Case Study: a leading travel aggregator and booking engine improves chatbot responses with Innodata

a leading travel aggregator and booking engine improves net promoter score with Innodata's multilingual booking assistant chatbot

A leading travel aggregator and booking engine required highly accurate, multilingual datasets to train its AI-powered booking assistant chatbot. The challenge was to enable the chatbot to accurately identify customer intent and respond to questions about specific hotels and global locations in multiple languages. They partnered with Innodata for this data annotation service.

Innodata's solution was to annotate incoming chatbot messages by identifying mentions of hotels and locations and categorizing the intent of customer utterances in English, Chinese, and French. They utilized a double-blind annotation process with an adjudicator to ensure high accuracy and quality. The impact was that the travel aggregator received the highly accurate datasets needed for their chatbot to provide relevant responses in multiple languages, which improved their net promoter score.