Factos essenciais
- Datas
- 15–19 fev. 2027 (5 dias)
- Local
- Best Western Hotel Ikibin-2000, Ancara, Turquia
- Preço por participante
- £2.870 (GBP)
- Pagamento
- Por fatura e transferência bancária
- Encerramento das inscrições
- 6 jan. 2027
- Certificado
- Certificado de conclusão
- Opção interna
- Realize-o para a sua equipa, nas datas que preferir
Nesta página 15 secções
Visão geral do programa
Applied Data Science with Python Training in Ankara gives government agencies, donor-financed project units, public service providers and private-sector teams a practical way to strengthen Python, statistics, data cleaning, visualisation and machine learning. The five-day learning journey concentrates on tasks that professionals can demonstrate at work, rather than a collection of unrelated management topics.
Free self-check: the digital readiness assessment gives how ready your institution is for digital services, data and AI, in about 10 minutes.
Türkiye published its Artificial Intelligence Action Plan 2026–2030, including public-sector applications and a national data library objective. This makes practical data preparation and model governance relevant to teams responsible for turning data into useful decisions.
The World Bank GovTech Maturity Index 2025 assesses core government systems, online services, digital citizen engagement and enabling capabilities, including digital skills. These categories offer a structured backdrop for evaluating whether an analytics initiative is institutionally ready.
The World Bank documents an implemented municipal AI application-routing example in Türkiye, where historical citizen submissions were labelled and classified to improve administrative routing. The case illustrates the often underestimated importance of data labelling and operational supervision.
In a mixed professional group, participants compare the procedures they use, the documents that demonstrate compliance and the decisions for which they are accountable. Every technical case is assessed for practicality, evidence quality and risks that could arise during implementation. The programme is a skills course, not a promise of institutional accreditation or an externally arranged visit.
The Programme Strengthens Institutional Capability
- Python environment and reproducible notebooks
- Data lifecycle and documented analysis questions
- Structured data import and validation
- NumPy arrays and numerical operations
- Pandas data cleaning and transformations
- Exploratory analysis and data profiling
- Descriptive statistics and uncertainty
- Sampling design and selection bias
- Hypothesis testing and interpretation
- Regression and predictive modelling
- Classification and model evaluation
- Feature engineering and leakage prevention
- Matplotlib and Seaborn charts
- Basic natural language processing
- Ethical data access and privacy controls
- Communicating uncertainty to decision makers
The output of the week is an institution-specific working package, combining documented decisions, review checklists, a short executive brief and a realistic adoption plan. Participants are encouraged to bring non-sensitive examples from their own work while all teaching datasets and scenarios can be completed with fictional material.
The Growing Importance of Applied Data Science with Python
Technical proficiency in Python analytics for public institutions is increasingly connected with institutional accountability. Teams need to know not just how a tool or procedure operates but when it is appropriate, who authorises it, what evidence is retained and how problems are escalated. Reliable practice improves the quality of decisions while reducing avoidable rework.
Türkiye published its Artificial Intelligence Action Plan 2026–2030, including public-sector applications and a national data library objective. This makes practical data preparation and model governance relevant to teams responsible for turning data into useful decisions.
As organizações precisam cada vez mais de profissionais capazes de:
- Planning python environment and reproducible notebooks for a defined operational purpose
- Planning data lifecycle and documented analysis questions for a defined operational purpose
- Planning structured data import and validation for a defined operational purpose
- Planning numpy arrays and numerical operations for a defined operational purpose
- Planning pandas data cleaning and transformations for a defined operational purpose
- Planning exploratory analysis and data profiling for a defined operational purpose
- Planning descriptive statistics and uncertainty for a defined operational purpose
- Planning sampling design and selection bias for a defined operational purpose
- Planning hypothesis testing and interpretation for a defined operational purpose
- Planning regression and predictive modelling for a defined operational purpose
- Planning classification and model evaluation for a defined operational purpose
- Planning feature engineering and leakage prevention for a defined operational purpose
Strong Python analytics for public institutions systems help organisations:
- Make structured data import and validation a documented routine instead of an informal assumption
- Make numpy arrays and numerical operations a documented routine instead of an informal assumption
- Make pandas data cleaning and transformations a documented routine instead of an informal assumption
- Make exploratory analysis and data profiling a documented routine instead of an informal assumption
- Make descriptive statistics and uncertainty a documented routine instead of an informal assumption
- Make sampling design and selection bias a documented routine instead of an informal assumption
- Make hypothesis testing and interpretation a documented routine instead of an informal assumption
- Make regression and predictive modelling a documented routine instead of an informal assumption
- Make classification and model evaluation a documented routine instead of an informal assumption
An effective institutional approach links senior management ownership with staff capability, written procedures and proportionate monitoring. This programme uses Ankara as a professional learning venue while inviting participants to compare methods across agencies and countries.
1. What is the most defensible way to evaluate a classification model?
Out-of-sample validation and error review reduce the risk of overstating predictive performance.
2. Why build a data dictionary before cleaning a dataset?
Documenting field meaning and provenance is central to reproducible, accountable analytics.
3. What does the World Bank GovTech Maturity Index 2025 assess?
The World Bank GTMI 2025 covers four GovTech dimensions, including digital skills and institutions.
Who Should Attend This Applied Data Science with Python Training?
Este programa destina-se a:
- Data analysts
- Data scientists
- Software and data engineers
- Economists and researchers
- Government statisticians
- Ministry planning officers
- Public sector ICT managers
- Monitoring and evaluation specialists
- Utility performance analysts
- Internal audit data teams
- Public finance analysts
- Programme implementation officers
- Digital transformation managers
- Policy evaluation researchers
The course is most useful to people who have a real operational responsibility or are preparing to supervise one. Mixed groups enable a technical specialist to test recommendations with finance, compliance, programme and management colleagues before applying them to an institution.
Assinale o que a sua equipa precisa de ser capaz de fazer.
Applied Data Science with Python Training Objectives
Upon successful completion of this Applied Data Science with Python Training, participants will be able to:
- Explain python environment and reproducible notebooks and its practical boundaries
- Diagnose gaps in data lifecycle and documented analysis questions using appropriate evidence
- Apply structured data import and validation to a realistic work case
- Choose suitable approaches for numpy arrays and numerical operations
- Prepare defensible records for pandas data cleaning and transformations
- Evaluate operational risks associated with exploratory analysis and data profiling
- Design checks for descriptive statistics and uncertainty
- Coordinate staff responsible for sampling design and selection bias
- Interpret data and feedback relating to hypothesis testing and interpretation
- Communicate decisions about regression and predictive modelling to leadership
- Develop a phased institutional plan for classification and model evaluation
Each objective is practised through an assessment, demonstration, structured case or group decision. The emphasis is on repeatable methods that can be embedded in existing workflows and adapted to the participant’s legal, organisational and financing constraints.
Conteúdo e estrutura do programa
- Define the institutional question before selecting a dataset
- Set up Python environments and notebook conventions
- Use variables, functions, lists and dictionaries safely
- Import CSV, spreadsheet and database extracts
- Inspect records, types, identifiers and missing values
- Apply NumPy for efficient numerical calculation
- Build a Pandas DataFrame and inspect its structure
- Profile distributions, outliers and inconsistent codes
- Handle missing observations with defensible rules
- Merge datasets with controlled keys and duplicates
- Group and aggregate observations using Pandas
- Choose appropriate sampling and summary methods
- Explore uncertainty, confidence intervals and bias
- Test a practical programme performance hypothesis
- Translate a service problem into a modelling task
- Prepare training, validation and test partitions
- Avoid leakage from outcomes into input features
- Fit interpretable linear and logistic regression models
- Use scikit-learn pipelines and transformations
- Compare classification precision, recall and specificity
- Interpret confusion matrices and error distributions
- Clean free-text citizen and service records responsibly
- Extract simple features for text classification
- Use scikit-learn for baseline natural language models
- Understand when SciPy supports statistical analysis
- Assess the need for Spark or other scalable tools
- Review web-data collection permissions and quality
- Apply access restrictions and de-identification checks
- Connect analysis findings to institutional decisions
- Design charts with clear units and uncertainty notes
- Build a simple dashboard or notebook briefing
- Write a short model card and data provenance statement
- Compare proposed action against operational capacity
- Develop validation checks for future data updates
- Assign ownership of outputs and periodic recalibration
Conteúdo completo, dia a dia
The five-day sequence moves from understanding the problem and evidence base through operational design and assurance to an actionable institutional plan. The themes are:
- Day one: python foundations and analytic workflows
- Day two: cleaning data and interpreting statistical evidence
- Day three: prediction classification and responsible model evaluation
- Day four: text analytics scalable workflows and model safeguards
- Day five: operational dashboards and institutional application
Conteúdo completo, dia a dia
First Day | Python foundations and analytic workflows
- Define the institutional question before selecting a dataset
- Set up Python environments and notebook conventions
- Use variables, functions, lists and dictionaries safely
- Import CSV, spreadsheet and database extracts
- Inspect records, types, identifiers and missing values
- Apply NumPy for efficient numerical calculation
- Build a Pandas DataFrame and inspect its structure
- Document reproducible file and folder practices
- Develop a data dictionary and initial quality log
Workshop: Baseline assessment
Create a reproducible notebook from a fictional agency service dataset, including a data dictionary and issue log. The group records evidence used, unresolved assumptions, responsible officers and decisions requiring verification. Participants then compare approaches with colleagues, revise their work and identify a practical step for their own institution.
Second Day | Cleaning data and interpreting statistical evidence
- Profile distributions, outliers and inconsistent codes
- Handle missing observations with defensible rules
- Merge datasets with controlled keys and duplicates
- Group and aggregate observations using Pandas
- Choose appropriate sampling and summary methods
- Explore uncertainty, confidence intervals and bias
- Test a practical programme performance hypothesis
- Create clear charts using Matplotlib and Seaborn
- Prepare a defensible exploratory analysis brief
Workshop: Design choices
Investigate a fictitious district service dataset, compare two groups and identify whether the observed gap survives quality checks. The group records evidence used, unresolved assumptions, responsible officers and decisions requiring verification. Participants then compare approaches with colleagues, revise their work and identify a practical step for their own institution.
Third Day | Prediction classification and responsible model evaluation
- Translate a service problem into a modelling task
- Prepare training, validation and test partitions
- Avoid leakage from outcomes into input features
- Fit interpretable linear and logistic regression models
- Use scikit-learn pipelines and transformations
- Compare classification precision, recall and specificity
- Interpret confusion matrices and error distributions
- Document model limitations and subgroup fairness
- Communicate prediction results without overclaiming causality
Case Study: Risk and decisions
Build a baseline classification model for delayed service requests and explain false-positive and false-negative risks. The group records evidence used, unresolved assumptions, responsible officers and decisions requiring verification. Participants then compare approaches with colleagues, revise their work and identify a practical step for their own institution.
Fourth Day | Text analytics scalable workflows and model safeguards
- Clean free-text citizen and service records responsibly
- Extract simple features for text classification
- Use scikit-learn for baseline natural language models
- Understand when SciPy supports statistical analysis
- Assess the need for Spark or other scalable tools
- Review web-data collection permissions and quality
- Apply access restrictions and de-identification checks
- Create version control and reproducibility standards
- Plan a human review procedure for model outputs
Workshop: Operations and assurance
Classify anonymised complaint themes and document how human review, privacy and model drift should be managed. The group records evidence used, unresolved assumptions, responsible officers and decisions requiring verification. Participants then compare approaches with colleagues, revise their work and identify a practical step for their own institution.
Fifth Day | Operational dashboards and institutional application
- Connect analysis findings to institutional decisions
- Design charts with clear units and uncertainty notes
- Build a simple dashboard or notebook briefing
- Write a short model card and data provenance statement
- Compare proposed action against operational capacity
- Develop validation checks for future data updates
- Assign ownership of outputs and periodic recalibration
- Present a policy and service delivery case succinctly
- Produce a 30, 60 and 90-day analytics adoption roadmap
Capstone Exercise: Institutional action plan
Present an executive dashboard, code notebook and implementation plan specifying owners, safeguards and monitoring milestones. The group records evidence used, unresolved assumptions, responsible officers and decisions requiring verification. Participants then compare approaches with colleagues, revise their work and identify a practical step for their own institution.
- Recolher
- Limpar
- Armazenar
- Analisar
- Decidir
Applied Data Science with Python Training Methodology
Facilitators introduce concepts through short explanations and then ask participants to work through realistic decisions related to Python, statistics, data cleaning, visualisation and machine learning. The format uses controlled datasets, fictional records and carefully selected public references; participants are never required to share confidential customer information, protected case files or unpublished institutional records.
O programa combina:
- Instructor-led explanations of terminology and current guidance
- Daily applied workshops using topic-specific exercises
- Structured comparison of official policies and standards
- Small-group analysis of competing operational options
- Individual review of an institutional readiness baseline
- Short presentations of decisions and their evidence
- Peer scrutiny of assumptions and implementation risks
- Simulation of procedures and escalation decisions
- Template development and checking against requirements
- Daily reflective notes on changes to working practice
- Facilitator feedback on participant-produced documents
- Final presentation of a feasible institutional action plan
Before arrival, a short pre-assessment identifies job roles, expected outputs and practical barriers. The trainer uses it to adjust examples while retaining the published technical content. Exercises favour local adaptation over copying another country’s arrangements.
The final peer review tests feasibility, ownership, evidence needs and a realistic timetable for introducing the proposed changes.
Resultados esperados
Upon completion of this Applied Data Science with Python Training, participants will be able to:
- Produce a usable review or working document for python environment and reproducible notebooks
- Produce a usable review or working document for data lifecycle and documented analysis questions
- Produce a usable review or working document for structured data import and validation
- Produce a usable review or working document for numpy arrays and numerical operations
- Produce a usable review or working document for pandas data cleaning and transformations
- Produce a usable review or working document for exploratory analysis and data profiling
- Produce a usable review or working document for descriptive statistics and uncertainty
- Produce a usable review or working document for sampling design and selection bias
- Produce a usable review or working document for hypothesis testing and interpretation
Participants should be able to explain why a selected method is appropriate, what information is still missing and who must authorise its implementation. Their final products are designed for managerial discussion and further adaptation, not as substitutes for professional certification, legal advice or regulatory approval.
£2.870 por participante · fatura e transferência bancária · desconto para grupos de 5 ou mais pessoas
Cinco ou mais participantes da mesma organização: peça a tarifa de grupo no pedido de fatura.
Ten or more: an in-house version for your institution may cost less per person. Ask for an in-house quote.
Registration closes in 50 days (6 Jan 2027)
Institutional Application of Applied Data Science with Python Training
Este programa ajuda as organizações a reforçar:
- Governance and staff ownership of python environment and reproducible notebooks
- Governance and staff ownership of data lifecycle and documented analysis questions
- Governance and staff ownership of structured data import and validation
- Governance and staff ownership of numpy arrays and numerical operations
- Governance and staff ownership of pandas data cleaning and transformations
- Governance and staff ownership of exploratory analysis and data profiling
- Governance and staff ownership of descriptive statistics and uncertainty
- Governance and staff ownership of sampling design and selection bias
- Governance and staff ownership of hypothesis testing and interpretation
- Governance and staff ownership of regression and predictive modelling
- Governance and staff ownership of classification and model evaluation
- Governance and staff ownership of feature engineering and leakage prevention
Os participantes regressam com ferramentas práticas para melhorar:
- A practical checklist or decision record for numpy arrays and numerical operations
- A practical checklist or decision record for pandas data cleaning and transformations
- A practical checklist or decision record for exploratory analysis and data profiling
- A practical checklist or decision record for descriptive statistics and uncertainty
- A practical checklist or decision record for sampling design and selection bias
- A practical checklist or decision record for hypothesis testing and interpretation
- A practical checklist or decision record for regression and predictive modelling
- A practical checklist or decision record for classification and model evaluation
- A practical checklist or decision record for feature engineering and leakage prevention
Institutional adaptation begins with a named management sponsor and a narrow process where improvement can be observed. Participants identify baseline records, reporting responsibilities, resources and review dates, then recommend whether to pilot, revise or defer each intervention. A capacity assessment may help the organisation identify readiness gaps.
Relevância regional e aplicação
Europa e Ásia Central
Public agencies in Europe and Central Asia can examine administrative data quality, interoperability and statistical capacity while preserving local legal controls. Türkiye provides a relevant policy example through its 2026–2030 AI action plan.
Asia and the Pacific
Project implementation units and utilities across Asia can use reproducible analysis to test service delays, targeting and operational performance without assuming that predictive models replace professional judgement.
Africa and the Middle East
Ministries, municipalities and donor-financed projects can adapt the same datasets and review procedures to finance, public services and programme monitoring, provided country-specific privacy and access rules are followed.
The cases are intended as comparative learning, not universal legal advice. Each team must verify the applicable national rules, sector requirements and institutional capacity before adopting a template. For pre-course prioritisation, participants may use Risalat’s Training Needs Assessment.
Perfil do país anfitrião – Turquia
Ankara is the capital of Türkiye and a suitable convening city for professionals from ministries, universities and public institutions. The course uses internationally relevant datasets rather than implying a confirmed visit to a particular agency.
Türkiye’s AI Action Plan 2026–2030 provides a documented policy context for public data use and AI capability. The World Bank’s Turkish municipal routing example supplies a practical case for discussing how algorithms depend on well-maintained operational records.
Participants should confirm entry requirements against their nationality and residence, plan for travel to Ankara and verify current transport and accessibility arrangements. No external institutional meeting is included unless separately confirmed.
Applied Data Science with Python Training Deliverables
Risalat structures this programme around preparation, practical participation and follow-up. Topic-specific templates are designed for adaptation to the participant’s institutional context. Travel, accommodation and participation arrangements remain subject to the confirmed booking conditions issued by Risalat.
Antes do programa
- Avaliação prévia dos participantes
- Applied Data Science with Python Readiness Assessment
- Applied Data Science with Python Resource Pack
- Materiais de leitura preparatória
- Apoio e aconselhamento gratuitos em matéria de vistos
Durante o programa
- Recolha no aeroporto e apoio de boas-vindas
- Accommodation with breakfast: 6 nights, check-in Sunday 14 February 2027, check-out Saturday 20 February 2027 (one night before and one night after the programme)
- Jantar de boas-vindas e networking
- Kit de formação Risalat (saco, garrafa, caneca, t-shirt, boné, capa de passaporte, canetas, bloco de notas, íman e porta-chaves)
- Comprehensive Applied Data Science with Python Training Manual
- Practical diagnostic checklist for python environment and reproducible notebooks
- Process mapping and baseline analysis worksheet
- Risk and controls assessment matrix
- Performance indicator and evidence register
- Case study decision log and peer review form
- Institutional 30, 60 and 90-day action plan template
- Workshops práticos e simulações
- Discussões de grupo e aprendizagem entre pares
- Almoço e pausas para café diários
- Acesso total às instalações do local de formação
- Mini tablet com materiais de aprendizagem digitais
- Visita de campo (quando aplicável)
- Passeio recreativo pela cidade
Depois do programa
- Certificado de Conclusão
- Prémio Risalat de Reconhecimento da Participação no Programa
- Materiais de referência pós-formação
- Completed Applied Data Science with Python case and decision toolkit
- Institutional implementation roadmap
- Selected monitoring and quality review templates
- Apoio pós-formação e orientação de acompanhamento
- Relatório do programa
Organisations seeking dedicated group delivery, tailored case exercises or additional implementation support can request a tailored in-house proposal.
Common Challenges in Applied Data Science with Python, And How This Training Helps
How can an organisation set priorities when several problems demand attention?
Day one uses python environment and reproducible notebooks to identify what is known, what is assumed and which operational outcome matters most. The exercise assigns a review owner and produces a practical record of what should change.
How can an organisation find reliable evidence for a difficult decision?
Days one and two develop a documented baseline for structured data import and validation and test whether information is complete and fit for purpose. The exercise assigns a review owner and produces a practical record of what should change.
How can an organisation translate general standards into procedures staff can use?
Day two turns numpy arrays and numerical operations into a practical checklist, record and decision sequence. The exercise assigns a review owner and produces a practical record of what should change.
How can an organisation manage risks and mistakes without hiding them?
Days three and four create clear review gates for exploratory analysis and data profiling and assign accountability for correcting failures. The exercise assigns a review owner and produces a practical record of what should change.
How can an organisation coordinate different units without losing ownership?
The day-four exercises map roles, handoffs and escalation around sampling design and selection bias. The exercise assigns a review owner and produces a practical record of what should change.
How can an organisation measure improvement when resources and time are limited?
Day five chooses a small set of indicators linked to hypothesis testing and interpretation and a realistic review cycle. The exercise assigns a review owner and produces a practical record of what should change.
How can an organisation make training useful after participants return to work?
The final capstone turns classification and model evaluation into a 90-day sequence with named owners, management decisions and observable outputs. The exercise assigns a review owner and produces a practical record of what should change.
Perguntas frequentes
Who should attend Applied Data Science with Python Training in Ankara?
The course is intended for responsible professionals such as data analysts, data scientists, software and data engineers, and cross-functional colleagues whose decisions affect Python, statistics, data cleaning, visualisation and machine learning.
Is this a practical course or mainly lectures?
It combines technical explanations with five applied cases, review templates, group decisions and a final action plan linked to Python analytics for public institutions.
What should participants prepare before attending?
Complete the pre-assessment, identify one relevant process or institutional challenge, and bring only public or appropriately anonymised records. Confidential files are not required.
Can the methods be used in government and donor-financed projects?
Yes, the cases can be adapted to public agencies, implementation units and regulated organisations, provided users check applicable national law, donor requirements and organisational controls.
Is a particular institution or technical visit confirmed?
No external institution or visit is represented as confirmed. Any field or technical exposure component depends on prior host acceptance and appropriate safeguarding arrangements.
Who Will Deliver This Programme
Each date is delivered by a team assigned from Risalat's expert network for this subject. For this programme we assign:
How we choose them: Experience in the region of the participants or comparable settings; Command of the financier rules that apply (World Bank, ADB, AfDB, EU, UN) where relevant; Delivery in English; other languages on request for in-house groups; No conflict of interest with participants' institutions.
Names and short CVs of the assigned experts are shared with registered participants and sponsoring organisations before the programme, on request. Are you an expert in this subject? Join our expert network.
Training courses in AnkaraData Science & Excel Modelling training courses
Porquê Ancara para este programa
Ankara is the administrative capital of Türkiye, hosting central government ministries, national agencies, and foreign diplomatic missions. Delegations can observe public administration and sector-specific operations, including visits to the Grand National Assembly and key state institutions. The city is easily reached via international and domestic flights through its dedicated airport, as well as by high-speed rail.
The Applied Data Science with Python Training in Ankara programme in February 2027 brings together officials and professionals from across the Middle East region and beyond. Sessions combine short inputs, case studies and group work on participants' own files, and the city's institutions can be added as a study visit on request.
Getting there: Ankara Esenboğa Airport (ESB).
Verifique se precisa de visto com o Verificador de Vistos gratuito e peça uma carta-convite depois de se inscrever.
O mesmo tema noutras cidades
- OCDE (Organização para a Cooperação e Desenvolvimento Económico) OCDE
- Programa das Nações Unidas para o Desenvolvimento PNUD
- Open Contracting Partnership Open Contracting Partnership
For teams and institutions
Bring your team
Teams that learn together apply more at work. Tell us how many people and how you would like to take part: we reply with a group quote within two working days.
Mais cursos de formação na Turquia · Todos os programas na Europa · Calendário de formação completo
Plan this programme in a roadmap: why it matters, why to approve it, the dates to plan and the career it builds.



