Twitter Sentimental Analysis using webMethods

Knowledge base
, , , , , ,
1

*products versions - webMethods 10.11 , REST API, azure cognitive service

Introduction

The idea is to be able to have an on premise package for customer who want to do social media sentimental analysis of the of individuals who interact with the organization.

Use case which can be helpful

  • Understand Employee thought process

  • Use this to do background check of existing or new employee / vendors or partners

Pre-requisite

  • Developer account on social media for example say twitter in this case

  • Access to some sort of report generation

  • Email server if emails needed to be triggered (optional)

Steps to follow

image
image1691×376 73 KB

p

image
image1137×675 29.2 KB

Reasons for Using webMethods vs Pure Azure solution

  1. You would need full access to data hub and data lake etc apart from a server to fetch data from social medias in case you want to do live streaming. The cons is social media developer accounts in free tire have limited no of innovations allowed .Plus using webMethods you could leverage existing hardware to do such social media analysis in case you have the same available.
  2. Why use Azure cogitative services to build sentimental analysis ? - This service already is a learning algorithm to identify the sentiment so why redesign the wheel use existing wheel plug and play into your main code and have faster time to market your overall solution with this capability.
  3. Why not use all azure services for integration ? - if you calculate the overall cost for each such transaction across different azure services hosting and building custom solutions on webMethods would be easier for example some custom reports there are webMethods package which you can use on existing server to generate reports which are customized .

Azure Sentimental Analysis Service

Sentiment Analysis applies sentiment labels to text, which are returned at a sentence and document level, with a confidence score for each.
The labels are positive, negative, and neutral. At the document level, the mixed sentiment label also can be returned. The sentiment of the document is determined below

Next steps

The implementation package will be shared separately link will be updated

Useful links | Relevant resources

Link for implementation https://tech.forums.softwareag.com/t/code-implementation-twitter-sentimental-analysis-using-webmethods/268495

@VijayBhaskar.VenkataBandaru_emp will be sharing the code package

2 Likes