Deep sentiments in Roman Urdu text using Recurrent Convolutional Neural Network model

The presented paper proposes a new speech command recognition model for novel engineering applications with limited resources. We built the proposed model with the help of a Convolutional Recurrent Neural Network (CRNN). The use of CRNN instead of Convolutional Neural Network (CNN) helps us to reduce the model parameters and memory requirement as per resource constraints. Furthermore, we insert transmute and curtailment layer between the layers of CRNN. By doing this we further reduce model parameters and float number of operations to half of the CRNN requirement. The proposed model is tested on Google’s speech command dataset. The obtained result shows that the proposed CRNN model requires 1/3 parameters as compared to the CNN model. The number of parameters of the CRNN model is further reduced by 45% and the float numbers of operations between 2% to 12 % in different recognition tasks. The recognition accuracy of the proposed model is 96% on Google’s speech command dataset, and on laboratory recording, its recognition accuracy is 89%.

Multimodal emotion recognition has become a hot topic in human-computer interaction and intelligent healthcare fields. However, combining information from different human different modalities for emotion computation is still challenging. In this paper, we propose a three-dimensional convolutional recurrent neural network model (referred to as 3FACRNN network) based on multimodal fusion and attention mechanism. The 3FACRNN network model consists of a visual network and an EEG network. The visual network is composed of a cascaded convolutional neural network-time convolutional network (CNN-TCN). In the EEG network, the 3D feature building module was added to integrate band information, spatial information and temporal information of the EEG signal, and the band attention and self-attention modules were added to the convolutional recurrent neural network (CRNN). The former explores the effect of different frequency bands on network recognition performance, while the latter is to obtain the intrinsic similarity of different EEG samples. To investigate the effect of different frequency bands on the experiment, we obtained the average attention mask for all subjects in different frequency bands. The distribution of the attention masks across the different frequency bands suggests that signals more relevant to human emotions may be active in the high frequency bands γ (31-50 Hz). Finally, we try to use the multi-task loss function Lc to force the approximation of the intermediate feature vectors of the visual and EEG modalities, with the aim of using the knowledge of the visual modalities to improve the performance of the EEG network model. The mean recognition accuracy and standard deviation of the proposed method on the two multimodal sentiment datasets DEAP and MAHNOB-HCI (arousal, valence) were 96.75 ± 1.75, 96.86 ± 1.33; 97.55 ± 1.51, 98.37 ± 1.07, better than those of the state-of-the-art multimodal recognition approaches. The experimental results show that starting from the multimodal information, the facial video frames and electroencephalogram (EEG) signals of the subjects are used as inputs to the emotion recognition network, which can enhance the stability of the emotion network and improve the recognition accuracy of the emotion network. In addition, in future work, we will try to utilize sparse matrix methods and deep convolutional networks to improve the performance of multimodal emotion networks.

Self-attention based recurrent convolutional neural network for disease prediction using healthcare data

The latest wireless sensor network (WSN) developments in critical applications have introduced security risks, like jamming. Intrusion Detection System (IDS) in WSN is the method of recognizing malevolent or unauthorized activities in the network. The intruder's presence to launch different attacks within the network cannot be disregarded. Despite a great deal of effort by the researcher workers, IDS still experienced difficulties enhancing recognition performance while minimizing the false alarm rate and identifying novel intrusions. Recently, Deep Learning (DL) and Machine Learning (ML) based IDS system has been deployed as promising solution to effectively identify intrusion across the network. Therefore, the study presents a Moth Search Optimization with DL-based Intrusion Detection (MSODL-ID) method in the WSN. The MSODL-ID technique aims to effectually identify the occurrence of malicious activities or intrusions in the network. To accomplish this, the MSODL-ID technique undergoes two stages of preprocessing: data conversion and data scaling. In addition, the MSODL-ID technique employs Convolutional Recurrent Neural Network (CRNN) model with a Hopfield layer for intrusion detection purposes. For optimal hyperparameter selection of the CRNN model, the MSO algorithm is used and thereby enhances the classification performance of the CRNN model. The stimulation analysis of the MSODL-ID system is tested by means of Kaggle datasets, and the outcomes exhibit the promising performance of the MSODL-ID system over other current DL approaches.

The majority of online comments/opinions are written in text-free format. Sentiment Analysis can be used as a measure to express the polarity (positive/negative) of comments/opinions. These comments/ opinions can be in different languages i.e. English, Urdu, Roman Urdu, Hindi, Arabic etc. Mostly, people have worked on the sentiment analysis of the English language. Very limited research work has been done in Urdu or Roman Urdu languages. Whereas, Hindi/Urdu is the third largest language in the world. In this paper, we focus on the sentiment analysis of comments/opinions in Roman Urdu. There is no publicly available Roman Urdu public opinion dataset. We prepare a dataset by taking comments/opinions of people in Roman Urdu from different websites. Three supervised machine learning algorithms namely NB (Naive Bayes), LRSGD (Logistic Regression with Stochastic Gradient Descent) and SVM (Support Vector Machine) have been applied on this dataset. From results of experiments, it can be concluded that SVM performs better than NB and LRSGD in terms of accuracy. In case of SVM, an accuracy of 87.22% is achieved.

This paper presents a viewpoint from computer vision to the radar echo extrapolation task in the precipitation nowcasting domain. Inspired by the success of some convolutional recurrent neural network models in this domain, including convolutional LSTM, convolutional GRU and trajectory GRU, we designed a new sequence-to-sequence neural network structure to leverage these models in a realistic data context. In this design, we decreased the numbers of channels in high abstract recurrent layers rather than increasing them. We formulated the task as a problem of encoding five radar images and predicting 10 steps ahead at the pixel level, and found that using only the common mean squared error can misguide the training and mislead the testing. Especially, the image quality of last predictions usually degraded rapidly. As a solution, we employed some visual image quality assessment techniques including Structural Similarity (SSIM) and multi-scale SSIM to train our models. Experimental results show that our structure was more tolerant to increasing uncertainty in the data, and the use of image quality metrics can significantly reduce the blurry image issue. Moreover, we found that using SSIM was very effective and a combination of SSIM with mean squared error and mean absolute error yielded the best prediction quality.

The lack of labeled data is a major obstacle in many music information retrieval tasks such as melody extraction, where labeling is extremely laborious or costly. Semi-supervised learning (SSL) provides a solution to alleviate the issue by leveraging a large amount of unlabeled data. In this paper, we propose an SSL method using teacher-student models for vocal melody extraction. The teacher model is pre-trained with labeled data and guides the student model to make identical predictions given unlabeled input in a self-training setting. We examine three setups of teacher-student models with different data augmentation schemes and loss functions. Also, considering the scarcity of labeled data in the test phase, we artificially generate large-scale testing data with pitch labels from unlabeled data using an analysis-synthesis method. The results show that the SSL method significantly increases the performance against supervised learning only and the improvement depends on the teacher-student models, the size of unlabeled data, the number of self-training iterations, and other training details. We also find that it is essential to ensure that the unlabeled audio has vocal parts. Finally, we show that the proposed SSL method allows a simple convolutional recurrent neural network model to achieve performance comparable to state-of-the-arts.

This work studies learning-based binaural sound source localization, under the influence of head rotation in rever-berant conditions. Emphasis is on whether knowledge of head rotation can improve localization performance over the non-rotating case for the same acoustic scene. Simulations of binaural head signals of a static and rotating head were conducted, for 5 different rotation speeds and a wide range of reverberant conditions. Several convolutional recurrent neural network mod-els were evaluated including a static head scenario, a model without rotation information, and distinct models differentiated on the way of manipulating the quaternions. The results were analyzed based on the direction-of-arrival error, and they show the importance of using quaternions as additional features, with the best localization accuracy obtained when using an additional convolutional branch that merges the features through addition or concatenation. Nevertheless, raw quaternion features presented lower performance than the static baseline model. Additionally, the study shows the importance of the analysis time window length when using information about head rotation.

Emotion detection (ED) and sentiment analysis (SA) play a vital role in identifying an individual’s level of interest in any given field. Humans use facial expressions, voice pitch, gestures, and words to convey their emotions. Emotion detection and sentiment analysis in English and Chinese have received much attention in the last decade. Still, poor-resource languages such as Urdu have been mostly disregarded, which is the primary focus of this research. Roman Urdu should also be investigated like other languages because social media platforms are frequently used for communication. Roman Urdu faces a significant challenge in the absence of corpus for emotion detection and sentiment analysis because linguistic resources are vital for natural language processing. In this study, we create a corpus of 1021 sentences for emotion detection and 20,251 sentences for sentiment analysis, both obtained from various areas, and annotate it with the aid of human annotators from six and three classes, respectively. In order to train large-scale unlabeled data, the bag-of-word, term frequency-inverse document frequency, and Skip-gram models are employed, and the learned word vector is then fed into the CNN-LSTM model. In addition to our proposed approach, we also use other fundamental algorithms, including a convolutional neural network, long short-term memory, artificial neural networks, and recurrent neural networks for comparison. The result indicates that the CNN-LSTM proposed method paired with Word2Vec is more effective than other approaches regarding emotion detection and evaluating sentiment analysis in Roman Urdu. Furthermore, we compare our based model with some previous work. Both emotion detection and sentiment analysis have seen significant improvements, jumping from an accuracy of 85% to 95% and from 89% to 93.3%, respectively.

Automatic sleep stage monitoring is an essential tool for the diagnosis and treatment of sleep-related disorders effectively. Although extensive studies focusing on single source data or information, such as single channel electroencephalogram (EEG), electrooculogram (EOG), and electromyogram (EMG), with machine learning (ML) and deep learning (DL) have been presented in this field of automatic sleep staging, there are very limited literature's that have investigated the impact of multisource information fusion on ML or DL based schemes for the sleep stage classification. In addition, exploiting recurrent neural network (RNN) to learn temporal information from the sequences of multi-source inputs can bring significant improvement in classifying the sleep stage. Therefore, we aim to develop an multi-source data fusion, more specifically two source's signal fusion, enabled end-to-end convolutional recurrent neural network (CRNN) in order to perform automatic sleep stage classification in this article. In the proposed scheme, no hand crafted features is used to train the model and hence the proposed model can be called an end-to-end approach. The proposed CRNN model can extract time-invariant features from raw data of two input sources and can fuse them to learn temporally correlated features. Sleep-EDF expanded benchmark dataset is used in this study and we have employed Fpz-Cz EEG channel signal and EOG signal as input with 3 variants of RNN layer in model architecture separately in order to evaluate and investigate the performance. The proposed CRNN model with standard RNN layers can classify the Sleep-EDF dataset with maximum average accuracy of 90.30% and maximum Cohen's kappa coefficient of 86.86 compared other two RNN variant's layer (e.g., gated recurrent unit and long short term memory) based models which is very promising in the studies on classification of sleep stages as well as can be regarded as an alternative to conventional sleep staging.

The world is heading towards more modernized and digitalized data and therefore a significant growth is observed in the active number of social media users with each passing day. Each post and comment can give an insight into valuable information about a certain topic or issue, a product or a brand, etc. Similarly, the process to uncover the underlying information from the opinion that a person keeps about any entity is called a sentiment analysis. The analysis can be carried out through two main approaches, i.e., either lexicon-based or machine learning algorithms. A significant amount of work in the different domains has been done in numerous languages for sentiment analysis, but minimal research has been conducted on the national language of Pakistan, which is Urdu. Twitter users who are familiar with Urdu update the tweets in two different textual formats either in Urdu Script (Nastaleeq) or in Roman Urdu. Thus, the paper is an attempt to perform the sentiment analysis on the Urdu language by extracting the tweets (Nastaleeq and Roman Urdu both) from Twitter using Tweepy API. A machine learning-based approach has been adopted for this study and the tool opted for the purpose is WEKA. The best algorithm was identified based on evaluation metrics, which comprise the number of correctly and incorrectly classified instances, accuracy, precision, and recall. SMO was found to be the most suitable machine learning algorithm for performing the sentiment analysis on Urdu (Nastaleeq) tweets, while the Roman Urdu Random Forest algorithm was identified as the best one.

BackgroundAccurately determining the softness level of pituitary tumors preoperatively by using their image textures can provide a basis for surgical options and prognosis. Existing methods for this problem require manual intervention, which could hinder the efficiency and accuracy considerably.MethodsWe present an automatic method for diagnosing the texture of pituitary tumors using unbalanced sequence image data. Firstly, for the small sample problem in our pituitary tumor MRI image dataset where T1 and T2 sequence data are unbalanced (due to data missing) and under-sampled, our method uses a CycleGAN (Cycle-Consistent Adversarial Networks) model for domain conversion to obtain fully sampled MRI spatial sequence. Then, it uses a DenseNet (Densely Connected Convolutional Networks)-ResNet(Deep Residual Networks) based Autoencoder framework to optimize the feature extraction process for pituitary tumor image data. Finally, to take advantage of sequence data, it uses a CRNN (Convolutional Recurrent Neural Network) model to classify pituitary tumors based on their predicted softness levels.ResultsExperiments show that our method is the best in terms of efficiency and accuracy (91.78%) compared to other methods.ConclusionsWe propose a semi-supervised method for grading pituitary tumor texture. This method can accurately determine the softness level of pituitary tumors, which provides convenience for surgical selection and prognosis, and improves the diagnostic efficiency of pituitary tumors.

Sentiment analysis is a dynamic field focused on understanding and predicting emotional sentiments in text or images. With the prevalence of smartphones, e-commerce and social networks, individuals readily express opinions, aiding businesses, political analysts and organizations in decision-making. Despite extensive research in sentiment analysis for various languages, challenges persist in low-resource languages like Roman Urdu. Roman Urdu, the use of Roman script to write Urdu, has gained popularity, yet limited linguistic resources hinder sentiment analysis research. This study addresses this gap by developing a bidirectional long short-term memory network with FastText embeddings and additional layers. A large Roman Urdu corpus for sentiment analysis, consisting of over 51 000 reviews, is crated and the proposed model is trained and compared with 14 other models, demonstrating an accuracy of 0.854 and an F1-score of 0.84.

People counting and pedestrian flow statistics are challenging tasks because of the perspective distortions, appearance changes and occlusion. In this paper, we address the two tasks: people counting in images of highly dense crowds and pedestrian flow statistics in a place over a period of time. Our first contribution is to propose a new convolution neural network (CNN) model which is composed of a deep and shallow fully convolution network to fulfill the task of people counting. We extract different layer features from the deep fully convolution network and the last layer features from the shallow fully convolution network, and concatenate them together. After that we add two deconvolution layers to make the output image have the same resolution with the input image. Our second contribution is to combine pedestrian flow statistics task with people counting task. According to the density maps that CNN model generates, we can calculate the number of people crossing a place based on the recurrent neural network (RNN). Besides, we also have collected two datasets and labelled them. Extensive experiments have been implemented, our people counting method outperforms other existing methods, and our pedestrian flow statistics method combined with CNN model also outperforms the model which only uses long-short term memory (LSTM).

To develop and evaluate deep learning models for cardiac arrest rhythm classification during cardiopulmonary resuscitation (CPR), we analyzed 508 electrocardiogram (ECG) segments (each 4 s in duration, recorded at 250 Hz) from 131 cardiac arrest patients. Compression-affected segments were recorded during chest compressions, while non-compression segments were extracted during compression pauses or immediately after return of spontaneous circulation (ROSC) declaration. One-dimensional convolutional neural network (1D-CNN) and recurrent neural network (RNN) models were employed for four binary classification tasks: (1) shockable rhythms (VF and pVT) versus non-shockable rhythms (asystole and PEA) in all ECG segments; (2) the same classification in compression-affected ECG segments; (3) pulse-generating rhythms (ROSC rhythm) versus non-pulse-generating rhythms (asystole, PEA, VF and pVT) in all ECG segments; and (4) the same classification in compression-affected ECG segments. The 1D-CNN model consistently outperformed the RNN model across all classification tasks. For shockable versus non-shockable rhythm classification, the 1D-CNN achieved accuracies of 91.3% and 89.8% for all ECG segments and compression-affected ECG segments, respectively, compared to 50.6% and 54.5% for the RNN. In detecting pulse-generating rhythms, the 1D-CNN demonstrated accuracies of 90.9% and 85.7% for all ECG segments and compression-affected ECG segments, respectively, while the RNN achieved 92.2% and 84.4%. The 1D-CNN model demonstrated superior performance in cardiac arrest rhythm classification, maintaining high accuracy even with compression-affected ECG data.