A convolutional neural network‐based comparative study for pepper seed classification: Analysis of selected deep features with support vector machine

Journal of Food Process Engineering, Journal Year: 2021, Volume and Issue: 45(6)

Published: Dec. 27, 2021

Abstract The seeds of high quality are very important for the cultivation pepper. required practices and growing conditions may be affected by cultivar. Also, productivity properties pepper depend on selection appropriate seed cultivars necessary breeding programs. cultivar differentiation tested human eye. However, small sizes visual similarities make it difficult to distinguish between cultivars. Computer vision artificial intelligence can provide discrimination accuracy procedures objective fast. This study aimed classify belonging different with convolutional neural network (CNN) models. were obtained from green, orange, red, yellow A flatbed scanner was used acquire images. After image acquisition, procedure applied preprocessing images, data augmentation using techniques then deep learning‐based classification. Two approaches have been proposed In first approach, CNN models (ResNet18 ResNet50) trained seeds. second first, features pretrained fused, feature fused features. Classification all selected performed support vector machine (SVM) kernel functions (Linear, Quadratic, Cubic, Gaussian). accuracies in approximation 98.05% 97.07% ResNet50 ResNet18, respectively. CNN‐SVM‐Cubic achieved up 99.02% Practical applications precision agriculture, is that same type purification standardization crop culture. Performing this classification manually assistance will result subjective, slow, low standard outcomes. To overcome such problems, supported systems emerges as an tool. study, a highly successful system presented according characteristics preferred practice identifying detecting falsification or ensuring their reliability. It prevent mixing attributes processing.

Language: Английский

---

Big data and artificial intelligence (AI) methodologies for computer-aided drug design (CADD)

Biochemical Society Transactions, Journal Year: 2022, Volume and Issue: 50(1), P. 241 - 252

Published: Jan. 25, 2022

There have been numerous advances in the development of computational and statistical methods applications big data artificial intelligence (AI) techniques for computer-aided drug design (CADD). Drug is a costly laborious process considering biological complexity diseases. To effectively efficiently develop new drug, CADD can be used to apply cutting-edge various limitations field. Data pre-processing approaches, which clean raw consistent reproducible AI are introduced. We include current status applicability areas such as identification binding sites target proteins, structure-based virtual screening (SBVS), absorption, distribution, metabolism, excretion toxicity (ADMET) property prediction. enable accurate comprehensive analysis massive biomedical predictive models field design. Understanding analyzing biological, chemical, or pharmaceutical architectures entities related will provide beneficial information era.

Language: Английский

---

A Comprehensive Survey on the Process, Methods, Evaluation, and Challenges of Feature Selection

IEEE Access, Journal Year: 2022, Volume and Issue: 10, P. 99595 - 99632

Published: Jan. 1, 2022

Feature selection is employed to reduce feature dimensions and computational complexity by eliminating irrelevant redundant features. A vast amount of increasing data its processing generate many sets, that are reduced the process improve performance in all sorts classification, regression, clustering models. This research performs a detailed analysis motivation concentrates on fundamental architecture selection. The study aims establish structured formation related popular methods such as filter, wrapper, embedded into search strategies, evaluation criteria, learning methods. Different organize comparison benefits drawbacks followed multiple classification algorithms standard validation measures. diversity applications domains retrieval, prediction analysis, medical, intrusion, industrial efficiently highlighted. focused some additional for handling big data. Nonetheless, new challenges have surfaced data, which also addressed this study. Reflecting commonly encountered clarifying how obtain absolute method significant components

Language: Английский

---

Enhanced text classification through an improved discrete laying chicken algorithm

Expert Systems, Journal Year: 2024, Volume and Issue: 41(8)

Published: Jan. 25, 2024

Abstract The exponential growth of digital text documents presents a significant challenge for classification algorithms, as the vast number words in each document can hinder their efficiency. Feature selection (FS) is crucial technique that aims to eliminate irrelevant features and enhance accuracy. In this study, we propose an improved version discrete laying chicken algorithm (IDLCA) utilizes noun‐based filtering reduce improve performance. Although LCA newly proposed algorithm, it has not been systematically applied problems before. Our enhanced employs different operators both exploration exploitation find better solutions mode. To evaluate effectiveness method, compared with some conventional nature‐inspired feature methods using various learning models such decision trees (DT), K‐nearest neighbor (KNN), Naive Bayes (NB), support vector machine (SVM) on five benchmark datasets three evaluation metrics. experimental results demonstrate comparison existing one. code available at https://github.com/m0javad/Improved-Discrete-Laying-Chicken-Algorithm .

Language: Английский

---

A review on advancements in feature selection and feature extraction for high-dimensional NGS data analysis

Functional & Integrative Genomics, Journal Year: 2024, Volume and Issue: 24(5)

Published: Aug. 19, 2024

Language: Английский

---