Portrait of Hatem Ghorbel, a researcher who bridges the gap between data and industry
Originally from Tunisia, Hatem Ghorbel obtained his PhD from EPFL, where he became familiar with language theory in the Theoretical Computer Science Laboratory, specializing in the study of the logical and mathematical foundations of computer science. A professor at HE-Arc Neuchâtel since 2004, he is in charge of the "Data Analysis" group. He is a regular swimmer and greatly enjoys being close to nature and the lake.
Meet a researcher who knows how to make the link between data and the industrial world.
What are your group's core competencies at He-Arc?
We work on 3 types of interconnected activities based on applied artificial intelligence. These are data analysis, machine learning and optimization.
Can you explain data analysis?
When we talk about data analysis, on the one hand we have textual data, i.e. digital documents (text, PowerPoint, pdf, etc.) and on the other, digital data from the web or various technological devices (internet, social networks, etc.).
Our main goal is to discover and extract knowledge from these documents by applying classification techniques and linguistic models.
We are working on keyword extraction to try and filter the content and keep what interests us. When using a search engine, the challenge is to be able to give relevant computer results to a query. Understanding the intention and context of the text is still the most complicated part of "translating".
We do a lot of work in the medical field, analyzing thousands of scientific publications to identify, for example, plausible new associations between a drug molecule and a protein in the body. It's the equivalent of several scientists brainstorming together. Let me give you a very simple everyday example: paracetamol is used for pain, but pain is often associated with fever, so maybe paracetamol could be used against fever.
For the time being, this is a research project, which we have already published. The project was initiated by Novartis, but the collaboration was unsuccessful. We hope to work with a local company. Today, biomedical work can be carried out on the basis of textual knowledge, internal documents, technical reports and so on.
Today, it's not easy to make this model generic across different industries. Indeed, when you move from MedTech to construction, for example, the vocabulary associated with describing a problem is not the same. However, the latest Transfer Learning technologies offer interesting avenues of exploration through the BERT family of language models.
What is the application of Machine Learning and what is your Group's contribution to this activity?
Machine Learning also applies to the industrial sector. The aim is to analyze large volumes of data to provide decision-support elements for optimizing processes and anticipating unplanned faults. Today, thanks to the many sensors installed on the machine (pressure, temperature, current consumption, etc.), it is possible to correlate a signal and an event, provided there is an algorithm to make the link. The real effort of the machine is taken into account, and the part is changed only when necessary, thus saving on maintenance costs.
The algorithms we develop are based on analyses, mathematically of temporal data series, which are process measurements of a phenomenon through operating time. These algorithms enable us to find automatic solutions for quality control and the detection of abnormalities. Sometimes, this abnormal behavior is not directly perceived by a human being. We're able to carry out quality control faster and more efficiently, and we're experienced in the field of machine tools.
We have also been involved in a number of projects, such as detecting abnormalities in nuclear power plants, sewage treatment plants and incineration plants, etc.
How do companies contact you?
There is a strong cohesion between the industry and the academy, which has been further strengthened in recent years through the regional economic fabric.
The engineering school is known to companies who contact us directly or through the canton's economic department, which acts as a link between us.
Optimization is your group's latest core business. What does this mean in concrete terms?
It can be difficult to optimize a process. Let's take the example of a robot that picks up parts from a rolling tray and has to deposit them in another location. Imagine that there are conditions to pick up a part, because not all positions of this part are possible. So, after analysis, we know that we need to vibrate the tray to be able to put the part in a pickable position. So we need to find the right vibration characteristic for the part and the tray to maximize the number of pickable parts on the conveyor belt, so that the robot doesn't stop. The robot has a camera that looks for the part in the pickable position. If the image doesn't conform, it doesn't pick it up. Basically, it's a mechanical problem, and our aim is to find the right parameter to optimize the process. First of all, we need to understand how the process works in each area. For each project, we visit the site to understand the mechanism, the machine, the mechanical aspect, which enables us to take into account the necessary parameters.
Is data collection within the company a sensitive issue?
That's for sure. Data loss due to attacks remains a major issue. Data is on internal servers, because companies lack confidence in the Cloud.
Before collecting the data, we create scenarios with the companies, to ensure that the data collected will be of high quality and will answer the question being asked.
We're still at a stage where we're not mixing data between companies. Indeed, we need to know each company's process in order to interpret the data.
Are your partners cantonal companies?
We have many partners in the canton, especially in the machine tool sector. We work with Mikron, Tornos, Providence Hospital, NOMAD, Johnson & Johnson and others. We are part of the in-house Mill project (MicroLean Lab). This is a very large project involving over thirty companies. A microfactory capable of hosting apps, manufacturing, assembly and control technologies. This microfactory can be configured at will according to what needs to be produced. The path opened up by the Micro5 makes machining more autonomous, with less human intervention.
What are your current challenges?
The main challenge for my group is to continue developing Innosuisse projects in the field of AI. These are projects that give us opportunities to innovate. We're in a field where new algorithms are appearing every day, and to innovate, we need a concrete case study and serious partners. We have carried out a number of projects with partners in the region, which has given us the experience and maturity to be able to solve problems quickly and innovatively, especially in the fields of machine tools and MedTech. It's a real challenge to create value by making sense of data.
What do you think of our ecosystem?
We receive many invitations to collaborate from the canton, which is a very good thing.
I think we need to further strengthen the networking that already exists. I find that the first step is a difficult one for companies wanting to innovate, so why not consider helping them to carry out their first innovation project, which would boost our ecosystem even more.
Source : Microcity
