Thai University RankingsRESEARCH RADAR
คัดผลงานใหม่จากแหล่งวิชาการสากล ตรวจสอบบทบาทนักวิจัยไทย และเล่าให้เห็นทั้งผลกระทบ ข้อจำกัด และแหล่งต้นทาง
สำรวจผลงานล่าสุด ↓The study developed a two-stage ThaiSook pipeline: a binary model first rejects non-produce images, after which a multiclass model identifies 43 fruit and vegetable types and three vegetable-heavy dishes. The 40,595-image dataset contained 11,210 labelled and 29,385 unlabelled user images. With label smoothing and temperature scaling, the pipeline reached 92.43% top-1, 94.56% top-3, and 94.91% top-5 accuracy; Noisy Student raised top-1 accuracy to 92.59%. However, results rely on one train–test split, without real on-device testing or evidence that classification improves dietary measurement or health outcomes.
Source-verified editorial analyses, ordered by latest publication date
In an international placebo-controlled phase 3 trial of 589 hospitalised adults, adding ensitrelvir to standard care did not significantly improve the day-60 recovery outcome. The result argues against broad inpatient benefit while leaving narrower timing or subgroup questions open.
The study optimises ultrasound-microwave extraction and aqueous two-phase separation for phenolics, flavonoids and in-vitro antioxidant activity from red water-lily petals and pollen. Process gains still require scale-up, composition, bioavailability and safety evidence.
The work couples a simple ionic-gelation synthesis with standardised antibacterial, phytotoxicity and biocompatibility testing. Its value is the safety workflow as much as the nanoparticle, although results remain sensitive to particle properties, feedstock and assay models.
A data-driven fuzzy controller adjusts a hybrid geothermal heat-pump system to changing demand rather than fixed rules, targeting comfort, energy use and ground-loop balance. Deployment evidence must be replicated across climates, buildings and tariff structures.
The team-level study links servant leadership with a compassionate climate and shared vision while examining leader grit and team size as boundary conditions. It offers testable management hypotheses, but perceptual and correlational data cannot establish causality.
Propolis from two stingless-bee species is profiled and screened against viral and cancer models to guide prototype products. Laboratory potency is only a starting point; geographic variation, active constituents, normal-tissue toxicity and dosing need further study.
Material testing combined with explainable AI predicts elastic properties of rubber-modified asphalt and ranks influential variables. The approach links accuracy with engineering interpretation, but must generalise across feedstocks, temperatures, ageing and field traffic.
This systematic review synthesises waterborne and vector-borne infection evidence after floods, considering timing, sanitation and surveillance pathways. It supports preparedness priorities, but heterogeneous flood definitions, diagnostics and study quality prevent a single universal risk estimate.
The review charts the shift from culture methods to 16S sequencing in studies of ageing, frailty and oral hypofunction. Recurring patterns are hypothesis-generating, but sampling, pipeline and confounding differences preclude a clinic-ready microbiome signature.
The study examines how native-speaker norms in Thai higher education can reproduce linguistic hierarchy and uses Global Englishes to frame more plural practices. Its policy value depends on evaluating implementation across diverse students, teachers and institutions.
The ecological analysis tests whether higher-order species interactions strengthen the latitudinal tree-diversity gradient. It challenges purely pairwise explanations of community structure, while requiring longitudinal and experimental confirmation across biomes.
Random Forest models compare environmental and spatial predictors of Thai Paphiopedilum distributions to prioritise surveys and conservation. Sampling bias, imperfect detection and changing habitat can make static suitability maps appear more certain than they are.
Thailand-affiliated bibliographic records. These are not full analyses; they are ordered by Impact Signal and should be verified against the original source.
Journal of Education and Innovation
Journal of Education and Innovation
International Journal of Building Urban Interior and Landscape Technology (BUILT)
Animals
International Journal of Molecular Sciences
Journalism
วารสารกฎหมาย จุฬาลงกรณ์มหาวิทยาลัย
วารสารกฎหมาย จุฬาลงกรณ์มหาวิทยาลัย
International Journal of Asian Studies
Egyptian Journal of Bronchology
Sustainability
Policy Design and Practice
European Journal of Social Sciences Education and Research
Global Health Action
European Journal of Social Sciences Education and Research
Systems Research and Behavioral Science
Discover Psychology
Forest Ecology and Management
City Territory and Architecture
Community and Social Development Journal
FWU Journal of Social Sciences
FWU Journal of Social Sciences
Technology in Language Teaching & Learning
Human Behavior, Development and Society
We use OpenAlex, Crossref, PubMed, DOI records, and publisher pages, then verify Thai researcher participation before publication.