This work proposes a hybrid Uzbek sentiment analysis model (sometimes referred to as tonality analysis in the local literature) that integrates contextual text representations with named-entity information from an NER module and emoji-based emotional cues that are common in short online messages. To provide a comprehensive baseline comparison, we evaluate seven approaches—SVM, LSTM, mBERT, XLM-RoBERTa-base, mDeBERTa-v3, LaBSE, and the proposed hybrid model—covering both classical machine learning and modern multilingual transformer architectures for low-resource sentiment tasks. The overall pipeline begins with Uzbek-specific text normalization to reduce noise from informal spellings, transliteration variants, and inconsistent apostrophe usage. In parallel, the system performs explicit emoji extraction to capture affective signals that are often expressed non-verbally in social media texts. Next, we construct three complementary feature streams: a context encoder for sentence-level semantics, NER-driven entity features that encode entity mentions and types, and an emotion module that models emoji priors and their interaction with contextual meaning. These streams are fused into a unified representation and fed to a final classifier to predict sentiment polarity. Experiments on an Uzbek test set demonstrate that the hybrid model reaches an F1-score of 0.92, consistently outperforming text-only baselines. The results indicate that entity-aware and emoji-informed features improve robustness under sarcasm/irony, mixed sentiment with multiple targets, and orthographic noise, making the approach suitable for social media analytics, public opinion monitoring, customer feedback triage, and recommendation-oriented text mining.
13 January 2025 in IEEE Transactions on Instrumentation and Measurement
Energy-resolved photon-counting detectors (PCD) present the ability to enhance material identification in different x-ray radiography and tomography systems employed for industrial and security applications. Their potential has been proven to be higher than the dual-energy scintillator detectors broadly used nowadays. However, detector effects, like photon pileup and charge sharing, distort the energy spectrum in PCDs. These effects lead to a significant performance deterioration of the PCDs when used for material identification where accurate measurements of multi-energy linear attenuation coefficients (LAC) are needed. We have proposed a method that rapidly corrects the measured raw LAC curve for significant spectral distortions in the image reconstruction domain. This method is based on polynomial fitting independently for each energy channel. The method has been validated with experimental data, and the calibration of the method is based on attenuation measurements of common materials with usual laboratory x-ray sources, allowing the method to be employed in any x-ray instrument. The experiments present that the method largely enhances the extraction accuracy of the multi-energy LACs, making it helpful for industrial, medical, and security purposes where PCDs are used.