We contributed new models, algorithms, and software for solving a number of important problems in the field of Information Retrieval. We developed the infrastructure of mobile agents and demonstrated that they could be used to solve the problem of distributed information retrieval. We developed a new model and a number of efficient algorithms for automatically organizing data, and we have shown that this algorithm is both highly accurate and efficient for its stated purpose and that it can be used to help solve a number of other longstanding problems in the field of Information Retrieval, such as information filtering and persistent query retrieval. We developed the first theoretically grounded models and algorithms for metasearch, and we demonstrated that these new algorithms equal or exceed the performance of well known, highly tuned, and largely ad hoc methods. Finally, we developed a unified framework for solving the longstanding problems of metasearch, pooling, and system evaluation, demonstrating for the first time that these three seemingly disparate problems are in fact ``three sides'' of the same coin.