Building Sentence Similarity Applications at Scale
Building Sentence Similarity Applications at Scale

Abstract: 

Comparing the similarity of two sentences is an integral part of many Natural Language Processing scenarios. These scenarios range from search and retrieval, nearest-neighbor to kernel-based classification methods, recommendation, and ranking tasks. Building state of the art models at production level scale can be difficult when you’re on a small team and not both an NLP and DevOps expert. In this workshop, we will walk through the Natural Language Processing Best Practices Github Repo (https://github.com/microsoft/nlp ) provided by Microsoft on how to create baseline representation models for Sentence Similarity scenarios from popular open source technologies like gensim and scikit-learn. We will then use Microsoft's Automated Machine Learning to create a competitive model with popular sentence encoders from Google and create reusable machine learning pipelines deployed at scale on Azure Kubernetes Services.

Bio: 

Santhosh Pillai is a principal program manager with the Azure machine learning team at Microsoft. Santhosh is responsible for data scientists’ experimentation experience with Azure machine learning service, specifically its highly optimized ML workflow orchestration engine, AzureML Pipelines, that can stitch together multistep workflows across heterogeneous computes. He’s been working on the machine learning platform (infrastructure, SDK, and graph authoring UX) for Microsoft and its customers over the last several years.