Training Accuracy Research Articles

Comparison of Activation Function Performance in the Resnet Algorithm for Rice Type Classification

Quality checking of rice seed varieties (Oryza sativa) is an important procedure for quality assessment in the agricultural sector. The application of transfer learning algorithms has shown good results in image recognition tasks, so this algorithm is suitable for classifying rice variety images automatically. The data classes to be analyzed are Arborio, Basmati, Ipsala, Jasmine and Karacadag based on morphological, shape and color features analysis using the ResNet algorithm. The experiment used three types of models, namely ResNet-TopHat-ReLU, ResNet-TopHat-LeakyReLU and ResNet-TopHat-eLU. The ResNet-TopHat-eLU model is the best model with training accuracy of 96.61%, validation accuracy of 95.12% and testing accuracy of 78.17%.

Improving river water quality prediction with hybrid machine learning and temporal analysis

River systems provide multiple ecosystem services to society globally, but these are already degraded or threatened in many areas of the world due to water quality issues linked to diffuse and point-source pollutant inputs. Water quality evaluation is essential to develop remediation and management strategies. Computational tools such as machine learning based predictive models have been developed to improve monitoring network capabilities. The model's performance is reduced when datasets composed of reductant information are used for training, on the other hand, the selection of most representative and variable water quality scenarios could result in higher precision. This study analyzed historical water quality behavior in the Santiago River, Mexico, to identify the most variable and representative data available to train machine learning models (Adaptive Neuro Fuzzy Inference System – ANFIS, Artificial Neural Network – ANN, and Support Vector Machine - SVM). Thirteen monitoring sites were clustered according to their water quality variability from 2009 to 2022. Subsequently, a Time Series Analysis (TSA) was used to select the most representative monitoring station from each cluster. Data for 6/13 monitoring sites were retained for the Best Training Subset (BTS) used to train restricted models that performed with similar (ANN and SMV) or higher (ANFIS) prediction accuracy (in terms of RMSE, MAE, MSE and R2) for both training and testing. This study provides evidence of water quality data containing redundant information that is not useful to improve machine learning model performance, in turn leading to overtraining. Combined analytical approaches can maximize the representativeness and variability of data selected for machine learning applications, leading to improved prediction.

Subjective brain fog: a four-dimensional characterization in 25,796 participants.

Brain fog is associated with significant morbidity and reduced productivity and gained increasing attention after COVID-19. However, this subjective state has not been systematically characterised. To characterise self-reported brain fog. We systematically studied the cross-sectional associations between 29 a priori variables with the presence of "brain fog." The variables were grouped into four categories: demographics, symptoms and functional impairments, comorbidities and potential risk factors (including lifestyle factors), and cognitive score. Univariate methods determined the correlates of brain fog, with long-COVID and non-long-COVID subgroups. XGBoost machine learning model retrospectively characterised subjective brain fog. Bonferroni-corrected statistical significance was set at 5%. Digital application for remote data collection. 25,796 individuals over the age of 18 who downloaded and completed the application. 7,280 of 25,796 individuals (28.2%) reported experiencing brain fog, who were generally older (mean brain fog 35.7 ± 11.9 years vs. 32.8 ± 11.6 years, p < 0.0001) and more likely to be female (OR = 1.2, p < 0.001). Associated symptoms and functional impairments included difficulty focusing or concentrating (OR = 3.3), feeling irritable (OR = 1.6), difficulty relaxing (OR = 1.2, all p < 0.0001), difficulty following conversations (OR = 2.2), remembering appointments (OR = 1.9), completing paperwork and performing mental arithmetic (ORs = 1.8, all p < 0.0001). Comorbidities included long-COVID-19 (OR = 3.8, p < 0.0001), concussions (OR = 2.4, p < 0.0001), and higher migraine disability assessment scores (MIDAS) (+34.1%, all p < 0.0001). Cognitive scores were marginally lower with brain fog (-0.1 std., p < 0.001). XGBoost achieved a training accuracy of 85% with cross-validated accuracy of 74%, and the features most predictive of brain fog in the model were difficulty focusing and following conversations, long-COVID, and severity of migraines. This is the largest study characterising subjective brain fog as an impairment of concentration associated with functional impairments in activities of daily living. Brain fog was particularly associated with a history of long-COVID-19, migraines, concussion, and with 0.1 standard deviations lower cognitive scores, especially on modified Stroop testing, suggesting impairments in the ability to inhibit cognitive interference. Further prospective studies in unselected brain fog sufferers should explore the full spectrum of brain fog symptoms to differentiate it from its associated conditions.

Convolutional neural networks to classify human stress that occurs during in‐field sugarcane harvesting: A case study

AbstractAssessing human stress in agriculture proves to be a complex and time‐intensive endeavor within the field of ergonomics, particularly for the development of agricultural systems. This methodology involves the utilization of instrumentation and the establishment of a dedicated laboratory setup. The complexity arises from the need to capture and analyze various physiological and psychological indicators, such as heart rate (HR), muscle activity, and subjective feedback to comprehensively assess the impact of farm operations on subjects. The instrumentation typically includes wearable devices, sensors, and monitoring equipment to gather real‐time data of subject during the performance of farm operations. Deep learning (DL) models currently achieve human performance levels on real‐world face recognition tasks. In this study, we went beyond face recognition and experimented with the recognition of human stress based on facial features during the drudgery‐prone agricultural operation of sugarcane harvesting. This is the first research study for deploying artificial intelligence‐driven DL techniques to identify human stress in agriculture instead of monitoring several ergonomic characteristics. A total of 20 (10 each for male and female) subjects comprising 4300 augmented RGB images (215 per subject) were acquired during sugarcane harvesting seasons and then these images were deployed for training (80%) and validation (20%). Human stress and nonstress states were determined based on four ergonomic physiological parameters: heart rate (ΔHR), oxygen consumption rate (OCR), energy expenditure rate (EER), and acceptable workload (AWL). Stress was defined when ΔHR, OCR, EER, and AWL reached or exceeded certain standard threshold values. Four convolutional neural network‐based DL models (1) DarkNet53, (2) InceptionV3, (3) MobileNetV2 and (4) ResNet50 were selected due to their remarkable feature extraction abilities, simple and effective implementation to edge computation devices. In all four DL models, training performance results delivered training accuracy ranging from 73.8% to 99.1% at combinations of two mini‐batch sizes and four levels of epochs. The maximum training accuracies were 99.1%, 99.0%, 97.7%, and 95.4% at the combination of mini‐batch size 16 and 25 epochs for DarkNet53, InceptionV3, ResNet50, and MobileNetV2, respectively. Due to the best performance, DarkNet53 was tested further on an independent data set of 100 images and found 89.8%–93.3% confident to classify stressed images for female subjects while 92.2%–94.5% for male subjects, though it was trained on the integrated data set. The comparative classification of the developed model and ergonomic measurements for stress classification was carried out with a net accuracy of 88% where there were few instances of wrong classifications.

A gradient-enhanced physics-informed neural networks method for the wave equation

Physics-informed neural networks (PINNs) have been proven to be a useful tool for solving general partial differential equations (PDEs), which is meshless and dimensionally free compared with traditional numerical solvers. Based on PINNs, gradient-enhanced physics-informed neural networks (gPINNs) add the partial derivative loss term of the independent variable and the physical constraint term, which improves the accuracy of network training. In this paper, the gPINNs method is proposed to solve the wave problem. The equation boundary value of the wave problem is added in the network construction, thus the network is forced to satisfy the equation boundary value during each training process. Two examples of wave equations are given in our paper, which are solved numerically by PINNs and gPINNs, respectively. It is found that using fewer data sets, gPINNs can learn data features more fully than PINNs and better results of gPINNs can be obtained.

FAKE PROFILE DETECTION ON SOCIAL NETWORKING WEBSITES USING MACHINE LEARNING

In an age where social media has become an integral part of our lives, the challenge of detecting fake accounts on platforms like Instagram has gained significant importance. This project, titled "Instagram Fake Account Detection using Machine Learning," employs Python as its primary tool to tackle this problem. It leverages two powerful machine learning algorithms, the Random Forest Classifier and the Decision Tree Classifier, to accomplish this task. The Random Forest Classifier demonstrates remarkable performance, achieving a 100% accuracy on the training dataset and an impressive 93% accuracy on the test dataset. Meanwhile, the Decision Tree Classifier exhibits its effectiveness with a training accuracy of 92% and a test accuracy of 92%. The dataset employed in this project is composed of 576 records, each characterized by 12 distinct features. These features encompass critical aspects of Instagram profiles, including the presence of a profile picture, the ratio of numerical characters in usernames, the breakdown of full names into word tokens, the ratio of numerical characters in full names, the equality between usernames and full names, the length of user bios, the existence of external URLs, the privacy status of accounts, the number of posts, the count of followers, the number of accounts followed, and the ultimate classification of an account as "Fake" or "Not." By harnessing the capabilities of Python and these advanced machine learning models, this project endeavors to provide a robust and efficient solution for the identification of fake Instagram accounts. In doing so, it contributes to the preservation of the platform's integrity and the security of its users.

Fake Profile Detection on Social Networking Websites Using Machine Learning

In an age where social media has become an integral part of our lives, the challenge of detecting fake accounts on platforms like Instagram has gained significant importance. This project, titled "Instagram Fake Account Detection using Machine Learning," employs Python as its primary tool to tackle this problem. It leverages two powerful machine learning algorithms, the Random Forest Classifier and the Decision Tree Classifier, to accomplish this task. The Random Forest Classifier demonstrates remarkable performance, achieving a 100% accuracy on the training dataset and an impressive 93% accuracy on the test dataset. Meanwhile, the Decision Tree Classifier exhibits its effectiveness with a training accuracy of 92% and a test accuracy of 92%. The dataset employed in this project is composed of 576 records, each characterized by 12 distinct features. These features encompass critical aspects of Instagram profiles, including the presence of a profile picture, the ratio of numerical characters in usernames, the breakdown of full names into word tokens, the ratio of numerical characters in full names, the equality between usernames and full names, the length of user bios, the existence of external URLs, the privacy status of accounts, the number of posts, the count of followers, the number of accounts followed, and the ultimate classification of an account as "Fake" or "Not." By harnessing the capabilities of Python and these advanced machine learning models, this project endeavors to provide a robust and efficient solution for the identification of fake Instagram accounts. In doing so, it contributes to the preservation of the platform's integrity and the security of its users.

Pioneering Data Processing for Convolutional Neural Networks to Enhance the Diagnostic Accuracy of Traditional Chinese Medicine Pulse Diagnosis for Diabetes.

Traditional Chinese medicine (TCM) has relied on pulse diagnosis as a cornerstone of healthcare assessment for thousands of years. Despite its long history and widespread use, TCM pulse diagnosis has faced challenges in terms of diagnostic accuracy and consistency due to its dependence on subjective interpretation and theoretical analysis. This study introduces an approach to enhance the accuracy of TCM pulse diagnosis for diabetes by leveraging the power of deep learning algorithms, specifically LeNet and ResNet models, for pulse waveform analysis. LeNet and ResNet models were applied to analyze TCM pulse waveforms using a diverse dataset comprising both healthy individuals and patients with diabetes. The integration of these advanced algorithms with modern TCM pulse measurement instruments shows great promise in reducing practitioner-dependent variability and improving the reliability of diagnoses. This research bridges the gap between ancient wisdom and cutting-edge technology in healthcare. LeNet-F, incorporating special feature extraction of a pulse based on TMC, showed improved training and test accuracies (73% and 67%, respectively, compared with LeNet's 70% and 65%). Moreover, ResNet models consistently outperformed LeNet, with ResNet18-F achieving the highest accuracy (82%) in training and 74% in testing. The advanced preprocessing techniques and additional features contribute significantly to ResNet18-F's superior performance, indicating the importance of feature engineering strategies. Furthermore, the study identifies potential avenues for future research, including optimizing preprocessing techniques to handle pulse waveform variations and noise levels, integrating additional time-frequency domain features, developing domain-specific feature selection algorithms, and expanding the scope to other diseases. These advancements aim to refine traditional Chinese medicine pulse diagnosis, enhancing its accuracy and reliability while integrating it into modern technology for more effective healthcare approaches.

Channel Adaptive and Sparsity Personalized Federated Learning for Privacy Protection in Smart Healthcare Systems.

With the booming development of Smart Healthcare Systems (SHSs), employing federated learning (FL) in SHS devices has become a research hotspot. FL, as a distributed learning framework, can train models without sharing the original data among users, and then protect the user privacy. Existing research has proposed many methods to improve the security and efficiency of FL, which may not fully consider the characteristics of SHSs. Specifically, the requirements of privacy protection and efficiency pose significant challenges to FL. Current studies have struggled to balance privacy security and efficiency, and the degradation of model training efficiency in SHSs can be critical to patient health. Therefore, to improve the privacy protection of healthcare data and ensure communication efficiency, this work proposes a novel personalized FL framework based on Communication quality and Adaptive Sparsification (pFedCAS). In order to achieve privacy protection, a control unit is proposed and introduced to adjust the sparsity of the local model adaptively. To further improve the training efficiency, a selection unit is added during global model aggregation to select suitable clients for parameter updates. Finally, we validate the proposed method operated on the HAM10000 dataset. Simulation results validate that pFedCAS can not only improve privacy protection, but also gain an improvement of 15% in training accuracy and a reduction of 30% in training costs based on communication quality. The simulation results also validate the excellent robustness of pFedCAS to non-iid data.

IoMT-Based Smart Healthcare Detection System Driven by Quantum Blockchain and Quantum Neural Network.

Electrocardiogram (ECG) is the main criterion for arrhythmia detection. As a means of identification, ECG leakage seems to be a common occurrence due to the development of the Internet of Medical Things. The advent of the quantum era makes it difficult for classical blockchain technology to provide security for ECG data storage. Therefore, from the perspective of safety and practicality, this article proposes a quantum arrhythmia detection system called QADS, which achieves secure storage and sharing of ECG data based on quantum blockchain technology. Furthermore, a quantum neural network is used in QADS to recognize abnormal ECG data, which contributes to further cardiovascular disease diagnosis. Each quantum block stores the hash of the current and previous block to construct a quantum block network. The new quantum blockchain algorithm introduces a controlled quantum walk hash function and a quantum authentication protocol to guarantee legitimacy and security while creating new blocks. In addition, this article constructs a hybrid quantum convolutional neural network called HQCNN to extract the temporal features of ECG to detect abnormal heartbeats. The simulation experimental results show that HQCNN achieves an average training and testing accuracy of 94.7% and 93.6%. And the detection stability is much higher than classical CNN with the same structure. HQCNN also has certain robustness under the perturbation of quantum noise. Besides, this article demonstrates through mathematical analysis that the proposed quantum blockchain algorithm has strong security and can effectively resist various quantum attacks, such as external attacks, Entanglement-Measure attack and Interception-Measurement-Repeat attack.

A new deep learning model with interface for fine needle aspiration cytology image-based breast cancer detection

Cytological evaluation through microscopic image analysis of fine needle aspiration cytology (FNAC) is pivotal in the initial screening of breast cancer. The sensitivity of FNAC as a screening tool relies on both image quality and the pathologist’s expertise. To enhance diagnostic accuracy and alleviate the pathologist’s workload, a computer-aided diagnosis (CAD) system was developed. A comparative study was conducted, assessing twelve candidate pre-trained models. Utilizing a locally gathered FNAC image dataset, three superior models-MobileNet-V2, DenseNet-121, and Inception-V3-were selected based on their training, validation, and testing accuracies. Further, these models underwent evaluation in four transfer learning scenarios to enhance testing accuracy. While the outcomes were promising, they left room for improvement, motivating us to create a novel deep convolutional neural network (CNN). The newly proposed model exhibited robust performance with testing accuracy at 85%. Our research concludes that the most lightweight, high-accuracy model is the one we propose. We’ve integrated it into our user-friendly Android App, “Breast Cancer Detection System,” in TensorFlow Lite format, with cloud database support, showcasing its effectiveness. Implementing an artificial intelligent (AI)-based diagnosis system with a user-friendly interface holds the potential to enhance early breast cancer detection using FNAC.

Enhancing intrusion detection system using rectified linear unit function in pigeon inspired optimization algorithm

The increasing rate of cybercrime in the digital world highlights the importance of having a reliable intrusion detection system (IDS) to detect unauthorized attacks and notify administrators. IDS can leverage machine learning techniques to identify patterns of attacks and provide real-time notifications. In building a successful IDS, selecting the right features is crucial as it determines the accuracy of the predictions made by the model. This paper presents a new IDS algorithm that combines the rectified linear unit (ReLU) activation function with a pigeon-inspired optimizer in feature selection. The proposed algorithm was evaluated on network security layer - knowledge discovery in databases (NSL-KDD) datasets and demonstrated improved performance in terms of training speed and accuracy compared to previous IDS models. Thus, the use of the ReLU activation function and a pigeon-inspired optimizer in feature selection can significantly enhance the effectiveness of an IDS in detecting unauthorized attacks.

Analysis of the possibility of using the satellite navigation system in the railway traffic control systems

The analysis of the possibility of using satellite navigation for rail vehicle positioning, which is the subject of many studies and projects, is aimed at assessing the uncertainty and accuracy of train positioning using satellite navigation systems, along with defining its impact on the capacity and safety of the rail infrastructure. Existing studies in this area prevent a clear assessment of the feasibility of implementing a satellite navigation system for rail vehicle positioning in rail traffic control systems. Therefore, it is necessary to analyze the possibility of using satellite navigation for rail vehicle positioning in terms of research problems arising from the limitations of satellite positioning system functionality, which include parameters related to accuracy, reliability, and speed of information transfer between the vehicle and the rail traffic control system. The implementation of the research work should be focused on developing a research model, subject to validation for the parameters and conditions of movement of rail vehicles in the defined area of railroad infrastructure, adopted within the framework of the work carried out. The research and the model form the basis for developing the knowledge of implementing satellite navigation in practical solutions on the railroad network.

Discrimination of maize transgenic and non-transgenic varieties by laser induced spectroscopy (LIBS) and machine learning algorithms

BackgroundIn the last decades, the production and consumption of genetically modified agricultural products have increased markedly due to the worldwide population growth and rising demand for food and feed. Consequently, genetically modified crops have been extensively produced and consumed, which required identifying and discriminating transgenic and non-transgenic products. ResultsLaser-induced breakdown spectroscopy (LIBS) combined with chemometric methods was applied to identify and discriminate two varieties of conventional (not-genetically modified, NGM) maize from four varieties of transgenic maize (genetically modified, GM). The LIBS spectra acquired under reduced pressure (100 Torr) conditions over two ranges, i.e., 175–330 nm and 275–770 nm, were subjected to Standard Normal Variation (SNV) and multivariate methods such as Principal Component Analysis (PCA) to reduce data matrices dimensionality and spectral noise. The supervised machine learning algorithms k-nearest neighbor (k-NN) and support vector machine (SVM) have been applied to discriminate among NGM and GM maize reserving 25 % of data for external validation. The training data were employed for hyperparameter optimization of classifiers using the Leave-One-Out Cross-Validation (LOOCV) method. Considering all six maize varieties simultaneously, the highest training accuracy achieved was 90.56 %, with an external validation accuracy of 88.33 %. In an alternative approach based on pairwise combinations of one GM variety against one NGM variety, the best outcome achieved was 100 % LOOCV and external validation accuracy. ConclusionsThese results showed that LIBS supported by appropriate chemometric methods represents an alternative screening technique for identifying and discriminating transgenic from non-transgenic maize.

AdaBoost‐powered cloud of things framework for low‐latency, energy‐efficient chronic kidney disease prediction

AbstractThe United Nations' sustainable development agenda has set an ambitious goal of reducing premature mortality from non‐communicable diseases by 33% by 2030. Among these diseases, chronic kidney disease (CKD) is a significant contributor to both morbidity and mortality. Integrating the Internet of Things (IoT) and cloud computing in healthcare has gained momentum, particularly in remote patient monitoring. However, it is essential to acknowledge that cloud computing has limitations, particularly in handling vast volumes of Big Data, mainly due to scalability and latency concerns. This article proposes a novel framework, AdaBoostCoTCKD, to mitigate latency issues, minimize response times, reduce power consumption, and optimize network resources for predicting CKD. The framework leverages the synergy between the AdaBoost machine learning technique and fog computing paradigms to enhance the precision and efficiency of CKD prediction methods. In addition, it introduces an auxiliary cloud‐based database, enriching the pool of future insights and facilitating prospective database infrastructure expansions. This augmentation is expected to impact predictive accuracy positively. We conducted comprehensive experiments to demonstrate the effectiveness of our approach. Our model achieved an impressive training accuracy of 99.928% and testing accuracy of 99.975%, while the fog environment reduced latency by 31% and energy consumption by 75% compared to traditional cloud‐based solutions. Our proposed system enables early CKD detection and offers advantages over cloud‐only solutions, providing a robust and efficient platform for healthcare IoT applications with significant clinical value. These promising results underscore the potential of combining fog computing and the AdaBoost machine learning technique to advance healthcare by addressing latency, response time, power consumption, and network resource optimization challenges in CKD prediction.

Classification of nutmeg ripeness using artificial intelligence

Nutmeg seeds can produce a lot of oil if they have optimal maturity, in other words, they have little moisture content. Based on observations I made at one of the refineries in Sukabumi, farmers do not pay attention to the maturity level of nutmeg seeds after drying which can cause a decrease in the quality of nutmeg seeds and the quality of the oil produced. This study aims to make it easier for nutmeg farmers to classify the maturity of nutmeg seeds. This study used the Convolutional Neural Network (CNN) method to help with classification problems and several image processing methods. This program will be run through an android application. The results of CNN model training accuracy are 97.92%. Thus, it can be concluded that the design and testing of a model to classify the maturity level of nutmeg seeds using artificial intelligence and the implementation of the model into an android application has been successfully carried out.

A Two-Stage Differential Privacy Scheme for Federated Learning Based on Edge Intelligence.

The issue of data privacy protection must be considered in distributed federated learning (FL) so as to ensure that sensitive information is not leaked. In this article, we propose a two-stage differential privacy (DP) framework for FL based on edge intelligence. Various levels of privacy preservation can be provided according to the degree of data sensitivity. In the first stage, the randomized response mechanism is used to perturb the original feature data by the user terminal for data desensitization, and the user can self-regulate the level of privacy preservation. In the second stage, noise is added to the local models by the edge server to further guarantee the privacy of the models. Finally, the model updates are aggregated in the cloud. In order to evaluate the performance of the proposed end-edge-cloud FL framework in terms of training accuracy and convergence, extensive experiments are conducted on a real electrocardiogram (ECG) signal dataset. Bi-directional long-short-term memory (BiLSTM) neural network is adopted to training classification model. The effect of different combinations of feature perturbation and noise addition on the model accuracy is analyzed depending on different privacy budgets and parameters. The experimental results demonstrate that the proposed privacy-preserving framework provides good accuracy and convergence while ensuring privacy.

Evaluating the Efficacy of the 3D U-Net Architecture For Glioblastoma Multiforme Tumor Segmentation

Glioblastoma is the deadliest form of brain cancer which begins as a congregation of cancerous cells within the brain but then progresses into invading and destroying healthy brain tissue. Radiation therapy, the most popular treatment option, is where neuro-oncologists apply intense radiation energy beams directly on the tumor region to kill the cancerous cells. However, for radiation therapy to be effective, the segmented magnetic resonance image (MRI) in which the oncologists base where to apply the radiation must be segmented nearly 100% accurately, or else the energy beams will mistakenly damage healthy brain tissue. Additionally, radiation therapy will be futile if the MRI segmentation is not complete in time for the therapy appointment. My research problem is evaluating the performance of notable segmentation models for accurate and immediate 3D segmentation of glioblastoma multiforme brain tumors from MRIs while focusing on one particular architecture, the 3D U-Net. Each of the models was given 850 MRIs from the same dataset called BraTS, which is an annual competition hosted by the University of Pennsylvania. My 3D U-Net model ended up having the highest training (0.9928), validation (0.968) and testing (0.9867) accuracies of all the models and took approximately 55 seconds to predict segmentations of the tumor region. The results reveal that the 3D U-Net model is capable of automating glioblastoma tumor segmentation in significantly fewer hours than a human oncologist would take, all while maintaining similar or higher accuracy, where even a minor difference can mean the distinction between life and death.

Detection of bacterial spot disease on tomato leaves using a Convolutional Neural Network (CNN)

Tomato diseases have become a major concern to the tomato production sector around the world. A huge proportion of tomato crops deteriorate yearly during growth or after harvesting due to the infections caused by fungus, viruses and bacteria. Early detection of these diseases plays a crucial role in alleviating overall production loss. Over the past decades, farmers have been using visual observation to identify a crop disease in a field. However, the visual observation method is labour intensive, time consuming, and prone to human error. Currently, intelligent approaches have been widely used to detect and classify these diseases. The objective of this study is to design a convolutional neural network VGG16 net architecture that is able to detect tomato diseases on tomato leaves. A model that can successfully detect 5 tomato leaf diseases was developed: (1) Bacterial spot, (2) Early blight, (3) Late blight, (4) Septoria leaf and (5) Yellow leaf curl with a training accuracy of 0.9328.

Emotional Speech Recognition for Marathi Language

A spontaneous mental state, emotion does not result from deliberate effort. There are many different kinds of emotions in a speech. Because it enhances interactions between people and technology, automatic emotion identification from human speech is becoming more common today. Several temporal and spectral features of human speech can be extracted. Several methods can be used to categorise pitch-related traits, Mel Frequency Cepstral Coefficients (MFCCs), and speech formants. This study looks at statistical characteristics, including MFCCs and linear discriminant analysis, which were used to categorise these properties (LDA). This article also describes a database of artificially emotionalized Marathi speech. The data samples were collected from Marathi speeches given by men and women that mimicked the emotions that resulted to Marathi utterances that could be utilised in everyday conversation and are interpreted in all analysed emotions. To identify data samples, three essential categories—happy, sad, and angry—were used. For MFCC and LPC, the training accuracy and testing accuracy are 98, 82 and 85,82 respectively.