Mining for interactive identification of users' information needs

To facilitate searching and navigation, much information has been hierarchically organized into categories with different levels of generality. However, users still suffer information overload in querying a hierarchical information space, since they often cannot make their aspects of interest precise enough. One way to alleviate the problem is to interactively identify those information categories that correspond to the users' information needs (INs). In that case, information of interest may be found in a more dedicated space (i.e. subset of categories), promoting both search precision and user satisfaction. This paper presents a technique that employs text mining to build each category's profile through which users' INs may be interactively identified. The profiles are mined incrementally so that the system may adapt to the ever-changing information space. The technique is shown to be effective in mapping users' INs to suitable categories without requiring the users to enter long queries, conduct many interactions, and suffer heavy cognitive load. It may serve as an intelligent intermediary in various applications that link users to suitable information categories and service departments.

[1]  Daphne Koller,et al.  Hierarchically Classifying Documents Using Very Few Words , 1997, ICML.

[2]  Yiming Yang,et al.  A re-examination of text categorization methods , 1999, SIGIR '99.

[3]  Tom M. Mitchell,et al.  Improving Text Classification by Shrinkage in a Hierarchy of Classes , 1998, ICML.

[4]  Koji Eguchi.  Adaptive cluster-based browsing using incrementally expanded queries and its effects (poster abstract) , 1999, SIGIR '99.

[5]  Yoram Singer,et al.  Context-sensitive learning methods for text categorization , 1996, SIGIR '96.

[6]  Yiming Yang,et al.  A Comparative Study on Feature Selection in Text Categorization , 1997, ICML.

[7]  James P. Callan.  Learning while filtering documents , 1998, SIGIR '98.

[8]  Sholom M. Weiss,et al.  Automated learning of decision rules for text categorization , 1994, TOIS.

[9]  Kristian J. Hammond,et al.  Question Answering from Frequently Asked Question Files: Experiences with the FAQ FINDER System , 1997, AI Mag..

[10]  Ragnar Nordlie,et al.  “User revealment”—a comparison of initial queries and ensuing question development in online searching and in human reference interactions , 1999, SIGIR '99.

[11]  Mark S. Staveley,et al.  Phrasier: a system for interactive document retrieval using keyphrases , 1999, SIGIR '99.

[12]  Amanda Spink,et al.  Real life information retrieval: a study of user queries on the Web , 1998, SIGF.

[13]  Dragomir R. Radev,et al.  Question-answering by predictive annotation , 2000, SIGIR '00.

[14]  Makoto Iwayama,et al.  Relevance feedback with a small number of relevance judgements: incremental relevance feedback vs. document clustering , 2000, SIGIR '00.

[15]  Hwee Tou Ng,et al.  Feature selection, perceptron learning, and a usability case study for text categorization , 1997, SIGIR '97.

[16]  Amanda Spink,et al.  Users and Intermediaries in Information Retrieval: What Are They Talking About? , 1997 .

[17]  Yoram Singer,et al.  Boosting and Rocchio applied to text filtering , 1998, SIGIR '98.

[18]  Wai Lam,et al.  Using a generalized instance set for automatic text categorization , 1998, SIGIR '98.

[19]  W. Bruce Croft,et al.  Combining classifiers in text categorization , 1996, SIGIR '96.

[20]  Yiming Yang,et al.  An example-based mapping method for text categorization and retrieval , 1994, TOIS.

[21]  Peter Bruza,et al.  Interactive Internet search: keyword, directory and query reformulation mechanisms compared , 2000, SIGIR '00.

[22]  Susan T. Dumais,et al.  Hierarchical classification of Web content , 2000, SIGIR '00.