Learning to Order Things

There are many applications in which it is desirable to order rather than classify instances. Here we consider the problem of learning how to order instances given feedback in the form of preference judgments, i.e., statements to the effect that one instance should be ranked ahead of another. We outline a two-stage approach in which one first learns by conventional means a binary preference function indicating whether it is advisable to rank one instance before another. Here we consider an on-line algorithm for learning preference functions that is based on Freund and Schapire's "Hedge" algorithm. In the second stage, new instances are ordered so as to maximize agreement with the learned preference function. We show that the problem of finding the ordering that agrees best with a learned preference function is NP-complete. Nevertheless, we describe simple greedy algorithms that are guaranteed to find a good approximation. Finally, we show how metasearch can be formulated as an ordering problem, and present experimental results on learning a combination of "search experts," each of which is a domain-specific query expansion strategy for a web search engine.

[1]  D. Black.  The theory of committees and elections , 1959 .

[2]  G. Thompson,et al.  The Theory of Committees and Elections. , 1959 .

[3]  John G. Kemeny,et al.  Mathematical models in the social sciences , 1964 .

[4]  W. S. Cooper.  Expected search length: A single measure of retrieval effectiveness based on the weak ordering action of retrieval systems , 1968 .

[5]  Peter C. Fishburn,et al.  Utility theory for decision making , 1970 .

[6]  John G. Kemeny,et al.  Mathematical models in the social sciences , 1964 .

[7]  P. Fishburn.  The Theory Of Social Choice , 1973 .

[8]  Zvi Galil,et al.  Cyclic Ordering is NP-Complete , 1977, Theor. Comput. Sci..

[9]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[10]  F. Roberts.  Measurement Theory with Applications to Decisionmaking, Utility, and the Social Sciences: Measurement Theory , 1984 .

[11]  Michael McGill,et al.  Introduction to Modern Information Retrieval , 1983 .

[12]  S. French.  Decision Theory: An Introduction to the Mathematics of Rationality , 1986 .

[13]  Paul E. Utgoff,et al.  Learning a Preference Predicate , 1987 .

[14]  N. Littlestone.  Learning Quickly When Irrelevant Attributes Abound: A New Linear-Threshold Algorithm , 1987, 28th Annual Symposium on Foundations of Computer Science (sfcs 1987).

[15]  Lynn A. Streeter,et al.  Comparing and combining the effectiveness of latent semantic indexing and the ordinary vector space model for information retrieval , 1989, Inf. Process. Manag..

[16]  M. Trick,et al.  Voting schemes for which it can be difficult to tell who won the election , 1989 .

[17]  Paul E. Utgoff,et al.  Two Kinds of Training Information For Evaluation Function Learning , 1991, AAAI.

[18]  S. Shapiro,et al.  Mathematics without Numbers , 1993 .

[19]  Garrison W. Cottrell,et al.  Automatic combination of multiple ranked retrieval systems , 1994, SIGIR '94.

[20]  Manfred K. Warmuth,et al.  The Weighted Majority Algorithm , 1994, Inf. Comput..

[21]  Paul B. Kantor,et al.  Decision Level Data Fusion for Routing of Documents in the TREC3 Context: A Base Case Analysis of Worst Case Results , 1994, TREC.

[22]  Ellen M. Voorhees,et al.  The Collection Fusion Problem , 1994, TREC.

[23]  Paul D. Seymour,et al.  Packing directed circuits fractionally , 1995, Comb..

[24]  Tom M. Mitchell,et al.  Using the Future to Sort Out the Present: Rankprop and Multitask Learning for Medical Risk Evaluation , 1995, NIPS.

[25]  Yiyu Yao.  Measuring retrieval effectiveness based on user preference of documents , 1995 .

[26]  Joseph Naor,et al.  Divide-and-conquer approximation algorithms via spreading metrics , 1995, Proceedings of IEEE 36th Annual Foundations of Computer Science.

[27]  Yiyu Yao,et al.  Measuring Retrieval Effectiveness Based on User Preference of Documents , 1995, J. Am. Soc. Inf. Sci..

[28]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[29]  Ellen M. Voorhees,et al.  Learning Collection Fusion Strategies for Information Retrieval , 1995, ICML.

[30]  David B. Shmoys,et al.  Cut problems and their application to divide-and-conquer , 1996 .

[31]  Oren Etzioni,et al.  Efficient information gathering on the Internet , 1996, Proceedings of 37th Conference on Foundations of Computer Science.

[32]  Dorit S. Hochbaum,et al.  Approximation Algorithms for NP-Hard Problems , 1996 .

[33]  Oren Etzioni,et al.  Dynamic Reference Sifting: A Case Study in the Homepage Domain , 1997, Comput. Networks.

[34]  Dorit S. Hochba,et al.  Approximation Algorithms for NP-Hard Problems , 1997, SIGA.

[35]  Yoav Shoham,et al.  Fab: content-based, collaborative recommendation , 1997, CACM.

[36]  Paul Resnick,et al.  Recommender systems , 1997, CACM.

[37]  Bonnie Berger,et al.  Tight Bounds for the Maximum Acyclic Subgraph Problem , 1997, J. Algorithms.

[38]  Guy,et al.  Approximating Minimum Feedba k Sets andMulti uts in Dire ted Graphs , 1998 .

[39]  Yoram Singer,et al.  An Efficient Boosting Algorithm for Combining Preferences by , 2013 .

[40]  Joseph Naor,et al.  Approximating Minimum Feedback Sets and Multicuts in Directed Graphs , 1998, Algorithmica.

[41]  Dayne Freitag,et al.  A Machine Learning Architecture for Optimizing Web Search Engines , 1999 .