An adaptive resource indexing and lookup method and system for a customer self service system that performs resource search and selection and includes a resource library having selectable resources. The method includes steps of : receiving a current user query for requesting resources; receiving a user context vector associated with the current user query, the user context vector comprising data associating an interaction state with the user; mapping each user query and associated context vector to a sub-set of resources from the resource library; and, generating a response set including the sub-set of resources that are most relevant to the user' s query. In an off-line process, an adaptive indexing function is applied for increasing the value of search results for a current user in their context, the adaptive indexing function enhancing the resource indexing functions by increasing their relevance and specificity for mapping user queries to resources. The adaptive indexing process implements a supervised learning algorithm for receiving user interaction data from among a database of user interaction records and resources from the resource library, and adapts resource indexing functions based on a history of user interactions and user feedback with the system as provided in user interaction records. In this manner the supervised learning algorithm optimizes performance of the resource indexing functions as measured by an evaluation metric applied to the user interaction feedback. The result of this invention is the ability to improve a set of resource indexing functions without the need for the user to explicitly train the system. The system and method is especially applicable for a self service system in a variety of customer self service domains including, but not limited to : education, real estate and travel.