SDU University research laboratory

We turn data into decisions.

MADRID Lab is a university research group working on machine learning, analytics and data research, with one goal: methods that help people make better, well-grounded decisions.

  1. Machine learning
  2. Analytics
  3. Data
  4. Research
  5. Intelligent
  6. Decisions
MADRID Lab members posing in front of the Freedom Holding Corp. sign
MADRID Lab team at Freedom Holding Corp.

About the lab

We bring together students and researchers who want to work on real data problems: building models, testing them honestly, and understanding when their results can be trusted.

Our work spans four connected areas, from learning algorithms to the decisions they support. Students join us for research projects, theses and internships.

Panel session on the main stage of AI & Digital Bridge 2026 in Astana

Highlight · October 2026

Through to the next stage at AI & Digital Bridge

At AI & Digital Bridge 2026 in Astana, lab members Aisulu, Aknur and Abilmansur presented and defended their team's project. The team advanced to the second qualification stage of the GovTech programme organised by the Ministry of Education.

  • Aisulu Igimbayeva
  • Aknur Mansurkhan
  • Abilmansur Alibekov

Latest news

All news
  • Lab team advances at AI & Digital Bridge 2026

    Aisulu Igimbayeva, Aknur Mansurkhan and Abilmansur Alibekov defended their team's project in Astana and moved on to the second qualification stage of the GovTech programme.

  • Three new members join the lab

    We welcome Abilmansur Alibekov, Akbike Zhurymbay and Aruzhan Iliyas to MADRID Lab.

  • Our new website is live

    Members, publications, open positions and news are now all in one place.

Want to do research with us?

We regularly take on students for research projects and theses. See what is open right now.

See openings

People

Members

The people behind MADRID Lab.

Founders & lab leadership

Bakhtiyor Meraliyev

Bakhtiyor Meraliyev

Professor

FounderLinkedIn
Aziza Nazarbayeva

Aziza Nazarbayeva

Head of MADRID Lab

FounderLinkedIn
Жансерік Сауатбек

Жансерік Сауатбек

FounderLinkedIn

Мерует Салижанова

Founder

Team

Aisulu Igimbayeva

Aisulu Igimbayeva

Research Assistant in DA Lab

Current memberLinkedIn
Aknur Mansurkhan

Aknur Mansurkhan

Research Assistant in DA Lab

Current memberLinkedIn
Abilmansur Alibekov

Abilmansur Alibekov

Research Assistant in DA Lab

New memberLinkedIn

Akbike Zhurymbay

Research Assistant in DA Lab

New memberLinkedIn

Aruzhan Iliyas

Research Assistant in DA Lab

New member

Former members

Әлихан Қошамет

Әлихан Қошамет

Ex memberLinkedIn
Ажар Кенжебек

Ажар Кенжебек

Ex memberLinkedIn
Алина Кенжегариева

Алина Кенжегариева

Ex memberLinkedIn
Лейлима Ақат

Лейлима Ақат

Ex memberLinkedIn

Айша Амирқызы

Ex member

Адель Кунуспекова

Ex memberLinkedIn

Нұрдәулет Серкетай

Ex memberLinkedIn

Жасұлан Тұрсынбай

Ex member

Explore our work

Research topics

Select a topic to see the publications connected to it.

Research output

Publications

Papers, proceedings and other research output from lab members.

2026

Using Machine Learning to Predict Arterial Hypertension From a Clinical Dataset

Aisulu Igimbayeva, Adel Kunuspekova, Bakhtiyor Meraliyev. 2026 18th International Conference on Electronics, Computer, and Computation (ICECCO), 2026.

  • Clinical prediction models

2025

Search Trees: A Review and Comparative Study

Mohammed Alaa Ala’anzy, Aknur Mansurkhan, Meruyert Oshayeva, Alaa F. M. Al-Hadeethi. 2025 28th International Symposium on Wireless Personal Multimedia Communications (WPMC), 2025.

  • Algorithms & data structures

2023

Attrition Rate Measuring in Human Resource Analytics Using Machine Learning

Meraliyev Bakhtiyor, Karabayeva Aruzhan, Altynbekova Tolganay, Nematov Yelaman. Proceedings — 2023 17th International Conference on Electronics, Computer, and Computation (ICECCO), 2023.

  • People & HR analytics

Machine Learning as an Effective Tool for Human Resource Management in the Higher Educational Field

Meraliyev Bakhtiyor, Alibekova Balzhan, Bekturganova Islana. Proceedings — 2023 17th International Conference on Electronics, Computer, and Computation (ICECCO), 2023.

  • People & HR analytics

2021

A Computer Vision Supported Investigation on Anxiety-Success Relationship in Educational Processes

Bakhtiyor Meraliyev, Kurmangazy Kongratbayev, Marzhan Bekbalanova, Cemil Turan, et al.. 2021 IEEE International Conference on Smart Information Systems and Technologies (SIST), 2021.

  • Computer vision
  • Learning analytics

Content Analysis of Extracted Suicide Texts From Social Media Networks Using Natural Language Processing and Machine Learning Techniques

Bakhtiyor Meraliyev, Kurmangazy Kongratbayev, Nazerke Sultanova. 2021 IEEE International Conference on Smart Information Systems and Technologies (SIST), 2021.

  • NLP & text analysis

2020

Applying Machine Learning Models for Predicting Forest Fires in Australia and the Influence of Weather on Their Spread

Bakhtiyor Askarovich Meraliyev, Kurmangazy Sakenuly Kongratbayev. Proceedings of the 2020 International Young Scholars Workshop, 2020.

  • Environmental modelling

What we build

Projects

Dashboards, models and data tools built by lab members.

All code on GitHub ↗

Freedom Customer Review Analytics

A dashboard that analyses 2GIS reviews of companies in the Freedom ecosystem to understand customer experience and spot service issues across branches.

  • Sentiment analysis
  • Topic analysis
  • Branch comparison
  • AI insights

Kazakhstan IT Job Market Analytics

Vacancies from HeadHunter, Kaspi, Kolesa and Telegram combined in one dashboard to explore salaries and skill demand in Kazakhstan's IT job market.

  • Data collection
  • Salary analysis
  • Job requirements
  • LLM enrichment

Kindergarten Accessibility in Almaty

An interactive map that compares kindergarten locations with population distribution, finds underserved areas and suggests where new kindergartens are needed.

  • Geospatial analysis
  • Population coverage
  • District comparison
  • Site recommendations

Astana Real Estate Analytics

A machine learning app that estimates apartment prices and compares Krisha.kz listings with model predictions to flag overpriced and underpriced offers.

  • CatBoost
  • Price prediction
  • Price per m²
  • District comparison

Employer Attractiveness Analytics

A survey dashboard showing how respondents rate employers and which factors drive their choice of company.

  • Company rankings
  • Group comparison
  • Selection factors
  • Survey data

SDU Research Publication Analytics

A platform for exploring SDU's publication activity and research collaboration using Scopus and Web of Science data.

  • Publication trends
  • Journal quartiles
  • Co-authorship networks
  • Faculty comparison

SDU University Review Analytics

A dashboard that analyses SDU University reviews on 2GIS across 23 thematic categories to track student experience over time.

  • 23 categories
  • Sentiment trends
  • Monthly reviews
  • AI insights

Join the lab

Openings

Ways to take part in research at MADRID Lab.

Undergraduate research assistant

Bachelor's students

Work alongside lab members on an ongoing project: preparing data, running experiments and writing up results.

  • Comfortable with Python
  • Basic statistics and linear algebra
  • Able to commit a few hours every week
Apply

Open application

Anyone

Don't see a position that fits? Tell us what you would like to work on and how you could contribute.

Apply

How to apply

Send your CV and a few lines about what you would like to work on. Mention the position in the subject line.

Email the lab

From the lab

Blog

Notes from lab members on projects, papers and what we learn along the way.

Latest activity

News

News from our lab and the wider AI and machine learning community.

AI & ML worldwide

Selected developments from primary sources. Reviewed October 12, 2026.

  • Anthropic launches the Cyber Mission

    A long-term programme to help defenders secure critical infrastructure and open-source software. It starts with a Critical Infrastructure Defense Program for operational-technology security providers and OSS Scanner, which offers free periodic vulnerability scans of enrolled open-source projects.

    Source: Anthropic ↗
  • Anthropic releases Claude Haiku 5.5

    Anthropic's new small model targets high-volume work such as summarisation, sub-agent tasks and browser use. According to Anthropic it costs about 75% less to run than Haiku 4.5, and it is the first Haiku model with effort controls.

    Source: Anthropic ↗
  • OpenAI rolls out GPT-6 with Intelligent UI in ChatGPT

    GPT-6 is rolling out globally in ChatGPT. With Intelligent UI, answers can combine text with charts, forms, buttons and interactive tools, so the interface adapts to each question.

    Source: OpenAI ↗
  • Google releases EmbeddingGemma 2

    An open 740M-parameter embedding model that places text, images, audio and video in one shared vector space and is small enough to run on-device. It is released under Apache 2.0, which makes it a practical choice for multimodal search and retrieval research.

    Source: Google ↗
  • Mistral opens a preview of Mistral Large 4

    A natively multimodal mixture-of-experts model with about 1 trillion parameters, 52 billion of them active, trained from scratch in Mistral's European data centres. The API preview is available now, and Mistral plans to release open weights by the end of the month.

    Source: Mistral AI ↗
  • Claude identifies a previously uncharacterised enzyme system

    Anthropic's life sciences team reports that Claude autonomously found an enzyme system in bacteriophages that pairs a reverse transcriptase with CRISPR-like repeats. Its function is still under investigation, but the result is an example of AI contributing to scientific discovery.

    Source: Anthropic ↗
  • Anthropic reports on AI misuse and defensive safeguards

    Anthropic describes detected misuse of Claude, including cyber operations and influence campaigns, and the steps taken to disrupt it. The report highlights why monitoring and safeguards matter as AI capabilities grow.

    Source: Anthropic ↗
  • WeatherNext 3 brings hourly AI weather forecasts

    Google DeepMind and Google Research introduced a forecasting model that uses live satellite observations, with hourly updates and selected surface variables at 5-kilometer resolution. Applications include renewable energy and local weather prediction.

    Source: Google DeepMind / Google Research ↗
  • Google introduces Gemini 3.8 Flash and Flash Cyber

    Google announced new models for coding, reasoning and agent workflows. The specialized Flash Cyber variant focuses on vulnerability detection and patching, with access for trusted defenders through the Fairwind Program.

    Source: Google ↗
Show more (5)Show less

Lab news

Updates from our team.

  • Lab team advances at AI & Digital Bridge 2026

    Aisulu Igimbayeva, Aknur Mansurkhan and Abilmansur Alibekov defended their team's project in Astana and moved on to the second qualification stage of the GovTech programme.

  • Three new members join the lab

    We welcome Abilmansur Alibekov, Akbike Zhurymbay and Aruzhan Iliyas to MADRID Lab.

  • Our new website is live

    Members, publications, open positions and news are now all in one place.