A Robust Deep Transfer Learning Model for Accurate Speech Emotion Classification

Electricity load forecasting is a crucial undertaking within all the deregulated markets globally. Among the research challenges on a global scale, the investigation of deep transfer learning (DTL) in the field of electricity load forecasting represents a fundamental effort that can inform artificial intelligence applications in general. In this paper, a comprehensive study is reported regarding day-ahead electricity load forecasting. For this purpose, three sequence-to-sequence (Seq2seq) deep learning (DL) models are used, namely the multilayer perceptron (MLP), the convolutional neural network (CNN) and the ensemble learning model (ELM), which consists of the weighted combination of the outputs of MLP and CNN models. Also, the study focuses on the development of different forecasting strategies based on DTL, emphasizing the way the datasets are trained and fine-tuned for higher forecasting accuracy. In order to implement the forecasting strategies using deep learning models, load datasets from three Greek islands, Rhodes, Lesvos, and Chios, are used. The main purpose is to apply DTL for day-ahead predictions (1–24 h) for each month of the year for the Chios dataset after training and fine-tuning the models using the datasets of the three islands in various combinations. Four DTL strategies are illustrated. In the first strategy (DTL Case 1), each of the three DL models is trained using only the Lesvos dataset, while fine-tuning is performed on the dataset of Chios island, in order to create day-ahead predictions for the Chios load. In the second strategy (DTL Case 2), data from both Lesvos and Rhodes concurrently are used for the DL model training period, and fine-tuning is performed on the data from Chios. The third DTL strategy (DTL Case 3) involves the training of the DL models using the Lesvos dataset, and the testing period is performed directly on the Chios dataset without fine-tuning. The fourth strategy is a multi-task deep learning (MTDL) approach, which has been extensively studied in recent years. In MTDL, the three DL models are trained simultaneously on all three datasets and the final predictions are made on the unknown part of the dataset of Chios. The results obtained demonstrate that DTL can be applied with high efficiency for day-ahead load forecasting. Specifically, DTL Case 1 and 2 outperformed MTDL in terms of load prediction accuracy. Regarding the DL models, all three exhibit very high prediction accuracy, especially in the two cases with fine-tuning. The ELM excels compared to the single models. More specifically, for conducting day-ahead predictions, it is concluded that the MLP model presents the best monthly forecasts with MAPE values of 6.24% and 6.01% for the first two cases, the CNN model presents the best monthly forecasts with MAPE values of 5.57% and 5.60%, respectively, and the ELM model achieves the best monthly forecasts with MAPE values of 5.29% and 5.31%, respectively, indicating the very high accuracy it can achieve.

Coronavirus, also known as COVID-19, has spread to several countries around the world. It was announced as a pandemic disease by The World Health Organization (WHO) in 2020 for its devastating impact on humans. With the advancements in computer science algorithms, the detection of this type of virus in the early stages is urgently needed for the fast recovery of patients. In this paper, a study of neutrosophic set significance on deep transfer learning models will be presented. The study will be conducted over a limited COVID-19 x-ray. The study relies on neutrosophic set and theory to convert the medical images from the grayscale spatial domain to the neutrosophic domain. The neutrosophic domain consists of three types of images, and they are the True (T) images, the Indeterminacy (I) images, and the Falsity (F) images. The dataset used in this research has been collected from different sources. The dataset is classified into four classes {COVID-19, normal, pneumonia bacterial, and pneumonia virus}. This study aims to review the effect of neutrosophic sets on deep transfer learning models. The selected deep learning models in this study are Alexnet, Googlenet, and Restnet18. Those models are selected as they have a small number of layers on their architectures. To test the performance of the conversion to the neutrosophic domain, more than 36 trials have been conducted and recorded. A combination of training and testing strategies by splitting the dataset into (90–10%, 80–20%, 70–30) is included in the experiments. Four domains of images are tested, and they are, the original domain, the True (T) domain, the Indeterminacy (I) domain, and the Falsity (F) domain. The four domains with the different training and testing strategies were tested using the selected deep transfer models. According to the experimental results, the Indeterminacy (I) neutrosophic domain achieves the highest accuracy possible with 87.1% in the testing accuracy and performance metrics such as Precision, Recall, and F1 Score. The study concludes that using the neutrosophic set with deep learning models may be an encouraging transition to achieve better testing accuracy, especially with limited COVID-19 datasets.

Tuberculosis (TB) remains a leading infectious disease worldwide, and early, reliable screening using chest X-rays (CXRs) is essential in low-resource settings. The scarcity of labeled TB-positive CXR images limits the effectiveness of deep learning models. This study investigates whether Conditional Generative Adversarial Networks (CGANs) can generate realistic TB-positive CXR images to balance training data and improve the classification performance of fine-tuned deep transfer learning (DTL) models. We trained a CGAN (LSGAN formulation) to synthesize class-conditional grayscale CXR images at 128x128 resolution and used the generated images to augment the Shenzhen TB dataset. Three pre-trained DTL architectures (DenseNet121, VGG16, and MobileNetV3Small) were fine-tuned on both original and CGAN-augmented datasets. Experiments used stratified 70/10/20 train/validation/test splits and a fixed random seed (random_state=42) to ensure reproducibility. Model performance was evaluated using accuracy, precision, recall (sensitivity), F1-score, confusion matrices, and ROC/AUC curves. The experiments were executed on an NVIDIA Tesla P100 GPU (16GB) in a Kaggle runtime environment; total CGAN+classifier processing reported a wall-clock runtime of 39 minutes 30 seconds for the baseline experimental run. CGAN augmentation produced consistent improvements across models: DenseNet121 improved from 93.0% to 94.6% test accuracy, VGG16 improved from 96.3% to 96.8%, and MobileNetV3Small improved from 93.0% to 93.5%. Class-conditional GAN augmentation can modestly but usefully improve DTL classifier performance in TB detection when labeled data are scarce, though further cross-dataset validation is required before clinical deployment.

In distribution system state estimation (DSSE), the physics-based approach such as the weighted least squares (WLS) technique suffers from convergence issues, whereas the problem with conventional deep learning (DL) models is the lack of generalizability. To address these aspects, this paper proposes a physics-guided deep learning (PGDL) framework, which incorporates the power system physical laws into the deep learning models for DSSE. The proposed approach has two stages. The first stage leverages the DL models to understand the relation between the measurement data and system states, while the second stage involves physics-guided equations to decode the measurement data from the estimated states. Further, a loss function incorporating the estimated states and estimated measurements is derived to guide the learning process of the considered deep learning model. The required database for DL models is generated by carrying out power flow studies on a modified IEEE 37-node unbalanced distribution test system. Various deep learning models such as multi-layer perceptron (MLP), convolutional neural networks (CNN), and hybrid CNN-MLP are implemented in the proposed PGDL framework for DSSE and their performance metrics are evaluated and compared under different scenarios. The results show that the PGDL approaches achieve better performance compared with their conventional DL models and the basic WLS algorithm.

COVID-19 is highly infectious and causes acute respiratory disease. Machine learning (ML) and deep learning (DL) models are vital in detecting disease from computerized chest tomography (CT) scans. The DL models outperformed the ML models. For COVID-19 detection from CT scan images, DL models are used as end-to-end models. Thus, the performance of the model is evaluated for the quality of the extracted feature and classification accuracy. There are four contributions included in this work. First, this research is motivated by studying the quality of the extracted feature from the DL by feeding these extracted to an ML model. In other words, we proposed comparing the end-to-end DL model performance against the approach of using DL for feature extraction and ML for the classification of COVID-19 CT scan images. Second, we proposed studying the effect of fusing extracted features from image descriptors, e.g., Scale-Invariant Feature Transform (SIFT), with extracted features from DL models. Third, we proposed a new Convolutional Neural Network (CNN) to be trained from scratch and then compared to the deep transfer learning on the same classification problem. Finally, we studied the performance gap between classic ML models against ensemble learning models. The proposed framework is evaluated using a CT dataset, where the obtained results are evaluated using five different metrics The obtained results revealed that using the proposed CNN model is better than using the well-known DL model for the purpose of feature extraction. Moreover, using a DL model for feature extraction and an ML model for the classification task achieved better results in comparison to using an end-to-end DL model for detecting COVID-19 CT scan images. Of note, the accuracy rate of the former method improved by using ensemble learning models instead of the classic ML models. The proposed method achieved the best accuracy rate of 99.39%.

Nowadays, the whole world is facing a pandemic situation in the form of coronavirus diseases (COVID-19). In connection with the spread of COVID-19 confirmed cases and deaths, various researchers have analysed the impact of temperature and humidity on the spread of coronavirus. In this paper, a deep transfer learning-based exhaustive analysis is performed by evaluating the influence of different weather factors, including temperature, sunlight hours, and humidity. To perform all the experiments, two data sets are used: one is taken from Kaggle consists of official COVID-19 case reports and another data set is related to weather. Moreover, COVID-19 data are also tested and validated using deep transfer learning models. From the experimental results, it is shown that the temperature, the wind speed, and the sunlight hours make a significant impact on COVID-19 cases and deaths. However, it is shown that the humidity does not affect coronavirus cases significantly. It is concluded that the convolutional neural network performs better than the competitive model.

Subject matter: Speech emotion recognition (SER) is an ongoing interesting research topic. Its purpose is to establish interactions between humans and computers through speech and emotion. To recognize speech emotions, five deep learning models: Convolution Neural Network, Long-Short Term Memory, Artificial Neural Network, Multi-Layer Perceptron, Merged CNN, and LSTM Network (CNN-LSTM) are used in this paper. The Toronto Emotional Speech Set (TESS), Surrey Audio-Visual Expressed Emotion (SAVEE) and Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS) datasets were used for this system. They were trained by merging 3 ways TESS+SAVEE, TESS+RAVDESS, and TESS+SAVEE+RAVDESS. These datasets are numerous audios spoken by both male and female speakers of the English language. This paper classifies seven emotions (sadness, happiness, anger, fear, disgust, neutral, and surprise) that is a challenge to identify seven emotions for both male and female data. Whereas most have worked with male-only or female-only speech and both male-female datasets have found low accuracy in emotion detection tasks. Features need to be extracted by a feature extraction technique to train a deep-learning model on audio data. Mel Frequency Cepstral Coefficients (MFCCs) extract all the necessary features from the audio data for speech emotion classification. After training five models with three datasets, the best accuracy of 84.35 % is achieved by CNN-LSTM with the TESS+SAVEE dataset.

The anterior cruciate ligaments (ACL) are the fundamental structures in preserving the common biomechanics of the knees and most frequently damaged knee ligaments. An ACL injury is a tear or sprain of the ACL, one of the fundamental ligaments in the knee. ACL damage most generally happens during sports, for example, soccer, ball, football, and downhill skiing, which include sudden stops or changes in direction, jumping, and landings. Magnetic resonance imaging (MRI) has a major role in the field of diagnosis these days. Specifically, it is effective for diagnosing the cruciate ligaments and any related meniscal tears. The primary objective of this research is to detect the ACL tear from MRI knee images, which can be useful to determine the knee abnormality. In this research, a Deep Convolution Neural Network (DCNN) based Inception-v3 deep transfer learning (DTL) model was proposed for classifying the ACL tear MRI images. Preprocessing, feature extraction, and classification are the main processes performed in this research. The dataset utilized in this work was collected from the MRNet database. A total of 1,370 knee MRI images are used for evaluation. 70% of data (959 images) are used for training and testing, and 30% of data (411 images) are used in this model for performance analysis. The proposed DCNN with the Inception-v3 DTL model is evaluated and compared with existing deep learning models like VGG16, VGG19, Xception, and Inception ResNet-v28. The performance metrics like accuracy, precision, recall, specificity, and F-measure are evaluated to estimate the performance analysis of the model. The model has obtained 99.04% training accuracy and 95.42% testing accuracy in performance analysis.

Daily scale streamflow forecasting in multiple stream orders of Cauvery River, India: Application of advanced ensemble and deep learning models

Brain Tumor is a deadly diagnosis that, if not detected early on, can cause significant damage to a patient’s brain, disrupting the body’s functions and even leading to death. However, the conventional method of detecting brain tumors is conducting MRI scans which a medical expert then consults for diagnosis. While time-consuming, this method also leaves room for human error, especially in cases where the tumor is in its early stages. Thus, the diagnosis of brain tumors must be made accurately in the least time possible. This paper aims to prevent premature mortality, provide health in resource-constrained settings, and seek patients’ healthy lives, which can be done through timely and accurate diagnosis of brain tumors. In this paper, three different deep transfer learning models are used after adding fine-tuned layers to detect brain tumors in a dataset of 251 Magnetic Resonance Imaging (MRI) scans. The transfer learning models used are VGG16, InceptionV3, and ResNet50, with fine-tuned Dropout, Flatten, and Dense added layers. The highest accuracy was achieved with the VGG16 model, which had an accuracy of 91.58%. Thus, deep learning models are proven effective in detecting brain tumors without the unwanted expense of time and resources.

PurposeTo conduct a head-to-head comparison between deep learning (DL) and radiomics models across institutions for predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC) and to investigate the model robustness and generalizability through rigorous internal and external validation.MethodsThis retrospective study included 2304 preoperative images of 576 HCC lesions from two centers, with MVI status determined by postoperative histopathology. We developed DL and radiomics models for predicting the presence of MVI using B-mode ultrasound, contrast-enhanced ultrasound (CEUS) at the arterial, portal, and delayed phases, and a combined modality (B + CEUS). For radiomics, we constructed models with enlarged vs. original regions of interest (ROIs). A cross-validation approach was performed by training models on one center’s dataset and validating the other, and vice versa. This allowed assessment of the validity of different ultrasound modalities and the cross-center robustness of the models. The optimal model combined with alpha-fetoprotein (AFP) was also validated. The head-to-head comparison was based on the area under the receiver operating characteristic curve (AUC).ResultsThirteen DL models and 25 radiomics models using different ultrasound modalities were constructed and compared. B + CEUS was the optimal modality for both DL and radiomics models. The DL model achieved AUCs of 0.802–0.818 internally and 0.667–0.688 externally across the two centers, whereas radiomics achieved AUCs of 0.749–0.869 internally and 0.646–0.697 externally. The radiomics models showed overall improvement with enlarged ROIs (P < 0.05 for both CEUS and B + CEUS modalities). The DL models showed good cross-institutional robustness (P > 0.05 for all modalities, 1.6–2.1% differences in AUC for the optimal modality), whereas the radiomics models had relatively limited robustness across the two centers (12% drop-off in AUC for the optimal modality). Adding AFP improved the DL models (P < 0.05 externally) and well maintained the robustness, but did not benefit the radiomics model (P > 0.05).ConclusionCross-institutional validation indicated that DL demonstrated better robustness than radiomics for preoperative MVI prediction in patients with HCC, representing a promising solution to non-standardized ultrasound examination procedures.

Graph convolutional network – Long short term memory neural network- multi layer perceptron- Gaussian progress regression model: A new deep learning model for predicting ozone concertation

Small and Medium-sized Enterprises (SMEs) in Sri Lanka are increasingly turning to social media to gain insights into consumer preferences; however, they frequently lack sophisticated analytical tools to derive actionable insights. This research introduces a hybrid deep learning framework aimed at sentiment and emotion analysis specifically for the Sri Lankan cake sector. Customer reviews and social media comments were gathered and processed through various methods including text cleaning, emoji translation, stopword elimination, tokenization, and sequence padding. Several deep learning models— Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM), and Gated Recurrent Units (GRU) were assessed, along with hybrid models that combined CNN and BiLSTM with GPT-2 embeddings. The CNN+GPT-2 model achieved the highest accuracy in sentiment classification at 96.14%, while BiLSTM performed best in emotion classification with an accuracy of 91.57%. In terms of trend forecasting, Long Short-Term Memory (LSTM) outshone both SARIMA and Prophet, yielding the lowest error metrics and the most reliable sentiment predictions. These results underscore the potential of hybrid deep learning and transformer-based models to effectively capture complex consumer sentiments and provide SMEs with practical, AI-driven insights. This study is among the first to implement such hybrid NLP models within a localized Sri Lankan context, presenting a scalable approach for business intelligence and informed decision-making.

Water quality prediction (WQP) plays an essential role in water quality management for aquaculture to make aquaculture production profitable and sustainable. In this work, we propose hybrid deep learning (DL) models, convolutional neural network (CNN) with the long short-term memory (LSTM) and gated recurrent unit (GRU) for aquaculture WQP. CNN can effectively fetch the aquaculture water quality characteristics, whereas GRU and LSTM can learn long-term dependencies in the time series data. We conduct experiments using the two different water quality datasets and present an extensive study on the impact of hyperparameters on the performance of the proposed hybrid DL models. Furthermore, the performance of hybrid CNN-LSTM and CNN-GRU models are compared with different baseline LSTM, GRU and CNN DL models and also with attention-based LSTM and attention-based GRU DL models. The results show that the hybrid CNN-LSTM outperformed all other models in terms of prediction accuracy and computation time.

We aimed at comparing intratumoral and peritumoral deep learning, radiomics, and fusion models in predicting KRAS mutations in rectal cancer using endorectal ultrasound imaging. This study included 304 patients with rectal cancer from Fujian Medical University Union Hospital. The patients were randomly divided into a training group (213 patients) and a test group (91 patients) at a 7:3 ratio. Radiomics and deep learning models were established using primary tumor and peritumoral images. In the optimally performing regions-of-interest, two fusion strategies, a feature-based and a decision-based model, were employed to build the fusion models. The Shapley additive explanation (SHAP) method was used to evaluate the significance of features in the optimal radiomics, deep learning, and fusion models. The performance of each model was assessed using the area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA). In the test cohort, both the radiomics and deep learning models exhibited optimal performance with a 10-pixel patch extension, yielding AUC values of 0.824 and 0.856, respectively. The feature-based DLRexpand10_FB model attained the highest AUC (0.896) across all study sets. In addition, the DLRexpand10_FB model demonstrated excellent sensitivity, specificity, and DCA. SHAP analysis underscored the deep learning feature (DL_1) as the most significant factor in the hybrid model. The feature-based fusion model DLRexpand10_FB can be employed to predict KRAS gene mutations based on pretreatment endorectal ultrasound images of rectal cancer. The integration of peritumoral regions enhanced the predictive performance of both the radiomics and deep learning models.