DataHour: Intelligent Knowledge Mining with Azure NLP and Graph databases

Online 29-09-2022 07:30 PM to 29-09-2022 08:30 PM
  • 11500


  • Knowledge and Learning


DataHour Recording

About the DataHour:

Over the years, a collection of more than 5 million reports received were searched and filtered  by the Portfolio managers using key words and key phrases for performing analysis. But now, with development of technology, the  Portfolio managers have started using a document search Engine with Azure Cognitive search for this task which enables them to search earnings call transcripts regarding companies being covered by them. With the advent of BERT, it is now possible to extend the simple keyword search to a much complex semantic search where user can search documents based on similar words and phrases like searching for "electric cars" might also highlight related terms like "Tesla" or searching for "Lithium" might prompt reports related to "ionic batteries" etc. which is more efficient and usable for the end users. 

In this DataHour Priyanka will discuss how to create a custom AI skill set in Azure cognitive search and add AI enrichment to the index and how to train the Sentence BERT on the text corpus and generate embeddings which can be further used to compute cosine similarity between words/ terms. She will introduce us with Azure Cognitive search features of synonym maps, Document extraction skill, dynamic document summarization features and the wonder of 2 powerful technologies: Azure Cognitive search and SBERT working in tandem to deliver a powerful semantic search engine.

Additionally, She will also cover the usage of Azure Cognitive search Semantic Q and A functionality and how users can search documents containing images using image tag for instance when the  users are searching for documents containing an image having a "compressor bulge" or "collapsed bridge". 

Prerequisites: Enthusiasm of learning Data Science.

Who is this DataHour for?

  • Students & Freshers who want to build a career in the Data-tech domain.
  • Working professionals who want to transition to the Data-tech domain.
  • Data science professionals who want to accelerate their career growth


Priyanka Shah

Group Manager at Avanade

Priyanka is currently working as a Group Manager at Avanade.With an experience of 10 + years in analysis, design and development of client/server, web based and n-tier application in both Java and technologies as well as UI Frameworks like Angular, Mobile development with NativeScript etc. she also has Expertise in Data science and is active blogger on AI/ML topics. 

She is also an influential Speaker at Microsoft and other technical events for Microsoft Technology stack and ML.NET, Conversation AI, Bot Framework, Azure Cognitive services like speech to text, Face recognition, Custom vision. 

She is Currently leading innovation technology team with projects architected with ElasticSearch, Bot framework, text analytics, spam detection, machine learning algorithms related to neural word embedding, topic modeling, LSTM for text completion, auto caption generation. Solution architect for micro services deployed with Azure Kubernetes cluster, Signal R, Azure Cognitive services and more. .

She is also contributing for the nation by Working with Government and retail clients for implementing AI for Sustainability, AI for accessibility. Mentor and trainer for AI Hackathons, AI training.

Connect with Priyanka at:


Please register/login to participate in the contest

Please register to participate in the contest

Please register to participate in the contest



We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy.


We believe in making Analytics Vidhya the best experience possible for Data Science enthusiasts. Help us by providing valuable Feedback.