Bobur Rashidovich Saidov

Dr. Bobur Rashidovich Saidov

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 September 2024
Verified
Conference Paper
Closed access
Davlatyor MenglievDavlatyor Mengliev
Davlatyor Mengliev
Vladimir BarakhninVladimir Barakhnin
Vladimir Barakhnin
Bobur Rashidovich SaidovBobur Rashidovich Saidov
Bobur Rashidovich Saidov
MAMukhammadjon Atakhanov
Mukhammadjon Atakhanov
Mukhriddin EshkulovMukhriddin Eshkulov
Mukhriddin Eshkulov
BIBahodir B. Ibragimov
Bahodir B. Ibragimov
In the research paper, the authors propose an algorithm for detecting poetic elements, words, and phrases in Uzbek texts. In addition, during analyzing the algorithm classifies poetic tokens in of the of 2 poetic genres (comedy, drama). In particular, the researchers trained a language model of artificial intelligence with an architectural combination of a convolutional neural network and long short-term memory. A custom language corpus was used as training data. This corpus was formed from more than 3,600 literary sentences, which in turn were selected from U zbek poetic works of the 20th century. Words and phrases in the corpus sentences were tagged according to the BIO-scheme, which the authors also talk about in the corresponding section of the article. Moreover, the authors also included information about the morphology of the U zbek language so that the context of this study was more understandable to all readers. In addition, the authors also conducted a comparative analysis of existing alternative works, where they provide the features of each similar work.

27 June 2025
Verified
Conference Paper
Closed access
Bobur Rashidovich SaidovBobur Rashidovich Saidov
Bobur Rashidovich Saidov
Vladimir BarakhninVladimir Barakhnin
Vladimir Barakhnin
URUlugbek T. Rixsibayev
Ulugbek T. Rixsibayev
OSOgabek O. Sobirov
Ogabek O. Sobirov
KBKhikmat M. Bekchanov
Khikmat M. Bekchanov
ESElbek J. Sharipov
Elbek J. Sharipov
This paper investigates integrated approaches for automatic detection of named entities (Named Entity Recognition, NER) and sentiment analysis in Uzbek texts. The development proposes an architecture that combines NER and sentiment analysis, taking into account the morphological features of the Uzbek language and the lack of a sufficiently annotated database. The study performed named entity detection using the BiLSTM+CRF model (F1=0.82) and sentiment analysis using the Uzbek-tuned XLM-RoBERTa model (accuracy=89%). The results showed that the NER system allows for more accurate context analysis in sentiment assessment. In particular, geographical entities such as “Tashkent city” were evaluated as positive with 95% accuracy, and sentences such as “The new performance at the Navoi Theater was great” were evaluated as positive with 91% reliability. At the same time, dialectical expressions and ambiguous words created difficulties in the analysis. The results of the study can serve as a basis for improving NER and tonality analysis in Uzbek. It is recommended that future work be carried out to adapt transformer models to the Uzbek language and expand annotated corpora.

22 December 2025 in The Herald of the Siberian State University of Telecommunications and Information Science
Verified
Journal Article
Open access
PDF file
Bobur Rashidovich SaidovBobur Rashidovich Saidov
Bobur Rashidovich Saidov
Vladimir BarakhninVladimir Barakhnin
Vladimir Barakhnin
This paper presents a comparative analysis of machine learning (SVM), deep learning (LSTM), and transformer-based (BERT) models for sentiment classification in Uzbek texts, enhanced by Named Entity Recognition (NER). The study addresses the challenge of accurately detecting sentiment in morphologically complex languages with limited resources, focusing on Uzbek–a Turkic language with rich agglutinative structures. A dataset of 10,000 user-generated comments from social platforms was annotated using a hybrid approach: manual labeling for sentiment (positive, negative, neutral) and a CRF-based NER system to identify entities (e.g., brands, locations, public figures). The integration of NER features aimed to resolve contextual ambiguities, such as distinguishing between "I love Samarkand’s history" (positive) and "Samarkand’s traffic is unbearable" (negative). Experimental results demonstrate that BERT, fine-tuned on Uzbek text, achieved the highest accuracy (90.2%) by leveraging contextualized embeddings to align entities with sentiment. LSTM showed competitive performance (85.1%) in sequential pattern learning but required extensive training data. SVM, while computationally efficient, lagged at 78.3% accuracy due to its inability to capture nuanced linguistic dependencies. The findings emphasize the critical role of NER in low-resource languages for disambiguating sentiment triggers and propose practical guidelines for deploying BERT in real-world applications, such as customer feedback analysis. Limitations, including data scarcity and computational costs, are discussed to inform future research on optimizing lightweight models for Uzbek NLP tasks.

1 February 2026 in Data
Verified
Journal Article
Open access
PDF file
Bobur Rashidovich SaidovBobur Rashidovich Saidov
Bobur Rashidovich Saidov
Vladimir BarakhninVladimir Barakhnin
Vladimir Barakhnin
SMShohrux Uktam ugli Madirimov
Shohrux Uktam ugli Madirimov
UIUmidjon Farkhodjonovich Ibragimov
Umidjon Farkhodjonovich Ibragimov
SMShakhboz Meylikulov
Shakhboz Meylikulov
ZFZarnigor Fayzullaeva
Zarnigor Fayzullaeva
This data descriptor presents two fully synthetic corpora for sentiment analysis and named entity recognition (NER) in Uzbek. The first corpus contains 12,000 hybrid synthetic sentences generated from templates with lexical randomization, automatic insertion of named entities (PER/ORG/LOC), lexicon-based polarity scoring, and a controlled emoji distribution. The second corpus includes 3000 “manual-style” sentences designed to resemble short, naturally structured messages. Although the manual-style subset was initially intended to be emoji-free, the released version includes a 39.6% emoji presence (sentences containing at least one emoji) to maintain comparability in emotional markers across corpora. Both corpora are released in CSV, XLSX, and JSONL formats and share a unified schema (id, text, sentiment, entities, entity_type, polarity_score, polarity_source, token_count, emojis, emoji_position, emoji_sentiment, conflict_flag, sentiment_from_polarity_score, split). The dataset is publicly available via Mendeley Data (DOI: 10.17632/y2d5pcyrzz.3).

Verified
Journal Article
Open access
PDF file
Bobur Rashidovich SaidovBobur Rashidovich Saidov
Bobur Rashidovich Saidov
Vladimir BarakhninVladimir Barakhnin
Vladimir Barakhnin
RSRakhmon Saparbaev
Rakhmon Saparbaev
ZNZayniddin Narmuratov
Zayniddin Narmuratov
RMRustamova Manzura
Rustamova Manzura
AAAnorgul Atajanova
Anorgul Atajanova
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.

22 December 2025 in The Herald of the Siberian State University of Telecommunications and Information Science
Verified
Journal Article
Open access
PDF file
Bobur Rashidovich SaidovBobur Rashidovich Saidov
Bobur Rashidovich Saidov
Vladimir BarakhninVladimir Barakhnin
Vladimir Barakhnin
This article presents a study on the creation of an emotional lexicon of the Uzbek language, taking into account its cultural and linguistic specificities. The emotional vocabulary of Uzbek reflects a complex interplay between language, national identity, and collective emotional experience. The purpose of the research is to develop a culturally sensitive framework for identifying and categorizing emotional expressions in Uzbek across diverse communicative contexts. The study analyzes a wide range of traditional sources, including folklore, proverbs, and classical literature, alongside modern discourse materials such as media texts, online forums, and social networks. Particular attention is given to dialectal variations and culturally embedded connotations, such as the double meaning of words like “achchiq” (bitter taste and emotional bitterness) and “suyanmoq” (physical leaning and emotional reliance). Methodologically, the research integrates corpus linguistics, sociolinguistic surveys, and AIbased sentiment analysis to ensure both empirical validity and cultural depth. As a result, the study proposes a prototype of an Uzbek emotional lexicon that captures emotional polarity, intensity, and contextual usage. The practical applications of this lexicon include improving human-computer interaction (e.g., chatbots), enriching language learning tools, and supporting sociolinguistic and affective computing research. The findings underscore the necessity of a standardized, culturally informed affective lexicon in Uzbek for preserving linguistic richness and enhancing emotional nuance in digital communication. This work contributes to the broader field of Turkic affective linguistics and emphasizes the importance of integrating cultural semantics into computational models.