South KoreaTravel and aviation2025 – ongoingNOL Universe
Machine learning categorization of customer consultation records
NOL Universe cut customer consultation categorization from over a day to under two hours with machine learning
Problem
NOL Universe, the Korean travel and leisure platform formerly known as Yanolja, had to sort a large volume of customer service consultation records into categories. The job took more than a day.
Approach
The customer service team used the machine learning capabilities of the Databricks platform to categorize consultation history records automatically. The work ran alongside a wider rebuild of the company's data environment on Databricks.
Results
Takeaway
Sorting support records by category is a routine task that machine learning can take over. Here it turned a day-long job into one that finishes within a working morning.
Categories
- Industry
- Travel and aviation
- Business function
- Customer service
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes how NOL Universe built a new data management environment on its platform. The customer service team used machine learning to categorize consultation records automatically, a task that took more than a day and now finishes in under two hours. The page also gives batch processing time and cost figures, which come from the data platform.
Full details are in the original sources.