian company for electricity and gas consumption. It was also evaluated with four machine learning classifiers: Logistic Regression (LR), Naïve Bayes (NB), Random Forest, and XGBoost. Standard evaluation metrics like precision, recall, F1-score, and accuracy have been used to assess the findings. The experimental results clearly revealed that the RF model provided the best performance and outperformed all other matched classifiers with attained a classification accuracy of 89% using NearMiss undersampling and 99% using Random oversampling.