Large language models (LLMs) can store factual knowledge within their parameters and have achieved superior results in question-answering tasks. However, challenges persist in providing provenance for their decisions and keeping their knowledge up to date. Some approaches aim to address these challenges by combining external knowledge with parametric memory. In contrast, our proposed QA-RAG solution relies solely on the data stored within an external knowledge base, specifically a dense vector index database. In this paper, we compare RAG configurations using two LLMs—Llama 2b and 13b—systematically examining their performance in three key RAG capabilities: noise robustness, knowledge gap detection, and external truth integration. The evaluation reveals that while our approach achieves an accuracy of 83.3%, showcasing its effectiveness across all baselines, the model still struggles significantly in terms of external truth integration. These findings suggest that considerable work is still required to fully leverage RAG in question-answering tasks.

Public transport agencies must triage thousands of multilingual complaints every day, yet the cost of training and serving fine-grained sentiment analysis models limits real-time deployment. The proposed “one encoder, any facet” framework therefore offers a reproducible, resource-efficient alternative to heavy fine-tuning for domain-specific sentiment analysis or opinion mining tasks on digital service data. To the best of our knowledge, we are the first to test this paradigm on operational multilingual complaints, where public transport agencies must prioritize thousands of Russian- and Kazakh-language messages each day. A human-labelled corpus of 2400 complaints is embedded with five open-source universal models. Obtained embeddings are matched to semantic “anchor” queries that describe three distinct facets: service aspect (eight classes), implicit frustration, and explicit customer request. In the strict zero-shot setting, the best encoder reaches 77% accuracy for aspect detection, 74% for frustration, and 80% for request; taken together, these signals reproduce human four-level priority in 60% of cases. Attaching a single-layer logistic probe on top of the frozen embeddings boosts performance to 89% for aspect, 83–87% for the binary facets, and 72% for end-to-end triage. Compared with recent fine-tuned sentiment analysis systems, our pipeline cuts memory demands by two orders of magnitude and eliminates task-specific training yet narrows the accuracy gap to under five percentage points. These findings indicate that a single frozen encoder, guided by handcrafted anchors and an ultra-light head, can deliver near-human triage quality across multiple pragmatic dimensions, opening the door to low-cost, language-agnostic monitoring of digital-service feedback.

This study introduces an unsupervised term extraction approach that combines non-negative matrix factorization (NMF) with word embeddings. Inspired by a pioneering semantic NMF method that employs regularization to jointly optimize document–word and word–word matrix factorizations for document clustering, we adapt this strategy for term extraction. Typically, a word–word matrix representing semantic relationships between words is constructed using cosine similarities between word embeddings. However, it has been established that transformer encoder embeddings tend to reside within a narrow cone, leading to consistently high cosine similarities between words. To address this issue, we replace the conventional word–word matrix with a word–seed submatrix, restricting columns to ‘domain seeds’—specific words that encapsulate the essential semantic features of the domain. Therefore, we propose a modified NMF framework that jointly factorizes the document–word and word–seed matrices, producing more precise encoding vectors for words, which we utilize to extract high-relevancy topic-related terms. Our modification significantly improves term extraction effectiveness, marking the first implementation of semantically enhanced NMF, designed specifically for the task of term extraction. Comparative experiments demonstrate that our method outperforms both traditional NMF and advanced transformer-based methods such as KeyBERT and BERTopic. To support further research and application, we compile and manually annotate two new datasets, each containing 1000 sentences, from the ‘Geography and History’ and ‘National Heroes’ domains. These datasets are useful for both term extraction and document classification tasks. All related code and datasets are freely available.

Accurate bus travel time prediction is a critical requirement for managing urban mobility and enabling reliable public transit operations. While existing research often relies on the fusion of multi-source, heterogeneous data, such approaches are less feasible in cities with limited access to external information. To address this gap, this study proposes a dataefficient approach to travel time prediction using only commonly available historical transit data. Specifically, a Hybrid Bidirectional Long Short-Term Memory (BiLSTM) model is applied to predict bus travel times without requiring external contextual inputs. The model is evaluated on real-world bus routes in Kandy and Astana, where it consistently outperforms attention-based ConvLSTM and stacked ensemble baselines. On Route 1 in Kandy, the proposed model achieves a Mean Absolute Error (MAE) of 5.44 seconds and a Mean Absolute Percentage Error (MAPE) of $\mathbf{2. 4 1 \%}$, markedly better than competing methods. Across three routes in Astana, the model reports MAE values between 42.16 and 50.45 seconds and MAPE between 34.23 % and 39.84 %, improving upon baselines by up to 20 %. Cross-validation further confirms the robustness of the approach, with consistently lower errors across all folds. The results show that accurate bus travel time prediction is achievable, even in a variety of data-limited environments, offering a practical and versatile solution for transport agencies.

Citizen complaint platforms generate large volumes of multilingual textual data that can support urban service management, yet conventional text-based representations typically analyze each complaint independently and do not explicitly capture relationships with the surrounding urban context. This study investigates whether relational information can complement semantic text representations for the thematic analysis of citizen complaints. A multi-view heterogeneous graph was constructed from municipal requests collected through the iKOMEK109 service in Astana, Kazakhstan, integrating complaint texts with annotated urban entities, spatial and contextual entity relations, and semantic complaint–complaint relations. Entity, Spatial, and Semantic graph views were evaluated using R-GCN, HGT, and an attention-based multi-view R-GCN configuration, with BGE-M3 embeddings and Logistic Regression serving as the text-only baseline. The representations were evaluated for infrastructure-domain classification and related-complaint retrieval. The bidirectional Multi-View R-GCN achieved a mean Macro F1-score of 0.7615 ± 0.0221 across three model-initialization seeds on the reference partition. However, a non-graph baseline combining complaint-text and directly linked entity embeddings achieved a Macro F1-score of 0.7810 on the same partition, compared with 0.7220 for the text-only baseline and 0.7381 for the R-GCN in the fixed-seed comparison, indicating that part of the observed benefit arises from additional entity information rather than graph propagation alone. Repeated-split analysis across five stratified data partitions further showed that the relative performance differences were sensitive to data partitioning. Under fine-grained human relevance assessment, BGE-M3 and the Hybrid representation achieved comparable retrieval performance. The corrected two-hop Hybrid representation obtained MRR@10 = 0.7596 compared with 0.7628 for BGE-M3, while producing small numerical improvements in Precision and NDCG. Query-level paired bootstrap confidence intervals included zero for all evaluated metrics, and paired permutation tests found no statistically significant differences, indicating that the graph-derived signal did not provide a statistically reliable retrieval advantage over the text representation.

Статья посвящена сбору доступной информации о водных объектах Восточно-Казахстанской области, полученной из открытых источников Национальной гидрометеорологической службы Республики Казахстан и Министерства по чрезвычайным ситуациям Республики Казахстан с целью формирования базы атрибутивных данных с пространственной привязкой в ГИС и формированию основы последующих исследований, направленных на разработку и внедрение моделей прогнозирования наводнений. Проведен обзор исследований, посвященных прогнозу наводнений в разных регионах Казахстана и в других странах, в результате чего сформированы критерии оценки данных и системы гидрологического мониторинга. Приводится характеристика гидрологического режима основных репрезентативных водных объектов Восточного Казахстана. На основе пространственного анализа сети гидрологических постов и известных зон затопления дана оценка пространственному охвату государственной системы наблюдений и определены рекомендации по ее расширению. Сделаны выводы о возможности применения собранных данных при создании прогнозных моделей.