10

More intelligent travel developments The services being developed for AI’s cognitive computing platforms are growing at an exponential rate. Following are some services currently available that can be applied to the travel industry. Natural Language Classifier (NLC): This service applies cognitive computing techniques to predict the best matching, predefined class for any given input query. Depending on the output class, the cognitive travel platform is able to return more accurate results, or recognize if the query lacks specific relevant information. For example, if a user searches for “Where can I crash in Austin tonight,” the NLC understands that the query is actually asking for suggestions on places to stay or lodging options in Austin—not a car crash —even though it wasn’t explicitly specified in the question. Relationship Extraction: To help analytics engines more easily understand the meaning of a sentence, this service first extracts all entities that were mentioned in the sentence, followed by the specific relationships between those entities. Based on a user’s blogs, emails or tweets, the cognitive travel engine is able to identify all the places and events that he or she wrote about and thus provide more personalized recommendations in the future. For example, a user’s blog can easily be analyzed to pull out key themes. If every time the user mentions lodging in the blog, the focus is on small boutique hotels, a travel company would know to show boutique hotels at the top of the search results for this user. Concept Expansion: This service analyzes a body of text for words or phrases that are contextually related. For example, it knows that “The Big Apple” refers to New York City, “honeymoon” indicates a romantic type trip or that “wanting to go off the beaten path” is asking for a unique getaway travel experience. 10