System, representation, and method providing multilevel information retrieval with clarification dialog
First Claim
1. An information retrieval system, comprising:
- a hierarchal analysis component that receives a query and processes probabilities associated with N categories that are collectively assigned a top-level classifier and individually assigned sublevel classifiers, each category having one or more topics, N being an integer, at least one of the one or more topics associated with a prior probability defined prior to receipt of the query, the prior probability indicating a likelihood that a particular topic is desired absent additional information; and
an interactive component that provides feedback derived from the query, the probabilities associated with the N categories, and the prior probability associated with the at least one topic, the feedback utilized to determine at least one category of the N categories to facilitate retrieval of at least one of the one or more topics.
2 Assignments
0 Petitions
Accused Products
Abstract
An information retrieval system, including a learning and real-time classification methodology, is provided in accordance with the present invention. The system includes a hierarchal analysis component that receives a query and processes probabilities associated with N categories, each category having one or more topics, wherein N is an integer. An interactive component drives clarification dialog that is derived from the query and the probabilities associated with the N categories and the one or more topics. The clarification dialog, driven by a rule-based policy, a decision-theoretic analysis considering the costs of dialog to focus the results versus the costs of browsing larger lists, or combinations of rules and decision-theoretic analysis is employed when valuable to determine at least one category of the N categories to facilitate retrieval of at least one of the topics.
160 Citations
38 Claims
-
1. An information retrieval system, comprising:
-
a hierarchal analysis component that receives a query and processes probabilities associated with N categories that are collectively assigned a top-level classifier and individually assigned sublevel classifiers, each category having one or more topics, N being an integer, at least one of the one or more topics associated with a prior probability defined prior to receipt of the query, the prior probability indicating a likelihood that a particular topic is desired absent additional information; and an interactive component that provides feedback derived from the query, the probabilities associated with the N categories, and the prior probability associated with the at least one topic, the feedback utilized to determine at least one category of the N categories to facilitate retrieval of at least one of the one or more topics. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25)
-
-
26. A method providing information retrieval from a database, comprising:
-
assigning prior probabilities to one or more topics prior to receipt of a query, the prior probabilities relate to a likelihood that a particular topic is desired by a user absent additional information; determining probabilities associated with one or more categories that are associated with the one or more topics upon receipt of a query through employment of a top-level classifier assigned collectively to a plurality of categories that include the one or more categories and sublevel classifiers individually assigned to each category within the plurality of categories; providing feedback that is derived from the query, the prior probabilities, and the determined probabilities associated with the one or more categories and the one or more topics; and resolving at least one category of the one or more categories based upon the feedback to facilitate retrieval of at least one of the one or more associated topics. - View Dependent Claims (27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38)
-
-
37. A system providing information retrieval, comprising:
-
means for assigning prior probabilities to one or more topics prior to receipt of a query, the prior probabilities relate to a likelihood that a particular topic is desired by a user absent additional information; means for determining probabilities associated with N categories upon receipt of a query, each category having at least one of the one or more topics, N being an integer greater than one, the probabilities are determined through employment of a top-level classifier collectively assigned to the N categories and sublevel classifiers individually assigned to each of the N categories, the top-level classifier and sublevel classifiers are provided by at least one of a Support Vector Machine, Naive Bayes, Bayes Net decision tree, similarity-based, vector-based and a Bayesian-based classification model; means for providing feedback that is derived from a query, the prior probabilities, and the probabilities associated with the N categories; and means for determining at least one category of the N categories based upon the feedback to facilitate retrieval of at least one of the one or more topics.
-
Specification