Higher Order Statistical Features Research Articles

Impact of harmonization on the reproducibility of MRI radiomic features when using different scanners, acquisition parameters, and image pre-processing techniques: a phantom study.

This study investigated the impact of ComBat harmonization on the reproducibility of radiomic features extracted from magnetic resonance images (MRI) acquired on different scanners, using various data acquisition parameters and multiple image pre-processing techniques using a dedicated MRI phantom. Four scanners were used to acquire an MRI of a nonanatomic phantom as part of the TCIA RIDER database. In fast spin-echo inversion recovery (IR) sequences, several inversion durations were employed, including 50, 100, 250, 500, 750, 1000, 1500, 2000, 2500, and 3000 ms. In addition, a 3D fast spoiled gradient recalled echo (FSPGR) sequence was used to investigate several flip angles (FA): 2, 5, 10, 15, 20, 25, and 30 degrees. Nineteen phantom compartments were manually segmented. Different approaches were used to pre-process each image: Bin discretization, Wavelet filter, Laplacian of Gaussian, logarithm, square, square root, and gradient. Overall, 92 first-, second-, and higher-order statistical radiomic features were extracted. ComBat harmonization was also applied to the extracted radiomic features. Finally, the Intraclass Correlation Coefficient (ICC) and Kruskal-Wallis's (KW) tests were implemented to assess the robustness of radiomic features. The number of non-significant features in the KW test ranged between 0-5 and 29-74 for various scanners, 31-91 and 37-92 for three times tests, 0-33 to 34-90 for FAs, and 3-68 to 65-89 for IRs before and after ComBat harmonization, with different image pre-processing techniques, respectively. The number of features with ICC over 90% ranged between 0-8 and 6-60 for various scanners, 11-75 and 17-80 for three times tests, 3-83 to 9-84 for FAs, and 3-49 to 3-63 for IRs before and after ComBat harmonization, with different image pre-processing techniques, respectively. The use of various scanners, IRs, and FAs has a great impact on radiomic features. However, the majority of scanner-robust features is also robust to IR and FA. Among the effective parameters in MR images, several tests in one scanner have a negligible impact on radiomic features. Different scanners and acquisition parameters using various image pre-processing might affect radiomic features to a large extent. ComBat harmonization might significantly impact the reproducibility of MRI radiomic features.

Ensemble Model for Stock Price Forecasting: MapReduce Framework for Big Data Handling: An Optimal Trained Hybrid Model for Classification

A number of authors have focused on this study to examine how huge data are perceived. A novel big data classification paradigm is introduced by the work’s preprocessing, feature extraction and classification techniques. Data normalization is carried out at the preprocessing stage. The MapReduce framework is then utilized to manage the massive data. Statistical features (mean, median, min/max and SD), higher-order statistical features (skewness, kurtosis and enhanced entropy), and correlation-based features are all extracted prior to classification. The Bi-LSTM and deep maxout hybrid classification model classifies the data during the reduction stage. To assure classification accuracy, training will also be deployed by the new Hybrid Butterfly Positioned Coot Optimization (HBPCO) algorithm. The proposed method’s accuracy of 97.45% beats the methods of NN (85.13%), CNN (83.78%), RNN (78.37%), Bi-LSTM (82.43%) and SVM (87.83%).

Hybrid model for rainfall prediction with statistical and technical indicator feature set

As excessive rain may cause numerous disasters, rainfall prediction is very crucial and the prediction should be realistic since it encourages individuals to take preventative steps. This work proposes a new rainfall prediction model by following three major phases: Preprocessing, Extraction of Features and Prediction. The Improved Data Normalization technique is carried out for preprocessing the data. Then, it extracts the technical Indicator features like Average Directional Movement (ADX), Moving Average Convergence Divergence (MACD), RSI, Welles Wilder's Smoothing Average (WWS), statistical & higher order statistical features. According to these features, prediction takes place by a hybrid model that includes the Improved DBN & LSTM methods. By fine-tuning the best weights, the training will be carried out optimally to increase the prediction model's efficacy. For this, the Brownian Motion-based Pelican Optimization Algorithm (BMPOA) is implemented. Finally, the proposed technique has contrasted over existing models concerning different metrics.

Arrhythmia classification using ECG signal: A meta-heuristic improvement of optimal weighted feature integration and attention-based hybrid deep learning model

In healthcare facilities, the most common and least expensive diagnostic tool for monitoring electrical signals in the heart is the Electrocardiogram (ECG). Arrhythmia is nothing but abnormal rhythms in the heart. Arrhythmias can be dangerous and fatal. An ECG test can identify the type of arrhythmia that's causing the abnormal heartbeat. These necessitate lengthy, high-quality ECG segments, which may result in a few important episode omissions. Because of this, a low-cost, dependable, and automatic system for screening and treating arrhythmia becomes highly necessary. A novel method for the classification of arrhythmias is suggested on the basis of attention-based deep learning to satisfy the above-mentioned requirement. Signal collection, signal analysis, feature extraction, weighted fused features, and classification are the various phases of the proposed work. The input signal is initially obtained from benchmark data sources. The RR interval and QRS complex signal analysis are carried out next. A One-Dimensional Convolution Neural Network (1DCNN) is used to extract the deep features from the resulting RR interval and QRS complex, as well as the Higher-Order Statistical (HOS) features from the raw signal. The weighted fused feature selection is carried out after the above-mentioned three features have been obtained, where the weight is adjusted with the Recommended Adaptive Risk Rate-based Lemurs Optimization Algorithm (ARR-LO). Finally, the fused features are fed into the hybrid Attention model through the Long Short-Term Memory (LSTM) and Deep Temporal Convolution Network (DTCN), where ARR-LO is used to optimally determine some of the hyperparameters. The objectives of the study include developing an arrhythmia classification framework, constructing an adaptive attention-based hybrid deep learning model (HA-LSTM-DTCN), creating a new meta-heuristic algorithm (ARR-LO), and validating the suggested model against existing classifiers. The performance is compared with the conventional models and evaluated using a variety of metrics and the results from the simulation, the generated deep learning-based arrhythmia classification framework outperformed conventional prediction models such as DTCN, LSTM-DTCN, DNN, and LSTM classifiers in classifying the arrhythmia by 81.79%, 55.46%, and 80.73%, respectively and the overall accuracy of the proposed framework performs 94.425%. From the results, it was shown that generated AR-LO-HA-LSTM-DTCN model for arrhythmia classification improved the classification performance, thus making it useful in the quick diagnosis of arrhythmia disease. The findings show that the suggested arrhythmia classification model improves the classification performance, which can be used to quickly diagnose arrhythmia disease.

Deep learning based concrete compressive strength prediction model with hybrid meta-heuristic approach

In concrete design, the compressive strength of the concrete is the critical parameter that defines the quality of concrete. Determination of the compressive strength of concrete by laboratory testing is the conventional way, however, with the help of advancements in technology, the development of an exact predicting technique for the compressive strength of concrete will reduce time as well as cost to a greater extent. There are various types of concrete, as concrete is the most used material and has wide applications in the construction sector and hence huge article is being done to increase the performance of concrete. In this paper, the compressive strength of High-Performance Concrete and Fiber Reinforced High Strength Self-Compacting Concrete is predicted by next three main stages, namely (a) pre-processing, (b) feature extraction, and (c) prediction of concrete compressive strength. Initially, in the data pre-processing phase, duplicate data samples are removed. Subsequently, various features like statistical features, symmetrical uncertainty-based features, higher order statistical features, ReliefF feature, and are derived from the pre-processed data. In the prediction phase of concrete compressive strength, a hybrid algorithm is designed by mingling the concepts of Deep Belief Network and Long Short-Term Memory. Moreover, The Rock Hyraxes Updated Sparrow Algorithm is used to fine-tune the weights of DBN to improve the prediction accuracy of the suggested work. The suggested hybrid optimization method is the hybridization of the standard Sparrow Search Algorithm and Rock Hyraxes Swarm Optimization, respectively. Eventually, the output from the optimized hybrid method will reveal the compression energy of concrete. Ultimately, the efficiency of the suggested method is verified by the comparative estimation. The MAPE of the suggested method is 235.32, which is 30.6%, 21.64%, 20.4%, and 11.4% superior than the CLPSO, ABC, SSA + Hybrid Classifier and RHSO + Hybrid Classifier, respectively.

Task prediction on tablet based deep learning method

The ability of smartphones and other mobile technology to improve cognitive abilities has long been acknowledged by all. This model includes pre-processing, enhanced higher-order feature extraction, and prediction stages. At first, the collected data from different age groups is pre-processed using a data cleaning approach. The feature extraction stage extracts the crucial features from the pre-processed data, including enhanced higher-order statistical features. The final prediction regarding the automatic prediction of variations in pupil diameter of the left eye, right eye, and reading ability (count of characters read/second) is made using three mathematical models: optimized Deep Convolutional Neural Network 1 (DCNN1), optimized DCNN2, and optimized DCNN3. Furthermore, the weight functions of the three mathematical models are fine-tuned to increase prediction accuracy using a hybrid optimization model called Arithmetically Updated Poor Rich Optimization (AUPRO). The comparative assessment is undergone to verify the efficiency of the projected model in terms of various measures.

DDOFM: Dynamic malicious domain detection method based on feature mining

The domain name system is an essential part of the network, and target hosts are often attacked by malicious domain names to steal resources. Some traditional detection methods have low accuracy, poor generalization ability, and high resource overhead on model construction to deal with complex and variable malicious domain names. A three-level dynamic malicious domain detection method (DDOFM) is proposed in this paper. DDOFM only needs to combine a few high-order statistical features of benign domains with some DNS features, without flagging malicious samples and involving them in training. Firstly, the boundary recognition of passive DNS (PDNS) features extracted from DNS traffic is carried out to conduct an early- warning for some domains. Second, the Hidden Markov Model (HMM) forward algorithm and normal distribution probability density function are used to calculate the formation probabilities of the warned domains and their probability density values. Then the probabilities of every character in the warned domain name and the standard deviation between these probabilities are counted. Further, the probability density values and the standard deviations of these warned domain names are compared with their respective thresholds to identify the attribution of the warned domain names. Finally, if the domain name is not warned, the Jensen–Shannon divergence (JS divergence) between it and the previous domain name will be calculated. Then the local iterative threshold finding algorithm (LLTFA) proposed in this paper will be combined to determine its attribution and identify whether the host is connected to the command and control (C&C) server. Experiments show that the detection indexes of this method exceed 99% for multiple types of malicious domain names. The C&C servers can also be identified by DDOFM faster than similar methods.

Dual stage ensemble technique for intrusion detection in cloud computing

A capability of cloud-based IDS in identifying complicated and anonymous attacks is rising in the current era. However, unwanted delays hinder the detection rate. A malicious user might utilize vast quantities of computational power. The cloud provides to perform attacks both within and without the cloud. Furthermore, there are major challenges for intrusion detection due to the ease of the cloud and also the continual restructuring and movement of cloud resources. Intruder detection, feature extraction, and data processing are all included in the novel optimization-based Intrusion Detection System (IDS) paradigm that will be presented in this study. Data normalization is used to first pre-process the input data. Then, appropriate feature extraction is carried out, including the extraction of (a) raw features, (b) statistical features, then (c) higher-order statistical features using suggested kurtosis. The detection phase is then applied to the retrieved features. A two-stage ensemble method is suggested for finding intruders in clouds. Random forest (RF), Support Vector Machine (SVM), optimal Neural Network (NN), and RNN make up the suggested ensemble technique. The RF, SVM, and Optimized NN algorithms are directly fed the collected features. The output of these classifiers is then provided to the RNN classifier (i.e.), RF output to RNN1, SVM output to RNN2, and optimized NN output to RNN3. Then, the weighted average of RNN 1, 2, and 3 is considered as the final output. A Self Adaptive Salp Swarm Optimization optimizes the weights of NN for exact detection (SA-SSO). Finally, a test is conducted to confirm the developed model’s superiority.

Using biospeckle and LIBS techniques with artificial intelligence to monitor phthalocyanine-gold nanoconjugates as a new drug delivery mediator for in vivo PDT

Photodynamic therapy (PDT) is a promising treatment method that generates reactive oxygen species by exciting a photosensitizer in the presence of light. The positive treatment outcome of PDT is enhanced by the photosensitizer’s ability to get out quickly from normal tissue. This study non-destructively evaluated the biocompatibility of novel synthesized phthalocyanine-gold nanoconjugates (Pc-Au NCs) as a drug delivery mediator for in vivo PDT. We first proposed a novel noncontact and safe biospeckle optical imaging technique to visualize the quantitative changes in the internal dynamics of biological tissue in response to PDT treatment. Second, laser-induced breakdown spectroscopy (LIBS) was utilized to emphasize the accumulation of gold nanoparticles (Au NPs) within the normal and cancerous tissue, which in turn verifies the cytotoxic and treatment effects of the Pc-Au NCs. The biospeckle images and LIBS spectra were obtained from tumor and normal tissues of male BALB/c malignancy-bearing mice before and after exposure to PDT. Then, they were statistically treated via different artificial intelligence techniques. For biospeckle analysis, the proposed method was established from the combination of second and higher-order statistical features extracted from the gray-level co-occurrence matrix (GLCM) and local contrast analysis (LCA), respectively, through principal component analysis (PCA). Regarding LIBS analysis, the powerful, most popular ensemble sequential learning algorithm Adaptive Boost was implemented on highly uncorrelated variables extracted from the raw spectra. The results, confirmed by the histological investigation, indicated the potential of the biospeckle and LIBS statistical properties in demonstrating the efficiency of the Pc-Au NCs as a drug delivery mediator for in vivo PDT.

Hybrid classification model with tuned weight for cyber attack detection: Big data perspective

Cybercrime using big data is growing at an unprecedented rate, posing a serious threat to the Internet sector and global data. Traditional ways of mitigating cyber risks are becoming inadequate due to the more complex attack and offensive methods employed by cyber attackers, as well as the expanding importance of data-driven and intellect competitors. This work introduces new cyber attack detection (CAD) model in Big data that includes: “Preprocessing, Feature Extraction, Feature Selection, and Detection, Mitigation”. The preprocessing is done by using the improved class imbalance process. The variety of 3 features is extracted as “flow-based features, improved entropy-based features, and higher-order statistical features”. For feature selection, the Improved Independent component analysis (ICA) is used. Finally, the hybrid classifier includes LSTM and Deep Max out (DMO) in the detection process. Once the presence of an attack is detected, mitigation takes place via the proposed Bait mitigation process. The weights of Long Short-Term Memory (LSTM) are optimized by using the Self-Enhanced Sea Gull Optimization (SE-SGO) model. The maximum accuracy has been achieved (0.94) for the suggested approach which is 38%, 14.6%, 7.36%, 38.7%, and 10.5% superior to the other existing approaches like HC + SGO, HC + SSOA, HC + DHOA, HC + DOX, and HC + FF, respectively.

Blockchain Mechanism-Based Attack Detection in IoT with Hybrid Classification and Proposed Feature Selection

With IoT development, safety attacks are growing gradually as well. Numerous centralized attack-detecting schemes were developed for detecting IoT attacks namely, ML and DL schemes, which classify it as normal or attack. However, because to the various requirements of IoT devices, the most of the plans fall short of achieving noticeable results. In this study, a blockchain method is used to construct a novel attack-detecting scheme for IOT. Here, a system using blockchain is employed to process information in a logical sequence. At first, enhanced data normalization is deployed. Then, the features like “Exponential Moving Average, improved Holoentropy, higher-order statistical features, and mutual information” are extracted. Then feature selection is done via Improved Principal Component Analysis. The selected traits are evaluated “hybrid classification that includes Gated Recurrent Unit and Deep Belief Network” to discover the attacks in IoT. Finally, the advancements of this work are established over others. The proposed HC(GRU + DBN) scheme has attained higher accuracy of 97.2% for the best case of 92%,91%,94%,84%, and 92% to current schemes like SVM, RF, RNN, LSTM, and NB.

A novel intrusion detection system in cloud infrastructure using deep learning technique

One of the business strategies for selling computer resources with services and technology for better use of computing infrastructures is Cloud computing (CC). Nowadays, every IT company prefers cloud computing because it provides consumers with flexible, pay-per-use services. Due to its open and distributed structure, which is susceptible to attackers, thereby, privacy and security is a key obstacle to its sustainability. The most prevalent approach for detecting assaults on the cloud is known to be Intrusion Detection System (IDS). This article aims to propose a novel intrusion pattern detection system (IPDS) in cloud computing that includes three stages: (1) pre-processing, (2) feature extraction, and (3) classification. At first, pre-processing is performed on the input data via Z-score normalization and then feature extraction is performed along with statistical and higher-order statistical features. Subsequently, the extracted features are given to the classification phases that use the Optimized Quantum Neural Network (QNN) classifier. The hidden neuron optimization is performed by Cubic Chaotic Map integrated Cat and Mouse Based Optimization (CC-CMBO) Algorithm to make the classification more exact. Finally, the results of the proposed work are assessed to those of standard systems with respect to various measures.

Automatic Modulation Classification Using a Support Vector Machine-Based Pattern Recognition Algorithm

Modulation format recognition is an essential part of intelligent receivers of wireless communication systems, especially for adaptive radio systems (ARS). This paper presents a detailed investigation of automatic modulation classification (AMC) using pattern recognition classifiers (PRC) under fading and AWGN conditions. A variety of classifiers with different kernel functions and Support Vector Machine (SVM) classifiers have been developed for the classification of higher-order digital modulation signals. In addition, an extensive investigation of the extraction of various higher-order statistical features from each of the modulated classes and the choice of appropriate features for training classifiers are presented. In addition, the performance of the SVM classifier is evaluated under a variety of training rates and suboptimal channel conditions. Further, the performance of SVM classifiers is compared to that of existing techniques to demonstrate the effectiveness of the SVM classifiers for modulation categorization.

Attack Detection in Wireless Sensor Network: A Big Data Perspective

For securing the network, intrusion detection systems are frequently used in wireless sensor networks to fight against insider attacks by adopting the appropriate trust-based methods. Still, the sensors could create an enormous amount of data that reduces the efficacy of trust computation in the big data era. This paper aims to introduce a new attack detection system under the big data perspective. Originally, the input data is fed the preprocessing phase, in which the normalization process takes place. Further, the MapReduce framework is used to handle the bulk data by reducing it. The preprocessed data is subjected to extract the features, where it extracts the raw features, statistical features, and higher-order statistical features. Here, the feature selection is done by chi-square ranking and info-gain ranking. Further, these features are provided as the input to the classification phase, where multilayer perception (MLP) is used for detecting the presence of an attack. For making the classification more precise, weights of MLP are optimally tuned by proposed slap swarm updated sparrow optimized algorithm which integrated the concept of sparrow search algorithm and salp swarm algorithm. Finally, the performance of presented scheme is calculated to existing approaches under different metrics.

Decision Tree with Pearson Correlation-based Recursive Feature Elimination Model for Attack Detection in IoT Environment

The industrial revolution in recent years made massive uses of Internet of Things (IoT) applications like smart cities’ growth. This leads to automation in real-time applications to make human life easier. These IoT-enabled applications, technologies, and communications enhance the quality of life, quality of service, people’s well-being, and operational efficiency. The efficiency of these smart devices may harm the end-users, misuse their sensitive information increase cyber-attacks and threats. This smart city expansion is difficult due to cyber attacks. Consequently, it is needed to develop an efficient system model that can protect IoT devices from attacks and threats. To enhance product safety and security, the IoT-enabled applications should be monitored in real-time. This paper proposed an efficient feature selection with a feature fusion technique for the detection of intruders in IoT. The input IoT data is subjected to preprocessing to enhance the data. From the preprocessed data, the higher-order statistical features are selected using the proposed Decision tree-based Pearson Correlation Recursive Feature Elimination (DT-PCRFE) model. This method efficiently eliminates the redundant and uncorrelated features which will increase resource utilization and reduces the time complexity of the system. Then, the request from IoT devices is converted into word embedding using the feature fusion model to enhance the system robustness. Finally, a Deep Neural network (DNN) has been used to detect malicious attacks with the selected features. This proposed model experiments with the BoT-IoT dataset and the result shows the proposed model efficiency which outperforms other existing models with the accuracy of 99.2%.

Hybrid classifier strategy with tuned training weights for distributed denial of service attack detection

SummaryThe 5G wireless networks associated with higher data‐transferring speeds considerably affect the performance of IoT networks. Nowadays, the Internet has become a very significant aspect of human lives, and it aids in data transfers, processing, and storing. However, 5G networks are subjected to varied cyber security attacks, which are hard to detect. As a result, it is required to set up attack detection models that can recognize 5G network's distributed denial of service (DDoS) attack. Thereby, this article aims to introduce a new model for detecting DDoS attacks. Initially, from input data, features such as statistical features, improved exponential moving average, higher order statistical features, MI, and improved correlation based features are derived. Further, a gain ratio ranking model is used for picking fine features from the overall derived features. Finally, at the detection stage, bidirectional long short‐term memory as well as optimized deep belief network (DBN) are introduced that portray the detected results in a precise way. DBN weights get fine‐tuned by the combined shark smell and electric fish model with new distance based active eco‐location model. Finally, the created approach's advantages are demonstrated using a variety of metrics, including precision, accuracy, F‐measure, and others. In particular, the performance metrics of the proposed work at node = 6000 are accuracy = 98.59%, sensitivity = 96.82%, specificity = 97.52%, precision = 98.91%, F‐measure = 97.86%, MCC = 92.19%, NPV = 97.24%, FPR = 2.47%, and FNR = 3.17%, respectively.

DeepBP: A bilinear model integrating multi-order statistics for fine-grained recognition

The bilinear model using second-order statistical features is an important weakly supervised method for fine-grained recognition. Based on this, fusing higher-order statistical features to obtain more discriminant features is an effective approach for improving the performance of the model. However, the existing framework is difficult to fuse higher-order features due to a sharp increase in the number of parameters caused by the increase in fusion order. To address the issue, this paper proposes a DeepBP model composed of a deep network module and a bilinear pooling module. The bilinear module explicitly captures low-order statistical features, and the deep network module implicitly learns high-order features. The two modules are integrated to achieve multi-level information integration. To verify the model's ability, experiments are conducted on the CUB, Cars, and Aircrafts datasets, and the accuracy of 85.6%, 91.6%, and 88.6% is achieved, respectively.

Modulation Format Identification Based on Signal Constellation Diagrams and Support Vector Machine

In coherent optical communication systems, where multiple modulation formats are mixed and variable, the correct identification of signal modulation formats provides the foundation for subsequent performance improvement using digital algorithms. A modulation format identification (MFI) scheme based on signal constellation diagrams and support vector machine (SVM) is proposed. Firstly, the signal constellation diagrams are divided by the fractal dimension of the weighted linear least squares (WLS-FD) algorithm, and the fractal dimension (FD) in each region is calculated, which is regarded as one of the image features. Then, the feature values of the image in different directions are extracted by the gray-level co-occurrence matrix (GLCM), and their mean and variance are calculated, which is regarded as another feature. Finally, the two features are input into the modulation format classifier constructed by the SVM to achieve MFI in coherent optical communication systems. To verify the feasibility and superiority of the scheme, we compare it with the MFI scheme based on higher-order statistical (HOS) features, GLCM features, and FD features, respectively. Further, we built a 30 GBaud coherent optical communication system with fiber lengths of 80 km and 120 km, where the optical signal-to-noise ratio (OSNR) ranges from 0 dB to 30 dB. The proposed MFI scheme identifies seven modulation formats: QPSK, 8QAM, 16QAM, 32QAM, 64QAM, 128QAM, and 256QAM. The results show that compared with the other three schemes, our proposed scheme has a better identification accuracy at low OSNR. In addition, the identification accuracy of this scheme can reach 100% when the OSNR ≥ 10 dB.

Optimal prediction of viral host from genomic datasets using ensemble classifier

Viruses are common biological agents that are supposed to be the world's greatest repositories of undiscovered genetic diversity. One of the common problems in bioinformatics is gene-disease prediction. Techniques for taxonomic classification, host range, and biological properties of newly discovered viruses are needed for complete functional characterization and annotation. Understanding the behaviors as well as interactions of microbial populations needs research into virus-host infectious associations. The following three main steps of an unique viral host prediction method using genomic datasets are introduced in this research work: “(a) Pre-processing, (b) Feature extraction, and (c) Prediction phase”. In starting stage, raw genomic datasets are exposed to pre-processing, which would include data cleaning activities. The pre-processed data is then used to extract the statistical features, high order statistical features, weighted holoentropy, chi-squared features, relief-based features, and symmetric uncertainty-based features. The ensemble approach, which uses the Support Vector Machine (SVM), Neural Network (NN), Random Forest (RF), and Convolutional Neural Network (CNN), respectively, is then employed for the prediction. Here, SVM, NN, and RF classifiers are fed the retrieved features as input. These classifiers' outputs will be fed into an optimised CNN, which produces the final prediction outcome. Additionally, the Adaptive Flower Pollination Algorithm (AFPA), an upgraded variant of the conventional Flower Pollination Protocol, is used to fine-tune the weights of optimised CNN in order to increase prediction accuracy (FPA). The AFPA+EC's F1-Score is 0.75026, which is correspondingly 61%, 52.4%, 20.2%, and 26.5% better than the conventional methods SVM, KNN, RF, and CNN.

Software Defects Prediction Model with Self Improved Optimization

Software systems have significantly grown and increased its complexity to unprecedented levels. Because of these characteristics, preventing software faults is extremely difficult. Therefore, automatic forecasting of errors is required, and it might assist developers deploy with limited resources more efficiently. Different methods on identifying and correcting these flaws at low cost were offered, which, significantly improves the effectiveness of the techniques. This work includes 4 steps to offer a new SDP model. The input data is preprocessed and from that, the “statistical features, raw features, higher order statistical features and proposed MI and entropy features” are extracted. Then, feature selection is done and appropriate features are elected via chi-square scheme. The elected features are detected via LSTM and DBN to predict the defects. The weights of LSTM and DBN are optimized by Opposite Behavior Learning Integrated SDO (OBLI-SDO) algorithm. Finally, examination is done to prove the betterment of OBLI-SDO.