Background: Clear language makes communication easier between any two parties. A layman may have difficulty communicating with a professional due to not understanding the specialized terms common to the domain. In healthcare, it is rare to find a layman knowledgeable in medical terminology which can lead to poor understanding of their condition and/or treatment. To bridge this gap, several professional vocabularies and ontologies have been created to map laymen medical terms to professional medical terms and vice versa. Objective: Many of the presented vocabularies are built manually or semi-automatically requiring large investments of time and human effort and consequently the slow growth of these vocabularies. In this paper, we present an automatic method to enrich laymen’s vocabularies that has the benefit of being able to be applied to vocabularies in any domain. Methods: Our entirely automatic approach uses machine learning, specifically Global Vectors for Word Embeddings (GloVe), on a corpus collected from a social media healthcare platform to extend and enhance consumer health vocabularies. Our approach further improves the consumer health vocabularies by incorporating synonyms and hyponyms from the WordNet ontology. The basic GloVe and our novel algorithms incorporating WordNet were evaluated using two laymen datasets from the National Library of Medicine (NLM), Open-Access Consumer Health Vocabulary (OAC CHV) and MedlinePlus Healthcare Vocabulary. Results: The results show that GloVe was able to find new laymen terms with an F-score of 48.44%. Furthermore, our enhanced GloVe approach outperformed basic GloVe with an average F-score of 61%, a relative improvement of 25%. Furthermore, the enhanced GloVe showed a statistical significance over the two ground truth datasets with P < 0.001. Conclusions: This paper presents an automatic approach to enrich consumer health vocabularies using the GloVe word embeddings and an auxiliary lexical source, WordNet. Our approach was evaluated used healthcare text downloaded from MedHelp.org, a healthcare social media platform using two standard laymen vocabularies, OAC CHV, and MedlinePlus. We used the WordNet ontology to expand the healthcare corpus by including synonyms, hyponyms, and hypernyms for each layman term occurrence in the corpus. Given a seed term selected from a concept in the ontology, we measured our algorithms’ ability to automatically extract synonyms for those terms that appeared in the ground truth concept. We found that enhanced GloVe outperformed GloVe with a relative improvement of 25% in the F-score.

The fast growth of artificial intelligence technologies, especially language processing technology has obscured the lines in between human-generated text comparing to chatbot-generated message. Recognizing which generated such, a text is essential for applications like information generating and manipulated text in order to guarantee authenticity between communicated parties. This research applies to a set of machine learning models to identify text as either human-written or chatbot-generated. The methodology of this research starts with a dataset including text generated from different Large Language Models (LLMs) along with a text generated by a human. After that, Tf-Idf ranking vectorization was used to define word embedding has and represent the text numerically. Then, different Machine Learning (ML) models leveraged recognize whether a human or a chatbot generated a text. The ML models applied include Logistic Regression, Random Forest, Decision Tree, Gradient Boosting, Naïve Bayes, and XGBoost. For this study accuracy, precision, recall, F1-score were used to evaluate the system. The dataset first was split into 80% for training and 20% for testing. Out of all implemented models, the Random Forest model reported the best with accuracy of 88%. Logistic Regression reported a close accuracy of 85%. The Random Forest model showed an 8% improvement compared to previous studies that reported an accuracy of 80%. Confusion matrices revealed that the Random Forest model provided high precision and recall, minimizing classification misleading of human or chatbot text. The research focused on studying the ability of ML models in identifying human vs. chatbot-generated text. The results showed the RF model was the best among other models with 88% accuracy. This accuracy shows a possible usage of such models in real-world applications that requires the confidentiality of human writing.

Word vector representations open up new opportunities to extract useful information from unstructured text. Defining a word as a vector made it easy for the machine learning algorithms to understand a text and extract information from. Word vector representations have been used in many applications such word synonyms, word analogy, syntactic parsing, and many others. GloVe, based on word contexts and matrix vectorization, is an effective vector-learning algorithm. It improves on previous vector-learning algorithms. However, the GloVe model fails to explicitly consider the order in which words appear within their contexts. In this paper, multiple methods of incorporating word order in GloVe word embeddings are proposed. Experimental results show that our Word Order Vector (WOVe) word embeddings approach outperforms unmodified GloVe on the natural language tasks of analogy completion and word similarity. WOVe with direct concatenation slightly outperformed GloVe on the word similarity task, increasing average rank by 2%. However, it greatly improved on the GloVe baseline on a word analogy task, achieving an average 36.34% improvement in accuracy.

Malfunctioning of photovoltaic (PV) systems is a main issue affecting solar panels and other related components. Detecting such issues early leads to efficient energy production with low maintenance costs and high system performance consistency. This paper proposed an ensemble model (EM) for fault detection (FD) in PV systems. The proposed model utilized advanced machine learning algorithms containing random forest (RF), k-nearest neighbors (KNN), and gradient boosting (GB). Traditional approaches often do not handle the several situations that PV systems can have. Our EM leveraged the power of GB’s algorithm in handling complex data patterns through iterative boosting, KNN’s capability in capturing local data structures, and RF’s strength in handling overfitting and noise through its tree structure randomness. Combining these models enhanced fault detection capabilities, providing excellent accuracy compared to individual models. To evaluate the performance of our EM, different experiments were conducted. The results demonstrated substantial improvements in detection fault, achieving an accuracy rate of 95%. This accuracy rate considered high underscores the model’s capability to handle fault detection of PV systems, posing a consistent solution for instant fault detection and maintenance scheduling.

With the advent of attention mechanisms and the development of transformer-based architectures, a new era of large language models (LLMs) has emerged. These models have given rise to a wide range of fine-tuned applications capable of providing high-quality and informative responses. ChatGPT, a chatbot developed by OpenAI using a popular LLM called GPT 3.5, has demonstrated its capability to provide valuable information in a range of fields. The renewable energy field is a prominent field, which aims to explore alternative sources of energy that are more affordable and effective. In this study, we examine how accurate ChatGPT is when it comes to providing general, non-technical information on renewable energy compared to human experts in this field. A set of prompts was presented to ChatGPT as well as to a human expert. Responses have been collected and evaluated using a set of similarity measures. Further to this, a pre-trained Google vector has been applied to emphasize semantic similarity, and then a more sophisticated LLM competitor evaluation, Gemini, has been employed to evaluate the degree of similarity between ChatGPT responses and those of human experts. It surprisingly comes to a point where ChatGPT responses were more accurate and relevant than those of human experts’ responses on renewable energy prompts. This study concludes that ChatGPT is a promising and supportive resource for renewable-energy information, offering responses remarkably close to those of a human expert.