This is a specialised course where 50% of the seats are reserved to PhD students enrolled at the Faculty of SCIENCE at UCPH and 50% of the seats are reserved to other applicants. Seats will be allocated on a first-come, first-served basis and according to the applicable rules.

Anyone can apply for the course, but if you are not a PhD student at a Danish university (except CBS), you will be placed on the waiting list until enrollment deadline. After the enrollment deadline, available seats will be allocated to applicants on the waiting list.

Aim and Content
The ISC-2027 is a four-week school designed to introduce different key aspects of Data Science and Machine Learning in different branches of science (chemistry, food & feed, physics, environmental, political economics, etc).
It is addressed to BSc, MSc, PhD students/post-docs, professors, as well as industrial and private researchers.

IMPORTANT: The ISC-2027 is structured in FOUR different and independent modules:
PROGAMMING, BASICS, INTERMEDIATE, CHALLENGES.

The students CAN CHOOSE WHICH ONES TO DO.
Please make sure to register individually for each course you intend to participate in.

WEEK 04 - CHALLENGES MODULE (2.5 ECTS)
This seminar contains several general topics:
- HYPER – Hyperspectral Data Analysis: Hyperspectral imaging is an important analytical tool in a growing number of areas, including the chemical sciences, process monitoring and forensics, cultural heritage, and remote and standoff sensing. Images are acquired using a wide variety of spectroscopic techniques, including, but not limited to, infrared, mass spectroscopy, and Raman, among others, and it can be confusing how to efficiently extract information from the sizable, multiwavelength images. This course will cover a variety of multivariate methodologies that can be applied to the analysis and interpretation of hyperspectral data, starting with principal components analysis (PCA) and using linked plots. Other techniques that will be discussed will include maximum autocorrelation factors, maximum difference factors, multivariate curve resolution (endmember extraction), target detection and targeted anomaly detection. This course will utilize the MATLAB environment, along with PLS_Toolbox and MIA_Toolbox, as well as the standalone software Solo and MIA_Toolbox.

The course content will be useful for those involved in chemical, food, pharmaceutical, and medical imaging, as well as remote and standoff imaging.
- MCR - Multivariate Curve Resolution: The module will address the theoretical description and
hands-on application of Multivariate Curve Resolution (MCR).
MCR is a multivariate resolution (unmixing) method that can provide the description of a multicomponent data set through a bilinear model of chemically meaningful profiles, e.g., when analyzing an HPLC-DAD data set, MCR would provide the real elution profiles and the related UV spectra for each compound in the sample.
It has applications in diverse fields, such as process analysis, chromatographic data, hyperspectral images or environmental data, in any context where a mixture analysis problem can be encountered.
MCR can be applied to a single data matrix or to multiset structures formed by blocks of different information (data fusion). The module focuses mainly on the algorithm MCRALS (Multivariate Curve Resolution-Alternating Least Squares), and hands-on work will be done using a dedicated free GUI interface adapted to MATLAB environment.
Applications will cover many of the areas mentioned above.

- NonLin – Deep Learning: This seminar aims at providing a basic introduction to the techniques
which may be used in all those situations when a linear relation is not enough to provide accurate
results (e.g. due to the presence of multiple sources of variability).
In this respect, the most important aspects of data modeling will be considered (classification and calibration). Topics such as different artificial neural network architectures (shallow learning and deep learning) will be covered.

- MULTIWAY - Multi-way analysis: Multi-way data is gaining popularity due to the capability of
scientific devices to generate data with at least 3 dimensions (elution time – mz channel –
samples, excitation-emission – sample, etc). Therefore, learning the basics of multi-way analysis
will help to extract the most from that complex data structure. In this sense, methods such as
parallel factor analysis (PARAFAC) and PARAFAC2 will be studied and applied to various
examples.

- GLUE – 1000M - Glue and 1000 methods in one day!
This seminar is the final seminar of the School, and it is divided into 2 parts:
- Glue: We will take a very close look at all the most common mistakes that even experienced
people will make when doing multivariate analysis.
We will cover exploration, calibration, interpretation, visualization and many other subjects. And always with a focus on the most common problem, as well as a sounder alternative. Additionally, a comprehensive overview of all the models studied at the school will be provided.

- 1000M: All the methods you can imagine in the same bag. This seminar will also address various
other methods/models that were not studied in the School but that the student may find extremely
helpful.

Important: The whole week is offered as a whole due to operational constraints. The modules
are independent, and some of them are simultaneous. Therefore, the student must choose which
modules to attend. All the material of the modules will be freely available.
Previous knowledge needed: Basic Chemometrics and a bit of the intermediate.

Software needed:
Feel free to work with Matlab, Python or R, or any other software that you consider (e.g. Unscrambler).

The teachers will work with:
- Matlab
- PLS_Toolbox / SOLO / MIA. IMPORTANT: For the PLS_Toolbox / SOLO / MIA, a fully functional
demo will be available for the School.
- MCR-ALS toolbox: MCR-ALS toolbox can be freely downloaded here:
https://mcrals.wordpress.com/download/mcr-als-2-0-toolbox/

Learning outcomes
Intended learning outcome for the students who complete the course:
Knowledge:
• Learn the basics of data analysis methods.
• Learn to handle data and create proper datasets and libraries for further analysis
• Learn critical thinking regarding Machine Learning, Chemometrics and IA

Skills:
• Develop their own data analysis protocols
• Code basic algorithms and the resources available for data analysis
• Apply the acquired knowledge to any problem related to their own research

Competences:
• Understand the structure of a vast number of data types and the issues derived from the data
• Independent thinking for the solution of their problems
• Interaction with other peers and teachers

Target Group
The course is specifically addressed to PhD students.
Additionally, the course attracts a high number of BSc, MSc, postdoctoral researchers, and professors. Another relevant audience is Industry; the course receives students from 2 to 3 companies every year.

Recommended Academic Qualifications
None specifically required. We start with basic topics and go all the way to more advanced topics.

Research Area
Chemometrics, machine learning, spectroscopy, artificial intelligence, programming, statistics

Teaching and Learning Methods
The seminars of the International School of Chemometrics will comprise a mix of presentations
from world-leading researchers, combined with practical and theoretical exercises in data analytics software, which will provide students with hands-on experience in applying the tools taught. The exercises are done under the supervision of the teachers.
The initial week of programming offers instruction in three different languages (MATLAB, R and Python), and all the instruction in this part is based on e-learning. The remaining three weeks of
the School are dedicated to physical on-site training.

Type of Assessment

The course is completed by attending and development during the practical exercises will be
evaluated by interest of the student.

Literature
Peer-reviewed papers provided during the course.

Course coordinator
Main coordinator: Rasmus Bro, Professor (rb@food.ku.dk)

Co-coordinator:
Beatriz Quintanilla Casas, Assistant profesor (beatriz@food.ku.dk)

Guest Lecturers
- Prof. José Amigo Rubio, University of the Basque Country.
- Prof. Morten A. Rasmussen, University of Copenhagen.
- Assoc. Prof. Asmund Rinnan, University of Copenhagen.
- Assoc. Prof. Agnieszka Smolinska, Maastricht University.
- Prof. Davide Ballabio, University of Milano-Bicocca.
- Prof. Anna de Juan, University of Barcelona.
- Dr. Neal Galhaguer, Eigenvector Research.
- Dr. Carlos de Cos, The Mathworks.
- Assoc. Prof. Sergey Kucheryavskiy, University of Aalbrog.
- Dr. Anders Krogh Mortensen, The AI Lab.
- Prof. Federico Marini, University of Rome La Sapienza.

Dates
05-09/04/2027: Week 01 - Programming
12-16/04/2027: Week 02 - Basic
19-23/04/2027: Week 03 - Intermediate
26-30/04/2027: Week 04 - Challenges

Expected frequency
Once per year

Course location
Frederiksberg Campus

Registration
Deadline for registration: 18/01/2027

Requirements for signing up
No extra requirement

Seats to PhD students from other Danish universities will be allocated on a first-come, firstserved basis and according to the applicable rules.
Applications from other participants will be considered after the deadline for registration.