Machine Learning Implementation with SharePoint Online

Business Scenario

The client is a technology-driven company that delivers tailored solutions with core focus areas of Microsoft Office 365, Microsoft Dynamics 365, Development, print & document management, and associated service & support of these areas. The client is well-known for delivering innovative solutions that serve customers spread across different regions.

The Challenges

Their prime challenge was to find the ML keywords from the existing documents and forecast the value for future with a defined analysis. It was needed to develop an algorithm which can increase the accuracy from document segregation process.

Solution Highlights

  • Extraction of keywords for existing documents
  • Keywords will be extracted from each document using ‘Text Analytics API’
  • Machine learning keywords will be created in the document library to store the keywords
  • Scheduler that traverses through each document, extract keywords and store them as ML Keywords
  • Create a machine learning experiment
  • Learning model created using Microsoft machine learning studio
  • Generate output prediction for required category values
  • Deployment of a model as API to update category values


  • Self-learning algorithm, periodic and automatic improvement in result accuracy
  • Increased accuracy in document segregation
  • Almost elimination of manual efforts