Publication date : 10/02/2025
Course : Machine learning, the state of the art
Seminar - 2d
- 14h00 - Ref. MLE
|
Download in PDF format
Share this course by email
 | Positioning machine learning in the data processing chain |
| Distinguish between the skills needed and the profiles to be recruited |
| Identify the keys to a successful machine learning project |
| Understand machine learning concepts and the evolution from Big Data to machine learning |
| Understand the challenges of using machine learning, including expected benefits and usage examples |
Intended audience
Corporate executives (CEO, COO, CFO, SG, HRD, etc.), CIOs, CDOs, IT managers, consultants, big data project managers.
Prerequisites
General computer literacy, and notions of probability and statistics are recommended.
Practical details
Teaching methods
Illustrated by case studies. Presentation of the main use cases by sector (automotive, industry, consumer goods, finance, health, energy, agriculture, transport, telecommunications, etc.).
Course schedule
1 The history of machine learning and the big data context
- Put the concepts of artificial intelligence and machine learning back into perspective...
- The link with mathematics, (inferential) statistics, data mining and data science.
- Move from descriptive analysis to predictive and prescriptive analysis.
- Machine Learning applications (search engines, spam detection, check reading).
- Dominique Cardon's typology of algorithms.
- The data science community and Kaggle challenges (e.g. Netflix).
Case study
Studies of concrete machine learning applications (search engines, spam detection, check reading).
2 Available data: collection and preparation
- Structured, semi-structured and unstructured data.
- Statistical nature of data (qualitative or quantitative).
- Connected objects (IoT) and streaming.
- Opportunities and limits of open data.
- Identifying correlations, the problem of multicollinearity.
- Dimension reduction by Principal Component Analysis.
- Outlier detection and correction.
- ETL (Extract Transform Load).
- Web scraping.
Demonstration
ETL (Extract Transform Load) demonstration. Web data collection.
3 Market tools for data processing and machine learning
- Traditional software (SAS, SPSS, Stata...) and its openness to open source.
- Choose between the two open source leaders: Python and R.
- Cloud platforms (Azure, AWS, Google Cloud Platform) and SaaS solutions (IBM Watson, Dataïku).
- New corporate jobs: data engineer, data scientist, data analyst, etc.
- Match the right skills to these different tools.
- Online APIs (IBM Watson, Microsoft Cortana Intelligence...).
- Chatbots (conversational agents).
Demonstration
Demonstration of a chatbot (conversational agent) and Azure Machine Learning.
4 The different types of machine learning
- Supervised learning: repeating an example.
- Unsupervised learning: discovering the data.
- Online (machine) learning as opposed to batch techniques.
- Reinforcement learning: reward optimization.
- Other types of learning (transfer, sequential, active...).
- Illustrations (recommendation engines...).
Demonstration
Demonstrations of the different types of machine learning possible.
5 Machine learning algorithms
- Simple and multiple linear regression. Limitations of linear approaches.
- Polynomial regression (LASSO). Time series.
- Logistic regression and scoring applications.
- Hierarchical and non-hierarchical classification (KMeans).
- Classification using decision trees or the Naïve Bayes approach.
- Ramdom Forest (decision tree development).
- Gradiant Boosting. Neural networks. Vector support machine.
- Deep learning: examples and reasons for current success.
- Text mining: analysis of textual data corpora.
Demonstration
Demonstrate the various basic algorithms in R or Python.
6 Algorithm training and evaluation procedure
- Data set separation: training, testing and validation.
- Bootstrap (bagging) techniques.
- Example of cross-validation.
- Definition of a performance metric.
- Stochastic gradient descent (metric minimization).
- ROC and lift curves to evaluate and compare algorithms.
- Confusion matrix: false positives and false negatives.
Demonstration
Demonstrating the choice of the best algorithm.
7 Production launch of a machine learning algorithm
- Description of a big data platform.
- How PLCs work.
- From development to production.
- Corrective and evolutionary maintenance strategy.
- Evaluation of production operating costs.
Demonstration
Demonstration of geolocation and sentiment analysis APIs.
8 Ethical and legal aspects of artificial intelligence
- CNIL missions and future developments.
- Right of access to personal data.
- The question of the intellectual property of algorithms.
- New corporate roles: chief data officer and data protection officer.
- The question of the impartiality of algorithms.
- Beware of confirmation bias.
- Sectors and professions affected by automation.
Group discussion
Brainstorming to identify the keys to success.
Customer reviews
4,4 / 5
Customer reviews are based on end-of-course evaluations. The score is calculated from all evaluations within the past year. Only reviews with a textual comment are displayed.
SEBASTIEN A.
10/03/26
4 / 5
RAS
SIMON A.
27/11/25
5 / 5
The first half-day dedicated to discussion seemed a little long to me. I would have appreciated getting to the heart of the matter a little earlier.
CAMILLE F.
27/11/25
5 / 5
content in line with requirementsgood teaching methodsvery interesting exchanges between participants and trainer
PARTICIPANTS
Corporate executives (CEO, COO, CFO, SG, HRD, etc.), CIOs, CDOs, IT managers, consultants, big data project managers.
PREREQUISITES
General computer literacy, and notions of probability and statistics are recommended.
TRAINER QUALIFICATIONS
The experts leading the training are specialists in the covered subjects. They have been approved by our instructional teams for both their professional knowledge and their teaching ability, for each course they teach. They have at least five to ten years of experience in their field and hold (or have held) decision-making positions in companies.
ASSESSMENT TERMS
The trainer evaluates each participant’s academic progress throughout the training using multiple choice, scenarios, hands-on work and more.
Participants also complete a placement test before and after the course to measure the skills they’ve developed.
TERMS AND DEADLINES
Registration must be completed 24 hours before the start of the training.
ACCESSIBILITY FOR PEOPLE WITH DISABILITIES
Do you need special accessibility accommodations? Contact Mrs. Fosse, Disability Manager, at psh-accueil@orsys.fr to review your request and its feasibility.
Corporate executives (CEO, COO, CFO, SG, HRD, etc.), CIOs, CDOs, IT managers, consultants, big data project managers.
PREREQUISITES
General computer literacy, and notions of probability and statistics are recommended.
TRAINER QUALIFICATIONS
The experts leading the training are specialists in the covered subjects. They have been approved by our instructional teams for both their professional knowledge and their teaching ability, for each course they teach. They have at least five to ten years of experience in their field and hold (or have held) decision-making positions in companies.
ASSESSMENT TERMS
The trainer evaluates each participant’s academic progress throughout the training using multiple choice, scenarios, hands-on work and more.
Participants also complete a placement test before and after the course to measure the skills they’ve developed.
TEACHING AIDS AND TECHNICAL RESOURCES
• The main teaching aids and instructional methods used in the training are audiovisual aids, documentation and course material, hands-on application exercises and corrected exercises for practical training courses, case studies and coverage of real cases for training seminars.
• At the end of each course or seminar, ORSYS provides participants with a course evaluation questionnaire that is analysed by our instructional teams.
• A check-in sheet for each half-day of attendance is provided at the end of the training, along with a course completion certificate if the trainee attended the entire session.
• The main teaching aids and instructional methods used in the training are audiovisual aids, documentation and course material, hands-on application exercises and corrected exercises for practical training courses, case studies and coverage of real cases for training seminars.
• At the end of each course or seminar, ORSYS provides participants with a course evaluation questionnaire that is analysed by our instructional teams.
• A check-in sheet for each half-day of attendance is provided at the end of the training, along with a course completion certificate if the trainee attended the entire session.
TERMS AND DEADLINES
Registration must be completed 24 hours before the start of the training.
ACCESSIBILITY FOR PEOPLE WITH DISABILITIES
Do you need special accessibility accommodations? Contact Mrs. Fosse, Disability Manager, at psh-accueil@orsys.fr to review your request and its feasibility.
Dates and locations
Select your location or opt for the remote class then choose your date.
Remote class
Last places available
Guaranteed date, in person or remotely
Guaranteed session