Paris NLP Meetup #1 Season 2

Seating is on a first come, first served basis whether you have RSVPed or not, so we suggest arriving early. We can host 80 people.

La salle permet d'accueillir 80 personnes. L'inscription est obligatoire mais ne garantit pas que vous pourrez entrer, nous vous recommandons donc d'arriver un peu en avance.

----------

Gilles Moyse - Récital.ai

"From text classification to text summarization"

Alice Coucke, Snips

"Benchmarking NLU providers: how and why?"

There are currently many solutions for natural language understanding on the market, some developed by the main actors in the field (Google, Apple, Amazon, Microsoft, etc). When you’re building your own slot filling solution, as we did at Snips, you may want to compare yourself to other providers. Not only as an end in itself, but more as a way to evaluate the complexity of the task and to visualize the road still ahead. A benchmark may also give you interesting insights about the relative complexity of given intents (is asking for a movie recommendation intrinsically more complex that setting an alarm clock? Do I need more data?).That is why we have benchmarked several natural language engines, and our results show there are important discrepancies between different solutions. They also show that no one has fully cracked the problem of understanding natural language.In this talk, I will share the motivations and methodology behind the benchmark, give a bunch of advice from my own experience and share the raw data we have been using.

Dr. Gregory Grefenstette, Chief Scientist Officer @Biggerpan,

Senior Research Associate at the Institute for Human and Machine Cognition :

"Predictive Artificial Intelligence : The Future of Human to Machine Interactions"

Touchscreen interfaces are everywhere: from smartphones, to cars, and connected refrigerators. Although we may think the world is getting smarter, we have lost the convenience of the keyboard and mouse, making it difficult to tell a smarter device what we want to do: basic tasks like search, selecting, and copy-pasting are tedious in this newly connected world. Artificial intelligence is solving this problem, and filling the evolutionary gap of human-machine interaction, using voice-based assistants, and predictive interfaces, capable of understanding what people want in real time. By leveraging Natural Language Processing (NLP), the field of AI that extracts meaning from human language, devices can become more aware of what we are doing. They can analyze the things we read, the conversations we have with other people, the visual content we interact with. NLP can provide the meaning needed to make sense of all the things we do on our computers and connected devices.

In this talk, we will examine the latest progress in NLP (context classification, entity recognition, automated knowledge construction) that can provide us with more advanced applications able to analyze the context and the content of our activity. We will also explain how this leads to a predictive AI which can anticipate and serve people’s needs in real time, and why this is the future of human to machine interactions.

Biggerpan is a French-American startup which develops a predictive artificial intelligence that leverages context to make real-time recommendations. The company's mission statement is to build a brain for the mobile web, to allow a better integration of the technology into our lives, without all the pain and frictions that are found in traditional mobile online activities.

Liste complète des intervenants à venir. Si vous êtes intéressé, contactez-nous sur Twitter @ParisNLP!

Speakers will be announced soon. If want to have a talk contact us on Twitter @ParisNLP!