Session 1: Improving Customer Experience Through Natural Language Processing
Presentation Type
Invited
Student
No
Abstract
Unstructured data such as spoken words can be transcribed into text, transformed into structured data, then analyzed using Natural Language Processing (NLP) methods. Putting meaning behind this complex data is beneficial to a business by identifying newly arising issues and efficiently monitoring call topics. This can be done by two common NLP methods, sentiment analysis and topic modeling. Sentiment analysis is a form of text mining used by data scientists to determine emotion behind text. SAS Viya has developed a tool using a combination of NLP and machine learning methods to help users perform sentiment analyses. Topic modeling identifies groups of text that pertain to a similar subject. Identifying the point in a call with the highest friction and understanding the potential cause of that friction can help identify potential process improvements. Through word scoring and topic modeling we can identify this specific point in the call and determine what topics were discussed prior to the friction. In this presentation, we will cover the process, methods, and benefits of performing sentiment analysis and go through a case study utilizing topic modeling.
Start Date
2-7-2023 9:50 AM
End Date
2-7-2023 10:50 AM
Session 1: Improving Customer Experience Through Natural Language Processing
Dakota Room 250 A/C
Unstructured data such as spoken words can be transcribed into text, transformed into structured data, then analyzed using Natural Language Processing (NLP) methods. Putting meaning behind this complex data is beneficial to a business by identifying newly arising issues and efficiently monitoring call topics. This can be done by two common NLP methods, sentiment analysis and topic modeling. Sentiment analysis is a form of text mining used by data scientists to determine emotion behind text. SAS Viya has developed a tool using a combination of NLP and machine learning methods to help users perform sentiment analyses. Topic modeling identifies groups of text that pertain to a similar subject. Identifying the point in a call with the highest friction and understanding the potential cause of that friction can help identify potential process improvements. Through word scoring and topic modeling we can identify this specific point in the call and determine what topics were discussed prior to the friction. In this presentation, we will cover the process, methods, and benefits of performing sentiment analysis and go through a case study utilizing topic modeling.