Journal of Intelligent & Fuzzy Systems - Volume Pre-press, issue Pre-press

Authors: Brännström, Andreas | Nieves, Juan Carlos

Article Type: Research Article

Abstract: This paper introduces an automated decision-making framework for providing controlled agent behavior in systems dealing with human behavior-change. Controlled behavior in such settings is important in order to reduce unexpected side-effects of a system’s actions. The general structure of the framework is based on a psychological theory, the Theory of Planned Behavior (TPB), capturing causes to human motivational states, which enables reasoning about dynamics of human motivation. The framework consists of two main components: 1) an ontological knowledge-base that models an individual’s behavioral challenges to infer motivation states and 2) a transition system that, in a given motivation state, decides …on motivational support, resulting in transitions between motivational states. The system generates plans (sequences of actions) for an agent to facilitate behavior change. A particular use-case is modeled regarding children with Autism Spectrum Conditions (ASC) who commonly experience difficulties in everyday social situations. An evaluation of a proof-of-concept prototype is performed that presents consistencies between ASC experts’ suggestions and plans generated by the system. Show more

Keywords: Interactive agents, strategic decision-making, behavior-change systems, theory of planned behavior, Autism

DOI: 10.3233/JIFS-219335

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-11, 2024

Authors: López-López, Aurelio | Garcıa-Gorrostieta, Jesús Miguel | González-López, Samuel

Article Type: Research Article

Abstract: Emotion detection in educational dialogues, particularly within student-teacher interactions, has become a crucial research area for improving the learning experience. In this paper, we employ two models, one generic Bidirectional Encoder Representations from Transformers (BERT) and the Emotion detection model Robustly Optimized BERT Approach (EmoRoBERTa), to automatically classify emotions in a corpus of student-teacher chat interactions. Then subsequently, we validate these classifications using a scheme based on oracles, employing two generative large language models (ChatGPT and Bard). Experiments on emotion detection in dialogues between students and teachers revealed that EmoRoBERTa exhibited a reasonable level of agreement with the oracles, while …ChatGPT demonstrated the highest consistency with EmoRoBERTa’s predictions. Furthermore, we identified the impact of specific words on emotion classification, offering insights into the decision-making process of these models. The results not only highlight the prominent presence of emotions like approval, gratitude, curiosity, disapproval, amusement, confusion, remorse, joy , and surprise but also provide substantial support for the utilization of the proposed emotion detection model to enhance the student learning environment. Exploring the emotional aspects of educational dialogues holds the potential to enhance instruction methods, provide timely assistance to students in need, and create an improved learning atmosphere. Show more

Keywords: Emotion detection, learning interaction, transfer learning, large language models, active learning

DOI: 10.3233/JIFS-219340

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-11, 2024

Authors: Shi, Xiaolong | Kosari, Saeed | Rangasamy, Parvathi | Nivedhaa, R.K. | Rashmanlou, Hossein

Article Type: Research Article

Abstract: Modern image processing techniques are improving beyond old methods, which include advanced approaches, for example deep learning. Convolutional Neural Networks (CNNs) are excellent at automatic feature extraction, whereas Generative Adversarial Networks (GANs) produce realistic images. Transfer learning uses pre-trained models, whereas semantic segmentation identifies pixels in images. Super-resolution, style transfer, and attention mechanisms can increase the quality of images and understanding. Adversarial defenses address purposeful manipulations, while 3D image processing handles three-dimensional data. These advancements make use of improved computational power and massive datasets to revolutionize image processing capabilities. Traditional image processing algorithms frequently fail to handle the complex and …multidimensional structure of color images, particularly when dealing with uncertainty and imprecision. In this study, the 3D-EIFIM frame work is extented and scaled aggregation operations 3D-EIFIM tailored for image data are proposed. By representing each pixel as an entry of 3D-EIFIM and applying aggregation techniques to enable more effective image analysis, manipulation, and enhancement. The practical implications of this research are significant, as it can lead to advancements in fields such as computer vision, medical imaging, and remote sensing. Show more

Keywords: IFP, conjunction, disjunction, IFIM, EIFIM, 3D-IFIM, 3D-EIFIM

DOI: 10.3233/JIFS-238252

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-17, 2024

Authors: Vimala, S. | Valarmathi, K.

Article Type: Research Article

Abstract: This study proposes a novel method using hybrid CNN-LSTM networks to measure and predict the effectiveness of speech and vision therapy. Traditional methods for evaluating therapy often rely on subjective assessments, lacking precision and efficiency. By combining CNN for visual data and MFCC for speech, alongside LSTM for temporal dependencies, the system captures dynamic changes in patients’ conditions. Pre-processing of audio and visual data enhances accuracy, and the model’s performance outperforms existing methods. This approach exhibits the potential of deep learning in monitoring patient progress effectively in speech and vision therapy, offering valuable insights for improving treatment outcomes. The proposed …system’s effectiveness is assessed by various performance metrics. The suggested system’s results are compared with those of other methods already in use. The study’s findings indicate that the suggested approach is more accurate than other existing models. In conclusion, this study offers important new information on how deep learning methods are being used to track patients’ progress in speech and vision therapy. Show more

Keywords: Monitor, speech and vision, deep learning, therapy patient, recording device, CNN-LSTM, categorization

DOI: 10.3233/JIFS-237363

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-17, 2024

Authors: Ravi, Vinayakumar

Article Type: Research Article

Abstract:  Deep learning-based models are employed in computer-aided diagnosis (CAD) tools development for pediatric pneumonia (P-Pneumonia) detection. The accuracy of the model depends on the scaling of the deep learning model. A survey on deep learning shows that models with a greater number of layers achieve better performances for P-Pneumonia detection. However, the identification of the optimal models is considered to be important work for P-Pneumonia detection. This work presents a hybrid deep learning model for P-Pneumonia detection. The model leverages the EfficientNetV2 model that employs various advanced methodologies to maintain the balance between the model scaling and the performance of …the model in P-Pneumonia detection. The features of EfficientNetV2 models are passed into global weighted average pooling (GWAP) which acts like an attention layer. It helps to extract the important features that point to the infected regions of the radiography image and discard all the unimportant information. The features from GWAP are high in dimension and using kernel-based principal component analysis (K-PCA), the features were reduced. Next, the reduced features are combined together and passed into a stacked classifier. The stacked classifier is a two-stage approach in which the first stage employs a support vector machine (SVM) and random forest tree (RFT) for the prediction of P-Pneumonia using the fused features and logistic regression (LRegr) on values of prediction for classification. Detailed experiments were done for the proposed method in P-Pneumonia detection using publically available benchmark datasets. Various settings in the experimental analysis are done to identify the best model. The proposed model outperformed the other methods by improving the accuracy by 4% in P-Pneumonia detection. To show that the proposed model is robust, the model performances were shown on the completely unseen dataset of P-Pneumonia. The hybrid deep learning-based P-Pneumonia model showed good performance on completely unseen data samples of P-Pneumonia patients. The generalization of the proposed P-Pneumonia model is studied by evaluating the model on similar lung diseases such as COVID-19 (CV-19) and Tuberculosis (TBS). In all the experiments, the P-Pneumonia model has shown good performances on similar lung diseases. This indicates that the model is robust and generalizable on data samples of different patients with similar lung diseases. The P-Pneumonia models can be used in healthcare and clinical environments to assist doctors and healthcare professionals in improving the detection rate of P-Pneumonia. Show more

Keywords: Pediatric pneumonia, machine learning, deep learning, dimensionality reduction, feature fusion

DOI: 10.3233/JIFS-219397

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-18, 2024

Authors: Vaikunta Pai, T. | Nethravathi, P.S. | Birau, Ramona | Popescu, Virgil | Karthik Pai, B.H. | Naik, Pramod Vishnu

Article Type: Research Article

Abstract: Multimodal conversational AI systems have gained significant attention due to their potential to enhance user experience and enable more interactive and engaging interactions. This vital and complex research field seeks to integrate diverse modalities, including text, images, and speech, to develop conversational AI systems capable of comprehending, perceiving, and generating responses within a multimodal framework. By seamlessly incorporating various modalities, these systems can provide a more comprehensive and immersive conversational experience, enabling users to communicate in a more natural and intuitively. This research presents a novel multimodal architecture empowered by Deep Neural Networks (DNNs) for simultaneous integration and processing of …diverse modalities. Multimodal data encompasses various sources like text, images, audio, video, or sensor data. The objective is to merge and harness information from these modalities to amplify learning and enhance performance across a spectrum of tasks. This research explores the extension of ChatGPT, a state-of-the-art conversational AI model, to handle multimodal inputs, including text and images or text and speech. We present a comprehensive analysis of the benefits and challenges of integrating various options into ChatGPT, examining their impact on understanding, interaction, and overall system performance. Through extensive experimentation and evaluation, we demonstrate the potential of multimodal ChatGPT to provide richer, more context-aware conversations, while also highlighting the existing limitations and open research questions in this evolving field. Multimodal ChatGPT outperform the current GPT-3.5 by 16.51% and it is clear that multimodal ChatGPTis capable of better performance and offer a pathway for further progress in the field of language models. Show more

Keywords: Large language model, generative pre-trained transformer, deep learning, State-Of-The-Art (SOTA), artificial intelligence (AI), reinforcement training from human feedback, natural language processing (NLP), convolutional neural networks (CNN), recurrent neural networks (RNN)

DOI: 10.3233/JIFS-239465

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-17, 2024

Authors: Li, Ye | Zhou, Jingkang

Article Type: Research Article

Abstract: Semi-supervised learning (SSL) aims to reduce reliance on labeled data. Achieving high performance often requires more complex algorithms, therefore, generic SSL algorithms are less effective when it comes to image classification tasks. In this study, we propose ComMatch, a simpler and more effective algorithm that combines negative learning, dynamic thresholding, and predictive stability discriminations into the consistency regularization approach. The introduction of negative learning is to help facilitate training by selecting negative pseudo-labels during stages when the network has low confidence. And ComMatch filters positive and negative pseudo-labels more accurately as training progresses by dynamic thresholds. Since high confidence does …not always mean high accuracy due to network calibration issues, we also introduce network predictive stability, which filters out samples by comparing the standard deviation of the network output with a set threshold, thus largely reducing the influence of noise in the training process. ComMatch significantly outperforms existing algorithms over several datasets, especially when there is less labeled data available. For example, ComMatch achieves 1.82% and 3.6% error rate reduction over FlexMatch and FixMatch on CIFAR-10 with 40 labels respectively. And with 4000 labeled samples, ComMatch achieves 0.54% and 2.65% lower error rates than FixMatch and MixMatch, respectively. Show more

Keywords: Semi-supervised learning, negative learning, dynamic threshold, predictive stability

DOI: 10.3233/JIFS-233940

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-12, 2024

Authors: Sun, Haobin | Chen, Bingsan | Zhang, Wenshui | Wei, Songma | Lian, Changwei

Article Type: Research Article

Abstract: In the process of production, the label on the product provides the basic product information. Due to the complex text contained on the product labels, the high accuracy recognition for online production labels has always been a challenging problem. To address this issue, a more effective method for complex text detection by improving the convolutional recurrent neural network has been proposed to enhance the recognition accuracy of complex text. Firstly, the SE-DenseNet feature extraction network has been introduced for feature extraction, aiming to improve the model’s depth and feature extraction capacity. Then, the Bi-GRU network is utilized to learn and …model the hidden states and spatial features extracted by SE-DenseNet, anticipate preliminary sequence results, reduce model parameters, and improve the model’s calculation performance. Finally, the CTC network is employed for transcription to convert each feature sequence prediction output by Bi-GRU into a label sequence, achieving complex text recognition. Experimental results on the SVT, IIIT-5K, ICDAR2013 public dataset, and a self-built dataset demonstrate that the proposed model achieves superior outcomes on both public and self-built datasets. Remarkably, the model exhibits the highest recognition accuracy of 93.2% on the ICDAR2013 public dataset, demonstrating its potential to support complex text recognition for online production labels. Show more

Keywords: Online production labels, complex text recognition, SE-DenseNet, Bi-GRU

DOI: 10.3233/JIFS-234748

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-14, 2024

Authors: Lv, Zhangwei

Article Type: Research Article

Abstract: In the context of China’s cultural and tourism industry, cultural equipment plays a critical role in cultural dissemination, especially in remote areas with harsh road conditions and unique environmental factors. However, the efficiency and stability of manual analysis are significantly challenged by these conditions and the vast yet sparsely collected monitoring data. This study aims to develop a method for extracting valuable information from monitoring data to assess the health status of cultural equipment. We introduce a deep learning-based algorithm that leverages convolutional neural networks (CNNs) to extract local features from multidimensional monitoring indicators and long short-term memory (LSTM) networks …to capture time series features, facilitating the classification of cultural equipment’s health status. The algorithm’s effectiveness is demonstrated through simulation results, highlighting its practicality and applicability in real-world scenarios. This research not only provides a novel approach for cultural equipment health assessment but also contributes significantly to the field by addressing the challenges of data analysis in complex environments, underscoring the importance of technological advancements in preserving cultural heritage. Show more

Keywords: Environmental evaluation, convolutional neural network, long short term memory, health status

DOI: 10.3233/JIFS-241607

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-15, 2024

Authors: Shamma, Aashitha L. | Vekkot, Susmitha | Gupta, Deepa | Zakariah, Mohammed | Alotaibi, Yousef Ajami

Article Type: Research Article

Abstract:  This paper investigates the potential of COVID-19 detection using cough, breathing, and voice patterns. Speech-based features, such as MFCC, zero crossing rate, spectral centroid, spectral bandwidth, and chroma STFT are extracted from audio recordings and evaluated for their effectiveness in identifying COVID-19 cases from Coswara dataset. The explainable AI SHAP tool is employed which identified MFCC, zero crossing rate, and spectral bandwidth as the most influential features. Data augmentation techniques like random sampling, SMOTE, Tomek, and Edited Nearest Neighbours (ENN), are applied to improve the performance of various machine learning models used viz. Naive Bayes, K-nearest neighbours, support vector machines, …XGBoost, and Random Forest. Selecting the top 20 features achieves an accuracy of 73%, a precision of 74%, a recall of 94%, and an F1-score of 83% using the Random Forest model with the Tomek sampling technique. These findings demonstrate that a carefully selected subset of features can achieve comparable performance to the entire feature set while maintaining a high recall rate. The success of the Tomek undersampling technique highlights the ability of model to handle sparse clinical data and predict COVID-19 and associated diseases using speech-based features. Show more

Keywords: Covid-19, MFCC, spectral bandwidth, zero crossing rate, SHAP tool, Tomek

DOI: 10.3233/JIFS-219387

Citation: Journal of Intelligent & Fuzzy Systems, vol. Pre-press, no. Pre-press, pp. 1-14, 2024