arXiv:2510.15843v2 Announce Type: replace 
Abstract: Accurately detecting sentiment polarity and intensity in product reviews and social media posts remains challenging due to informal and domain-specific language. To address this, we propose a novel hybrid lexicon-fuzzy-transformer framework that combines rule-based heuristics, contextual deep learning, and fuzzy logic to generate continuous sentiment scores reflecting both polarity and strength. The pipeline begins with VADER-based initial sentiment estimations, which are refined through a two-stage adjustment process. This involves leveraging confidence scores from DistilBERT, a lightweight transformer and applying fuzzy logic principles to mitigate excessive neutrality bias and enhance granularity. A custom fuzzy inference system then maps the refined scores onto a 0 to 1 continuum, producing expert)like judgments. The framework is rigorously evaluated on four domain-specific datasets. food delivery, e-commerce, tourism, and fashion. Results show improved alignment with user ratings, better identification of sentiment extremes, and reduced misclassifications. Both quantitative metrics (distributional alignment, confusion matrices) and qualitative insights (case studies, runtime analysis) affirm the models robustness and efficiency. This work demonstrates the value of integrating symbolic reasoning with neural models for interpretable, finegrained sentiment analysis in linguistically dynamic domains.

تم اقتراح إطار هجين جديد يجمع بين القاموس والمنطق الضبابي والمحولات لتحسين تفسير المشاعر في مراجعات المنتجات ومنشورات وسائل التواصل الاجتماعي. يدمج هذا الإطار بين القواعد القائمة على القواعد، والتعلم العميق السياقي، والمنطق الضبابي لإنتاج درجات مستمرة من المشاعر تعكس كل من القطبية والقوة، بدءًا من تقديرات مستندة إلى VADER وتكريرها من خلال عملية تعديل من مرحلتين تشمل DistilBERT ومبادئ المنطق الضبابي.

Se ha propuesto un nuevo marco híbrido de léxico-fuzzy-transformer para mejorar la interpretación de sentimientos en reseñas de productos y publicaciones en redes sociales. Este marco integra heurísticas basadas en reglas, aprendizaje profundo contextual y lógica difusa para producir puntuaciones de sentimiento continuas que reflejan tanto la polaridad como la intensidad, comenzando con estimaciones basadas en VADER y refinándolas a través de un proceso de ajuste en dos etapas que involucra a DistilBERT y principios de lógica difusa.

Un nouveau cadre hybride lexique-flou-transformateur a été proposé pour améliorer l'interprétation des sentiments dans les avis de produits et les publications sur les réseaux sociaux. Ce cadre intègre des heuristiques basées sur des règles, un apprentissage profond contextuel et une logique floue pour produire des scores de sentiment continus reflétant à la fois la polarité et l'intensité, en commençant par des estimations basées sur VADER et en les affinant par un processus d'ajustement en deux étapes impliquant DistilBERT et des principes de logique floue.

A novel hybrid lexicon-fuzzy-transformer framework has been proposed to enhance sentiment interpretation in product reviews and social media posts. This framework integrates rule-based heuristics, contextual deep learning, and fuzzy logic to produce continuous sentiment scores that reflect both polarity and strength, starting with VADER-based estimations and refining them through a two-stage adjustment process involving DistilBERT and fuzzy logic principles.

Enhanced Sentiment Interpretation via a Lexicon-Fuzzy-Transformer Framework

arXiv:2511.07299v2 Announce Type: replace 
Abstract: Video anomaly understanding (VAU) aims to provide detailed interpretation and semantic comprehension of anomalous events within videos, addressing limitations of traditional methods that focus solely on detecting and localizing anomalies. However, existing approaches often neglect the deeper causal relationships and interactions between objects, which are critical for understanding anomalous behaviors. In this paper, we propose VADER, an LLM-driven framework for Video Anomaly unDErstanding, which integrates keyframe object Relation features with visual cues to enhance anomaly comprehension from video. Specifically, VADER first applies an Anomaly Scorer to assign per-frame anomaly scores, followed by a Context-AwarE Sampling (CAES) strategy to capture the causal context of each anomalous event. A Relation Feature Extractor and a COntrastive Relation Encoder (CORE) jointly model dynamic object interactions, producing compact relational representations for downstream reasoning. These visual and relational cues are integrated with LLMs to generate detailed, causally grounded descriptions and support robust anomaly-related question answering. Experiments on multiple real-world VAU benchmarks demonstrate that VADER achieves strong results across anomaly description, explanation, and causal reasoning tasks, advancing the frontier of explainable video anomaly analysis.

تم تقديم إطار جديد يسمى VADER لتحسين فهم الشذوذ في الفيديو (VAU) من خلال دمج العلاقات السببية والتفاعلات بين الكائنات في مقاطع الفيديو. تستخدم هذه الطريقة نموذج لغة كبير (LLM) لتوفير تفسير أكثر دقة للأحداث الشاذة، متجاوزة الطرق التقليدية التي غالبًا ما تتجاهل العوامل السياقية الأعمق.

Se ha introducido un nuevo marco llamado VADER para mejorar la Comprensión de Anomalías en Videos (VAU) al integrar relaciones causales e interacciones entre objetos en videos. Este enfoque utiliza un modelo de lenguaje grande (LLM) para proporcionar una interpretación más matizada de eventos anómalos, yendo más allá de los métodos de detección tradicionales que a menudo pasan por alto factores contextuales más profundos.

Un nouveau cadre nommé VADER a été introduit pour améliorer la compréhension des anomalies vidéo (VAU) en intégrant les relations causales et les interactions entre objets dans les vidéos. Cette approche utilise un modèle de langage large (LLM) pour fournir une interprétation plus nuancée des événements anormaux, allant au-delà des méthodes de détection traditionnelles qui négligent souvent des facteurs contextuels plus profonds.

A new framework named VADER has been introduced to enhance Video Anomaly Understanding (VAU) by integrating causal relationships and object interactions within videos. This approach utilizes a large language model (LLM) to provide a more nuanced interpretation of anomalous events, moving beyond traditional detection methods that often overlook deeper contextual factors.

Enhanced Sentiment Interpretation via a Lexicon-Fuzzy-Transformer Framework

Was this article worth reading? Share it

finlight.me

Tweeteasy

Humanize AI

LucidQuery AI

Reviewradar

AI Humanizer

Ready to build your own newsroom?