Medet Mukushev

Dr. Medet Mukushev

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
2 June 2020 in PLOS ONE
Verified
Journal Article
Open access
PDF file
VKVadim Kimmelman
Vadim Kimmelman
AIAlfarabi Imashev
Alfarabi Imashev
MMMedet Mukushev
Medet Mukushev
ASAnara Sandygulova
Anara Sandygulova
Facial expressions in sign languages are used to express grammatical functions, such as question marking, but can also be used to express emotions (either the signer’s own or in constructed action contexts). Emotions and grammatical functions can utilize the same articulators, and the combinations can be congruent or incongruent. For instance, surprise and polar questions can be marked by raised eyebrows, while anger is usually marked by lowered eyebrows. We investigated what happens when different emotions (neutral/surprise/anger) are combined with different sentence types (statement/polar question/wh-question) in Kazakh-Russian Sign Language (KRSL), replicating studies previously made for other sign languages. We asked 9 native signers (5 deaf, 4 hearing children of deaf adults) to sign 10 simple sentences in 9 conditions (3 emotions * 3 sentence types). We used OpenPose software to track eyebrow position in the video recordings. We found that emotions and sentence types influence eyebrow position in KRSL: eyebrows are raised for polar questions and surprise, and lowered for anger. There are also some interactions between the two factors, as well as some differences between hearing and deaf native signers, namely a smaller effect of polar questions for the deaf group, and a different interaction between emotions and wh-question marking in the two groups. We thus find evidence for the complex influences on non-manual behavior in signers of sign languages, and showcase a quantitative approach to this field.

12 September 2022 in PLOS ONE
Verified
Journal Article
Open access
PDF file
MMMedet Mukushev
Medet Mukushev
AUAidyn Ubingazhibov
Aidyn Ubingazhibov
AKAigerim Kydyrbekova
Aigerim Kydyrbekova
AIAlfarabi Imashev
Alfarabi Imashev
VKVadim Kimmelman
Vadim Kimmelman
ASAnara Sandygulova
Anara Sandygulova
This paper presents a new large-scale signer independent dataset for Kazakh-Russian Sign Language (KRSL) for the purposes of Sign Language Processing. We envision it to serve as a new benchmark dataset for performance evaluations of Continuous Sign Language Recognition (CSLR) and Translation (CSLT) tasks. The proposed FluentSigners-50 dataset consists of 173 sentences performed by 50 KRSL signers resulting in 43,250 video samples. Dataset contributors recorded videos in real-life settings on a wide variety of backgrounds using various devices such as smartphones and web cameras. Therefore, distance to the camera, camera angles and aspect ratio, video quality, and frame rates varied for each dataset contributor. Additionally, the proposed dataset contains a high degree of linguistic and inter-signer variability and thus is a better training set for recognizing a real-life sign language. FluentSigners-50 baseline is established using two state-of-the-art methods, Stochastic CSLR and TSPNet. To this end, we carefully prepared three benchmark train-test splits for models’ evaluations in terms of: signer independence, age independence, and unseen sentences. FluentSigners-50 is publicly available at https://krslproject.github.io/FluentSigners-50/

5 March 2021
Verified
Book Chapter
Open access
PDF file
AAAlikhan Abutalipov
Alikhan Abutalipov
AJAigerim Janaliyeva
Aigerim Janaliyeva
MMMedet Mukushev
Medet Mukushev
ACAntonio Cerone
Antonio Cerone
ASAnara Sandygulova
Anara Sandygulova
This paper showcases the work that aims at building a user-friendly mobile application of a reverse dictionary to translate sign languages to spoken languages. The concept behind the reverse dictionary is the ability to perform a video-based search by demonstrating a handshape in front of a mobile phone’s camera. The user would be able to use this feature in two ways. Firstly, the user would be able to search for a word by showing a handshape for the application to provide a list of signs that contain that handshape. Secondly, the user could fingerspell the word letter by letter in front of the camera for the application to return the sign that corresponds to that word. The user can then look through the suggested videos and see their written translations. To offer other functionalities, the application also has Search by Category and Search by Word options. Currently, the reverse dictionary supports translations from Russian Sign Language (RSL) to Russian language.

10 April 2026
Verified
Conference Paper
Closed access
AZAibar Zhangabay
Aibar Zhangabay
MMMedet Mukushev
Medet Mukushev
This study develops a fairness-aware recommendation system for YouTube gaming content addressing underrepresentation of Kazakh-language videos and emerging creators. A post-processing re-ranking approach was applied to content-based and collaborative filtering baselines using multiplicative boost factors for Kazakh content (1.25x) and emerging creators (1.25x), with channel diversity penalties. Evaluation on 700 videos from seven channels demonstrated 75 % Kazakh representation and 85 % emerging creator representation in top-100 recommendations, exceeding fairness targets while maintaining 98 % engagement retention. Crossvalidation confirmed algorithm stability with coefficient of variation below 0.05 for primary metrics.

30 March 2026 in Scientific Journal of Astana IT University
Verified
Journal Article
Open access
PDF file
MMMedet Mukushev
Medet Mukushev
AMAnar Mengdigali
Anar Mengdigali
TKTemirlan Karibekov
Temirlan Karibekov
MZManzura Zholdasova
Manzura Zholdasova
DADiana Arman
Diana Arman
AKAlmira M Kustubayeva
Almira M Kustubayeva
Drowsiness reduces efficiency in perceptual processing, reaction time, and executive control, posing risks in safety-critical domains such as driving and long-duration monitoring tasks. EEG-based fatigue detection has emerged as a powerful approach for quantifying early neurophysiological signs of vigilance decline, yet many proposed algorithms are insufficiently evaluated in strictly subject-independent conditions. To address this gap, we systematically compare classical machine learning models, Riemannian geometry-based classification, and compact deep neural architectures on a publicly available electroencephalography (EEG) dataset containing 11 subjects. We employ a rigorous leave-one-subject-out (LOSO) protocol, ensuring that no individual contributes information simultaneously to the training and test sets. The study evaluates logistic regression, support vector machines with radial-basis kernels, random forests, a Log-Euclidean Riemannian classifier, EEGNet, a transformer encoder, and a bidirectional long short-term memory (BiLSTM) with temporal attention. Across folds, accuracy and macro-F1 scores were calculated and summarized with mean and standard deviation. The BiLSTM-attention model achieved the highest performance (accuracy ; macro-F1 ) but only moderately exceeded EEGNet and the classical baselines. Wilcoxon signed-rank tests revealed no significant difference between EEGNet and BiLSTM (), although BiLSTM significantly outperformed the transformer model (). Analysis of error structure demonstrated a notable asymmetry with 295 false positives and 184 false negatives aggregated across folds. Band-specific analysis revealed theta activity as the strongest contributor to class separation, followed by delta and alpha rhythms. Channel-importance analysis indicated that posterior and paracentral regions were consistently more informative. These findings highlight that model complexity does not guarantee superior performance in small datasets with large inter-subject variability. The study provides a transparent, fully reproducible baseline for future fatigue-classification research and demonstrates the practical relevance of compact architectures and Riemannian geometry in low-data conditions.

1 July 2026 in Economies
Verified
Journal Article
Open access
PDF file
NYNurgul Yesmagulova
Nurgul Yesmagulova
NINurkhat Ibadildin
Nurkhat Ibadildin
RIRymkul Ismailova
Rymkul Ismailova
MMMedet Mukushev
Medet Mukushev
ZMZhandos Mussabekov
Zhandos Mussabekov
This article examines the maturation of the Astana Hub technology park and its association with Kazakhstan’s national digital competitiveness during the country’s transition from resource dependency toward a higher value-added digital economy. As Central Asia’s largest international technology park, Astana Hub provides a policy-relevant case of a state-initiated open-innovation ecosystem. To proxy the national enabling environment, we construct the Kazakhstan Digital Performance Index (KDPI), a composite measure that normalizes and aggregates three publicly available global indices: the UN E-Government Development Index (EGDI), the Global Innovation Index (GII, WIPO), and the Network Readiness Index (NRI, Portulans Institute). Using an Interrupted Time Series (ITS) design over 2010–2024, we estimate the change in the level and slope of national IT service exports associated with the establishment of the Hub in 2018. The analysis identifies a statistically significant post-2018 increase in the export-growth slope of approximately $112.8 million per year relative to the pre-intervention trend, alongside a near 20-fold rise in IT service exports between 2020 and 2024. By contrast, the composite KDPI shows no comparable acceleration, and an exploratory correlation analysis over the short post-2018 window indicates a positive association between Hub growth and e-government infrastructure but a negative association with the GII. We interpret these patterns cautiously as descriptive evidence consistent with a temporary “island of efficiency,” in which commercial scaling has outpaced economy-wide innovation diffusion, rather than as confirmation of a bidirectional feedback loop or of causal national-level human-capital effects, which the single-country quasi-experimental design cannot establish. The paper discusses policy options for diffusing Hub capabilities into the wider economy and sets out the limitations of the design.