Senior Data scientist at GE Healthcare, working on Natural Language Processing for clinical projects. With an MSc. in Industrial Engineering, and a healthy sense of curiosity and exploration, I research, develop and implement AI products that drive business results. I am also a program manager at Baot, Israel’s largest community of women in R&D, where I help fellow data scientists find their next dream job.
Shani Gershtein
Senior Data scientist
English, Hebrew
Languages:
Location:
Hadera, Israel
Can also give an online talk/webinar
Paid only. Contact speaker for pricing!
MYTALKS
Boost Your Chat-Bot Using Unsupervised Machine Translation
Data / AI / ML
Let’s say you have a well speaking English chat-bot. You already have a successful solution for English, and you start thinking about your next users: those coming from Spain, Japan, perhaps Israel. You’ll need to translate utterances from each language to English, run your current English flow, and - finally - translate your bot’s English response to the user’s native language. Supervised machine translation is the obvious starting point, but getting paired corpora is very expensive, and even more so in a specific field, such as Healthcare or Finance. In this talk, I’ll go over the steps of creating an unsupervised machine translation solution model, discuss the various challenges and opportunities, and review the latest industry solutions for understanding and generating multilingual text in specific domains, thus enhancing your bot’s language capabilities. I will also discuss the evaluation metrics and quality estimation methods you'll need to master it in production.
Chat-Bots Response Generation: From a Rule-Based Solution to a Full Learning Model.
Data / AI / ML
When building a chat-bot, rule-based response generation is the go-to starting point in many use cases. This is indeed a good place to start, but we can do better. In this talk, we will survey methods for forging ahead your chat bot from a rule-based model to deep neural architectures, and the recommended steps in the middle. We'll discuss the challenges and opportunities in the space of natural language generation, how to generate text in specific domains, and how to formulate text from structured and unstructured data. We’ll also discuss evaluation metrics and quality estimation methods, to make sure you could also use your deep learning models in production.
Boost Your Chat-Bot Using Unsupervised Machine Translation
Completed
true
Visible
true
Order
1
Go to lecture page
Chat-Bots Response Generation: From a Rule-Based Solution to a Full Learning Model.