Announcing | TLM (Trustworthy Language Model) for reliable LLM outputs.Learn more.

Customer Service

Cleanlab Studio uses AI to detect and correct errors/outliers in data, and can also deploy robust ML models with 1-click. This data might be customer support requests (text) for intent classification. Or numeric/categorical customer information stored in a table for customer segmentation tasks like churn prediction.
Hero Picture

Case StudyCustomer Requests at Online Bank

To automatically triage future customer requests, intent classification is a standard Machine Learning task in customer service applications that requires well-labeled data. When run on a dataset of customer requests (text) annotated with 10 different intents, Cleanlab Studio found over 5% of dataset labels were incorrect and detected out-of-scope queries (outliers) like "how much is 1 share of aapl" and "is android better than iphone".

This dataset was also used for customer analytics, to determine the relative frequencies of different types of customer requests and which types are most common. However, some conclusions drawn from the original dataset are inaccurate due to the mislabeling and out-of-scope issues. Significantly more accurate conclusions were obtained by running the same analytics on the cleaned version of the dataset obtained from Cleanlab Studio.

Customer Requests at Online Bank

Using Cleanlab Studio to auto-fix label issues in this dataset led to a 16% improvement in prediction error without altering the existing LLM Transformer model or training code. Addressing additional data issues further enhanced accuracy without any model changes.

Businesses striving to make better decisions for customers must rely on accurate data-driven conclusions. These in turn rely on accurate data, which for this customer service application was easy to ensure with Cleanlab Studio.

Case StudyGavagai

Quote from Fredrik Olsson Head of Data Science at Gavagai
At Gavagai, we rely on labeled data to train our models, both publicly available datasets and data we have annotated ourselves. We know that the quality of the data is paramount when it comes to creating machine learning models that can produce business value for our customers.

Cleanlab Studio is a very effective solution to calm my nerves when it comes to label noise!

The tool allows me to upload a dataset and obtain a ranked list of all the potential label issues in the data in just a few clicks. The label issues can then be assessed and fixed right away in the GUI.

Cleanlab should be a go-to tool in every ML practitioners toolbox!
Gavagai provides multilingual text analytics for customer insights. Analyzing reviews, surveys, call transcripts, support tickets, and social media, their platform helps discover, track, and act on customer data to improve Customer Experience.
Company Logo


Help Section Icon
Read about how to improve LLMs through fine-tuning by systematically improving the training data.
Help Section Icon
Quickly produce an improved version of your customer dataset to produce more reliable versions of your existing models and data analyses — all without changing any of your existing code!
Help Section Icon
Supported ML tasks where Cleanlab is particularly effective include: intent recognition, conversational AI, multi-label classification, entity/product recognition, predicting sentiment or emotions. Read more
Help Section Icon
Practice data-centric AI to produce accurate ML models for messy real-world tabular or text data. Cleanlab offers powerful automated Machine Learning capabilities so you can focus on what matters — the data.
Help Section Icon
Read about ensuring high quality evaluation data for LLM prompt selection.
Help Section Icon
Read about handling noisy tabular data to improve XGBoost model.
Help Section Icon
Read about improving data stored in Databricks with Cleanlab Studio.
Help Section Icon
Videos on using Cleanlab Studio to find and fix incorrect labels for:

Cleanlab Studio auto-corrects raw data to ensure reliable insights so you can provide great customer service.