Shopping on the Internet has become a convenient way of purchasing commodities of choice. A key contribution of agent technology will be to develop personal agents that can assist users by generating, filtering, collecting, or transforming information available on the Internet. We believe that knowledgeable agents can increase the productivity and satisfaction on online shoppers by suggesting query reformulations to obtain products and services of particular interest to the user. A consumer interested in a particular product class, e.g., a consumer electronic item, can use such an agent to locate a desired product fitting a specified budget from one of several online stores. Apart from finding good deals on a product, such a “buyer’s agent” can also suggest a higher quality alternatives for even a slightly higher price. We identify the types of query reformulation that will be particularly useful for online shoppers. We characterize the information requirements of a domain o...