#39 Paris Women in Machine Learning & Data Science: application to travel case, transformers in health, and formal modeling
To kick off the year, we started with a joint meetup with two awesome comunities: Duchess France and PyLadies Paris! We were hosted at Datacraft, in their offices inside Jussieu.
Duchess is a group aiming at building relationships between women in IT, but everyone is welcomed to their technical parties! You can find them on instagram 📷, and they have a blog post with great experience sharing (in french) 🔈
PyLadies Paris, just like us, is the french chapter of an international organization: PyLadies. They aim at supporting diversity in the python community. They have a slack and a twitter 🐦
The event started with Begoña Ascaso, Director of Machine Learning and Data Science at Egencia. The title of her talk is “Machine Learning in the Travel Industry: Tales and legends”. What a promise, we are going to travel in the wonderlands of #MachineLearning!
The usage of Machine Learning has started a revolution in the last few years with the shift from university rooms to industrial prototypes. The travel industry has highly benefited from Machine Learning as the applicable business cases bloom in every single interaction with the customers or suppliers.
To give you an example of the use case Begoña presented:
The second talk was led by Ndèye Maguette Mbaye, Research Engineer at Institut Curie. She presented her awesome work on “Transformers for multimodal electronical health records”.
Interest has increased in the use of prognosis factors as a cursor for breast cancer personalized treatment. For clinicians, early detection of those factors can be helpful for a good management of the disease and for the choice of an efficient treatment. Moreover, it exists a huge amount of meaningful information in pathological reports, biological measurements and clinical information in a patient journey that remain unexploited. In that context, Maguette proposes to develop and apply novel machine learning techniques to predict cancer outcome such as recurrence or survival from multi-modal breast cancer patient data (including medical notes in natural languages and the outcome of various lab analyses). For that, Maguette used a deep neural sequence transduction for electronic health records called BEHRT1. This model is inspired from one of the most powerful transformer-based architecture in Natural Language Processing: BERT2.
Déborah Boyenval, PhD Student at Université Côté d’Azur, presented the last talk about “Formal modeling of biological cyclic behavior with control points: the case of the cell cycle”.
The main limitation of biologists rooted in an experimental practice is the ability to perform rigorous proofs in the absence of a language for formalizing the biological knowledge extracted from their experiments.
Biologists have identified numerous biochemical and genetic mechanisms involved in physiological functions or diseases, but once this knowledge is linked together it remains extremely difficult and expensive to predict the impact of genetic mechanisms on physiological functions.
Déborah presented to us her thesis, which focuses firstly on a reasonable mathematical specification of complex biological functions such as cell cycle checkpoints, which represent the main barrier against cancer. Secondly, she focused on the development of an automated proof method, using the mentioned tools, proving whether a set of genetic regulations is sufficient to generate cell cycle checkpoints.
You can access the code she explains in her presentation on github:
For this edition, PyLadies handled a technical setup to have a hybrid event, both on-site and remote! We were live on youtube, and there is the video:
That’s not the end, we are getting ready for new adventures!
Pyladies organize their next meetup on October 18th, at Talent.io and virtually. They will also have a mentoring session at Hugging Dace with Scikit-learn core developers: news to come on their meetup page!
Duchess just celebrated their 12th birthday on October 4th!
On our side, we will have our 40th meetup very soon #teasing! We will also have a new BBL, #teasingfurther.
If you want to keep posted of our news or talk to us, you can:
🔗 follow us on Twitter, Meetup, and LinkedIn
📑 check our Google spreadsheet if you want to speak 📣, host 💙, help 🌠
📍join our Slack channel for information, discussions, and opportunities
📩 send an email to the Paris WiMLDS team to paris@wimlds.org
🎬 subscribe to our WiMLDS Paris Youtube channel
📸 follow the global WiMLDS on Instagram, LinkedIn, and Facebook
🔥 share your company or lab’s job offers for free on the global WiMLDS’ website.
