AbstractConversational assistants (CAs) and Task‐oriented ones, in particular, are designed to interact with users in a natural language manner, assisting them in completing specific tasks or providing relevant information. These systems employ advanced natural language understanding (NLU) and dialogue management techniques to comprehend user inputs, infer their intentions, and generate appropriate responses or actions. Over time, the CAs have gradually diversified to today touch various fields such as e‐commerce, healthcare, tourism, fashion, travel, and many other sectors. NLU is fundamental in the natural language processing (NLP) field. Identifying user intents from natural language utterances is a sub‐task of NLU that is crucial for conversational systems. The diversity in user utterances makes intent detection (ID) even a challenging problem. Recently, with the emergence of Deep Neural Networks. New State of the Art (SOA) results have been achieved for different NLP tasks. Recurrent neural networks (RNNs) and Transformer architectures are two major players in those improvements. RNNs have significantly contributed to sequence modelling across various application areas. Conversely, Transformer models represent a newer architecture leveraging attention mechanisms, extensive training data sets, and computational power. This review paper begins with a detailed exploration of RNN and Transformer models. Subsequently, it conducts a comparative analysis of their performance in intent recognition for Task‐oriented (CAs). Finally, it concludes by addressing the main challenges and outlining future research directions.

The idea of a Cyber-Physical-Social System, or CPSS for short, is a relatively new concept that has emerged as a response to the requirement to comprehend the influence that Cyber-Physical Systems (CPS) have on people and vice versa. Conversational assistants (CAs), also called bots, are dedicated to oral or written communication. Over time, the CAs have gradually diversified to today touch various fields such as e-commerce, healthcare, tourism, fashion, travel, and many others sectors. Natural-language understanding (NLU) is fundamental in the Natural Language Processing (NLP) field. Identifying user intents from natural language utterances is a crucial step in conversational systems, and the diversity in user utterances makes intent detection even a challenging problem. Recently, with the emergence of Deep Neural Networks. New State of the Art (SOA) results have been achieved for different NLP tasks. Recurrent Neural networks (RNNs) and recent Transformer architectures are two major players in those improvements. In addition, RNNs have been playing an increasingly important role in sequence modeling in different application areas. On the other hand, Transformer models are new architectures that benefit from the attention mechanism, extensive training datasets, and compute power. First, this review paper presents a comprehensive overview of RNN and Transformer models. Then, a comparative study of the performance of different RNNs and Transformer architectures for the specific task of intent recognition for CAs which is a fundamental task of NLU.

This article aims to find the determinants that affect patient satisfaction regarding the Mawiidi public portal in Moroccan public hospitals and assess its outpatient online booking system effectiveness using a model that integrates the Technology Acceptance Model (TAM) with the Information Systems Success Model (ISSM) while adopting a quantitative research methodology. The analysis was conducted using 348 self-administered questionnaires to analyze eight key constructs, such as information quality, patient satisfaction, perceived ease of use, and privacy protection, among others. The results of PLS-SEM verified six out of eleven hypotheses tested, which reflected that information quality has a positive influence on perceived ease of use, which again enhances patient satisfaction. The major factors influencing the satisfaction and trust of patients in online appointment scheduling systems at public hospitals are highlighted. Indeed, privacy protection enhances patient satisfaction and trust. Service quality positively affects satisfaction but to a lesser degree. Website-related anxiety impacts perceived ease of use, although it has a limited influence on satisfaction. Such findings can inform suggestions for the managers of hospitals and portal designers to increase user satisfaction. This study uses a model from the TAM and ISSM frameworks, including cultural and socioeconomic aspects that apply to Morocco’s healthcare context.

Electricity is a crucial aspect of modern life, and with the increasing population and industrialization, energy demand has risen significantly. A swift transition to renewable energy sources such as wind and solar is essential for saving the planet. Solar energy is one of the most widely used renewable energy solutions, but choosing a PVSC poses a challenging problem that involves considering various factors, such as geographical location and energy consumption patterns. In this study, we investigate the effectiveness of using machine learning techniques to assist users in selecting the most suitable PVSC for their needs. We propose a new framework for PVSC recommendation, which encompasses a PV power forecasting model and a PV configuration recommendation system.

A dialogue state tracker is a component in a task-oriented dialogue system that monitors the current state of a conversation and gives information about its context and history to other system components. The dynamic and open-ended character of human interactions is one of the primary obstacles in dialogue state monitoring, necessitating robust and adaptable models to keep up with the quick context changes. Recently, numerous deep learning-based algorithms have been developed for this purpose. Still, these models are typically heavily-engineered and conceptually sophisticated, making them challenging to deploy, debug, and maintain in a production environment. To overcome these challenges, we offer the BERT-SIAM-DST model, a unique way to dialogue state monitoring employing a Siamese network with BERT as the base network. This model uses the robust representation capabilities of BERT and the ability of Siamese networks to record correlations between inputs to make accurate predictions regarding the current state of the discussion. In addition, the number of parameters does not increase proportionally with the size of the ontology, and the model is adaptable to alterations in the domain ontology. We test the performance of the BERT-SIAM-DST model on the standard WoZ 2.0 dataset of annotated dialogues and compare it to other approaches. Compared to numerous baseline models, the BERT-SIAM-DST model is effective at tracking the state of discussions, demonstrating the promise of BERT-based Siamese networks for this purpose.

In the age of information overload, customers are overwhelmed with the number of products available for sale. Search engines try to overcome this issue by filtering relevant items based on the users' queries. Traditional product search engines rely on the exact match of terms in the query and product meta-data. Recently, deep learning methods have grabbed more attention outperforming traditional methods in many circumstances. In this work, a new approach is presented to extract relevant items in the context of e-commerce product search using similarity measure on the top of query and product embeddings. To refine the search results for the users' convenience, a deep neural network architecture is proposed to rank products based on the sentiment polarity presented in their associated reviews. The performance of the proposed method is evaluated on the Amazon reviews dataset. The results showed that the proposed method is effective in retrieving relevant and high-quality products.

With the rapid development of artificial intelligence (AI), many companies are moving towards automating their services using automated conversational agents. Dialogue-based conversational recommender agents, in particular, have gained much attention recently. The successful development of such systems in the case of natural language input is conditioned by the ability to understand the users’ utterances. Predicting the users’ intents allows the system to adjust its dialogue strategy and gradually upgrade its preference profile. Nevertheless, little work has investigated this problem so far. This paper proposes an LSTM-based Neural Network model and compares its performance to seven baseline Machine Learning (ML) classifiers. Experiments on a new publicly available dataset revealed The superiority of the LSTM model with 95% Accuracy and 94% F1-score on the full dataset despite the relatively small dataset size (9300 messages and 17 intents) and label imbalance.

The Internet Of Things (IoT) is an emerging paradigm representing a network of infrastructure connecting different types of devices in different places. Recently, IoT-based Conversational systems gained more attention, providing users with a new human-computer interaction mode. These new systems allow users to interact with the IoT ecosystem through a virtual assistant. They also heavily rely on Natural Language Processing (NLP) techniques for intelligent and efficient communication. Understanding the user’s needs is the first step toward a more humanized IoT interaction system. This can be achieved using Natural language understanding (NLU) modules. Slot Filling is one of the core sub-tasks in NLU. It is an active research area focusing on extracting attribute values from the user’s utterance. In this study, we propose an LSTM-based model for slot filling. The model benefits from contextual embeddings extracted from the BERT transformer model, the attention mechanism, and the conditional random field (CRF) that is known to model strong dependence among adjacent tags in the output sequence. Experiments show that our model achieved competitive results on two widely used benchmark datasets.

Conversational Systems are increasingly substituting humans in many service industries. They aim to provide human-like interaction with users for task completion or chitchat in a conversation style. User sentiment analysis is an important task that can help better understand users’ behavior and satisfaction in conversations. Although some researchers have studied the problem of sentiment analysis, most of the existing methods are oriented toward general felds. To overcome the challenges of sentiment analysis, we propose a BE-Att-BiLSTM, which stands for an attention-based Bidirectional Long Short-Term Memory (BiLSTM) model. The proposed model uses pre-trained BERT, contextual embeddings and a combination of BiLSTM and attention mechanism for efficient sentiment analysis in conversations. In addition, text-augmentation techniques are leveraged to enhance the performance of the proposed model. Experimental results on a public benchmark dataset show an improved accuracy of 68.00% and an F1-score of 67.50%.

In recent years Natural language processing is one of the most active areas of research especially with the emergence of deep learning algorithms. More attention has been given to Latin descendent languages e.g English, French, and Spanish given the availability of high-quality datasets and compute resources. In this paper, we present a moroccan News Articles Corpus collected from four of the major moroccan news websites. The corpus contains more than 418k news articles corresponding to 19 different categories, thus considered to be one of the largest Arabic news articles corpora. A description of the collection and processing steps were presented and exploration analysis was performed. To prove the utility of the dataset. An evaluation step was conducted in the context of text classification using four different Machine Learning baselines: Random Forest (RF), Multinomial Naive Bayes (MNB), Support Vector Machine (SVC), and Gradient Boosting (GradBoost) Classifiers. The experimental results are presented in terms of accuracy, F1-score, and confusion matrix.

In the age of information overload, customers are overwhelmed with the number of products available for sale. Search engines try to overcome this issue by filtering relevant items to the users’ queries. Traditional search engines rely on the exact match of terms in the query and product meta-data. Recently, deep learning-based approaches grabbed more attention by outperforming traditional methods in many circumstances. In this work, we involve the power of embeddings to solve the challenging task of optimizing product search engines in e-commerce. This work proposes an e-commerce product search engine based on a similarity metric that works on top of query and product embeddings. Two pre-trained word embedding models were tested, the first representing a category of models that generate fixed embeddings and a second representing a newer category of models that generate context-aware embeddings. Furthermore, a re-ranking step was performed by incorporating a list of quality indicators that reflects the utility of the product to the customer as inputs to well-known ranking methods. To prove the reliability of the approach, the Amazon reviews dataset was used for experimentation. The results demonstrated the effectiveness of context-aware embeddings in retrieving relevant products and the quality indicators in ranking high-quality products.

Texture is a fundamental characteristic of many types of images, especially those with significant rotation, scale illumination, and viewpoint change. Texture image classification is one of the challenging problems that have various applications such as remote sensing, material recognition, and computer-aided medical diagnosis, etc. Various Computer vision techniques have been used. More recently, Deep learning architectures demonstrated impressive results. This paper aims to investigate combining two feature extraction methods: Handcrafted-based and CNN-based in a two-stream neural network architecture. We believe that Statistical features could enhance the performance of the CNN architecture, especially in the case of small datasets. To test our approach we used two challenging datasets, the Describable Textures Dataset (DTD) and Flicker Material Database (FMD). Results showed that our two-stream neural network which has an image as a first stream and a statistical feature vector as a second stream achieve better results than a Convolutional neural network achieved with just the RGB image as input. The Xception network [9] combined with SIFT-FV demonstrated an accuracy superiority for both datasets.