这是我们最新精选的自然语言处理(NLP)方面的全球英文出版物和专利,涉及许多科学在线期刊,分类并侧重于自然语言处理、标记化、词干化、词法化、语音部分、命名实体识别和情感分析。
Deep Learning Approaches with Pretrained Embeddings
Published on 2025-10-21 by Zheqi Shen, Incheon Paik @MDPI
Abstract: In the field of natural language processing, depression forecasting from social media has gained extensive attention, as platforms like X (formerly Twitter) offer real-time user-generated content that can reflect psychological states. Common approaches typically rely on static text analysis, which overlooks how users’ emotions change over time. To address this limitation, we propose a temporal modeling approach that applies deep learning models to capture both textual and temporal [...]
Our summary: This study explores deep learning techniques for predicting depression from social media data. It emphasizes the importance of temporal modeling to capture emotional changes over time. The best-performing model achieved 99.4% accuracy by integrating Llama 2 embeddings with personalized temporal features.
Deep Learning, Pretrained Embeddings, Temporal Modeling, Depression Forecasting
Publication
Sentiment Analysis of Air Pollution in Jakarta Using the Bidirectional Encoder Representations from Transformers (BERT) Method
Published on 2025-10-20 by Shiva Aulia Anjani, Mya Septiani, Fawwaz Alfauzi, Sudin Saepudin, Muhamad Muslih, Carti Irawan @MDPI
Abstract: Air pollution represents a critical environmental challenge in Jakarta, significantly affecting public health and overall quality of life. This study aims to examine public sentiment regarding air pollution in Jakarta through the application of the Bidirectional Encoder Representations from Transformers (BERT) methodology. The selection of this method is based on its proficiency in comprehending contextual nuances within textual data, thereby facilitating a more precise sentiment analysis. The d[...]
Our summary: This study analyzes public sentiment on air pollution in Jakarta using the BERT method. It processes data from social media, particularly Twitter, to gauge public opinion. Findings reveal a predominance of negative sentiments influenced by government policies and environmental conditions.
Sentiment Analysis, Air Pollution, BERT, Jakarta
Publication
How to Conduct AI-Assisted (Large Language Model-Assisted) Content Analysis in Information Science and Cyber Security Research
Published on 2025-10-20 by Monica Therese Whitty @MDPI
Abstract: The advent of Large Language Models (LLMs) has revolutionised natural language processing, providing unprecedented capabilities in text generation and analysis. This paper examines the utility of Artificial-Intelligence-assisted (AI-assisted) content analysis (CA), supported by LLMs, as a methodological tool for research in Information Science (IS) and Cyber Security. It reviews current applications, methodological practices, and challenges, illustrating how LLMs can augment traditional approach[...]
Our summary: This paper explores the use of AI-assisted content analysis in Information Science and Cyber Security research. It highlights the integration of Large Language Models to enhance traditional qualitative data analysis methods. A hybrid workflow is proposed, emphasizing human oversight alongside AI capabilities for ethical and efficient research practices.
AI-assisted content analysis, Large Language Models, Information Science, Cyber Security
Publication
An Interpretable Soft-Sensor Framework for Dissertation Peer Review Using BERT
Published on 2025-10-17 by Meng Wang, Jincheng Su, Zhide Chen, Wencheng Yang, Xu Yang @MDPI
Abstract: Graduate education has entered the era of big data, and systematic analysis of dissertation evaluations has become crucial for quality monitoring. However, the complexity and subjectivity inherent in peer-review texts pose significant challenges for automated analysis. While natural language processing (NLP) offers potential solutions, most existing methods fail to adequately capture nuanced disciplinary criteria or provide interpretable inferences for educators. Inspired by soft-sensor, this st[...]
Our summary: This study presents a BERT-based model for analyzing dissertation peer reviews, addressing the challenges of complexity and subjectivity in evaluations. It integrates SHAP for interpretability, allowing educators to understand the importance of various evaluation dimensions. The proposed framework outperforms existing methods in key performance metrics while providing actionable insights for dissertation improvement.
NLP, BERT, interpretability, soft-sensor
Publication
Systems and methods for automated assessment of media content for sincerity
Patent published on the 2025-10-16 in WO under Ref WO2025217604 by WINNICK DANIEL [US] (Blumenthal Jonathan [us], Winnick Daniel [us])
Abstract: A non-transitory computer-readable medium storing instructions that, when executed by a processor, cause a system to perform a method for analyzing media content, the method comprising: ingesting, via one or more web scraping techniques, media content from one or more online sources; preprocessing, via one or more natural language processing tools, the media content comprising the steps of: tokenizing the media content into a plurality of components, removing stop words from the plurality of com[...]
Our summary: The system automates the assessment of media content for sincerity using web scraping techniques. It preprocesses the content through tokenization, stop word removal, lemmatization, and normalization. Finally, it analyzes sentiments, generates scores, and displays the final score on client devices.
automated assessment, media content, natural language processing, sentiment analysis
Patent
Auditory user interfaces and associated systems, methods, devices, and non-transitory computer-readable media
Patent published on the 2025-10-16 in WO under Ref WO2025217644 by IYO INC [US] (Treat Neil David [us], Ervin Christian [us], Rugolo Jason Steven [us])
Abstract: An auditory operating system designed to facilitate context-aware, audio-based user interactions, particularly with artificial intelligence agents or applications. An auditory operating system shell serves as the primary interface, managing and coordinating multiple specialized agents that handle specific domains like music streaming, scheduling, or home automation. Using natural language processing, the auditory operating system shell identifies the appropriate agent or application for a user's[...]
Our summary: The auditory operating system facilitates audio-based user interactions with AI applications. It manages multiple specialized agents for tasks like music streaming and home automation. The system enhances privacy, stability, and context management while reducing the need for specific commands.
Auditory interfaces, Natural language processing, Context-aware systems, AI agents
Patent
Systems and methods for optimizing the conversion of feedstock into renewable energy
Patent published on the 2025-10-16 in WO under Ref WO2025217280 by VANGUARD RENEWABLES HOLDINGS LLC [US] (Hanselman John [us], Sunchukeshava Rajkumar [us], Bacha Kaylyn [us], Walker Kimberly [us])
Abstract: Provided are systems and methods configured to optimize processing of feedstock sources into renewable energy. Optimization over conventional approaches can begin with systematic functionality at the first steps of delivering feedstock to various digester locations. Optimizing transport of materials to the various locations can significantly impact production efficiency and resultant greenhouse gases gas emissions stemming from such processing. Various embodiments resolve the technical issues of[...]
Our summary: Systems and methods are designed to optimize the conversion of feedstock into renewable energy. They utilize trained machine learning models to enhance resource efficiency and minimize greenhouse gas emissions. The approach focuses on improving transport logistics and processing of varied feedstock sources.
Optimization, Renewable Energy, Feedstock, Machine Learning
Patent
A Cloud-Based Sentiment Analysis System with a BERT Algorithm for Fake News on Twitter
Published on 2025-10-15 by Nadire Cavus, Bora Oktekin, Murat Goksu @MDPI
Abstract: The rapid spread of the global COVID-19 pandemic has rapidly changed people’s communication demands and shifted them to digital channels, thus increasing the use of social networks more than ever. However, the increased use of social networks has also led to emotional confusion that has emerged with the fake news problem. As a result of limited studies on fine-grained sentiment analysis of fake news, this study comprehensively presents a sentiment analysis of fake news across seven[...]
Our summary: This study presents a cloud-based sentiment analysis system called SA-ES that utilizes the BERT algorithm to analyze fake news on Twitter. It categorizes sentiment into seven main categories and achieves 99% accuracy with extensive training datasets. The system aims to address emotional confusion and contribute to a healthier society by understanding the sentiments of individuals sharing fake news.
Sentiment Analysis, BERT Algorithm, Fake News, Cloud-Based System
Publication








