Problem

In the EIS categorization project, there was a need to efficiently categorize and analyze a large volume of Employee survey comments. The existing manual process was time-consuming and hindered productivity. The goal was to leverage machine learning algorithms to automate the categorization process and improve overall customer experience.

The manual process of categorizing employee survey comments was taking up too much time and hindering productivity. It was necessary to find a more efficient way to categorize these comments so that business partners could save time, enhance customer experience, and raise awareness of our capabilities and value proposition.

Key Approach

Identified suitable machine learning algorithms for language identification, topic extraction, and sentiment analysis.
Utilized AI insights capabilities in Power BI to gain deeper insights from your categorized comments.
Collaborated with management and IT teams to gather the necessary data and train the machine learning models.
Collaborated with business partners to create categories and sub-categories as per their requirements to gather feedback and incorporate their insights into the categorization process.
Analyzed data to identify trends in employee survey comments that could be used to improve overall customer experience.
Implemented row-level security in Power BI to restrict access to categorized comments based on user roles or permissions

Solution

To achieve these objectives, we identified suitable machine learning algorithms for language identification, topic extraction, and sentiment analysis. I collaborated with management and IT teams to gather the necessary data and train the machine learning models. We leveraged Power BI's capabilities to integrate and deploy the machine learning models.

We developed a mechanism for categorizing customer comments using PowerBi Topic Extraction Azure ML model and assessed sentiment using sentiment analysis techniques. We conducted iterative testing and fine-tuning of the models to ensure accurate categorization and sentiment analysis results. We also collaborated with business partners, worked on creating categories and sub-categories as per their requirements to gather feedback and incorporate their insights into the categorization process.

Impact

Improved customer experience by accurately identifying and extracting IT-related comments using the Power BI Topic Extraction model.
Increased productivity by implementing a document grouping model to efficiently categorize customer comments.

"The implementation of machine learning algorithms and the automation of the categorization process reduced manual work by 50% by automating the identification of IT Related comments and categorization."

Automating Employee Survey Analysis with Machine Learning Algorithms