Digital Equipment Corporation

Overture Services, Inc.

Compaq Computer Corporation, Inc.

Peter G Anick

Rex A Flynn

Suzanne O Artemieff

Farzin Maghoul

John Thrall

Alastair Robert Gourlay

Johannes Leon Marais

Laurent Chavet

Jeffrey D Brennan

David R Hanssen

Suresh Tipirneni

Jeffrey M Robbins

Bryan Alvey

A received query is processed so as to generate an initial group of ranked documents corresponding to the received query. Each document in all or a portion of the documents in the initial group of ranked documents is associated with a respective set of ranked candidate terms such that each candidate term in the respective set of ranked candidate terms is embedded within the document. Each respective set of ranked candidate terms is identified at a time prior to the processing of the received query. In accordance with a selection function, a subset of the candidate terms in one or more of the respective sets of candidate terms is selected. In response to the received query, the initial group of ranked documents and the subset of candidate terms are presented.

Systems and methods for interactive search query refinement

A search engine receives query terms from a client. In response, the search engine executes a search on a web directory to identify zero or more documents that match the query terms. The identified documents are associated with one or more categories. The search engine probabilistically selects one of the categories associated with the identified documents. Each message in a message database is also associated with one or more of the categories. The search engine accesses the message database and selects at least one message associated with the selected category. The search engine returns a web page containing references to the documents matching the query terms and the one or more messages selected from the message database to the client.

Search engine for selecting targeted messages

Iterative information retrieval from a large database of textual or text-containing documents is facilitated by automatic construction of faceted representations. Facets are chosen heuristically based on lexical dispersion, a measure of the number of different words with which a particular search expression co-occurs within a given type of lexical construct (e.g., a noun phrase) appearing in the document set. Words having high dispersion rates represent &ldquo;facets&rdquo; that may be used to organize the documents conceptually in accordance with the search expression, effectively providing a concise, structured summary of the contents of a result set as well as presenting a set of candidate terms for query reformulation.

Method and apparatus for automatic construction of faceted terminological feedback for document retrieval

A language syntax defines statement forms for statements that describe the inflectional morphology of the natural language. A set of language statements that follow the language syntax to describe the inflectional morphology of the natural language is accepted as input into to a computer. Rule statements define a set of morpho-syntactic features corresponding to grammatical distinctions within the parts of speech categories in the natural language and define a set of inflectional morphological paradigms. The inflectional morphological paradigms include form rule statements to describe the construction of word forms and associate with each construction pre-selected morpho-syntactic features.

Method and apparatus for efficient morphological text analysis using a

A language syntax defines statement forms for statements that describe the inflectional morphology of the natural language. A set of language statements that follow a language syntax to describe the inflectional morphology of the natural language is accepted as input into to a computer. Rule statements define a set of morpho-syntactic features corresponding to grammatical distinctions within the parts of speech categories in the natural language and define a set of inflectional morphological paradigms. The inflectional morphological paradigms include form rule statements to describe the construction of word forms and associate with each construction pre-selected morpho-syntactic features.

A method and apparatus for providing an historical perspective into a database of information objects through an efficient method and apparatus for versioning information objects stored in a database as well as an index representative of the information objects is disclosed. The latest versions of each one of the information objects and each one of the entries in the index are maintained in the database. Partial earlier versions of each one of the information objects and each one of the index entries are stored with the latest versions. The partial versions contain only sufficient information about the differences between the earlier versions and the later ones so that any earlier version may be reconstructed. Identification tags are employed as unique identifiers for each one of the plurality of information objects and for different versions of the information objects. Through the exploitation of the relative time ordering of generated identification tags, an efficient index representation and method for performing historical queries is provided.

System for reconstructing prior versions of indexes using records

An information retrieval system including a plurality of indices representative of information stored in the information retrieval system and a dynamic lexicon is disclosed. The system includes memory having a database stored therein, the database being logically divided to include the plurality of indices, an information database having information objects stored therein and a dynamic lexicon which includes a plurality of data items and groups of data items that appear in the information database. A predetermined time variable represents the last time the plurality of indices were reindexed. After changes are made to the lexicon, a time stamp is attached to each one of the plurality of changes to the lexicon to indicate when the change was made to the lexicon. At some specified time interval later, the reindexing process is invoked. This process involves selecting a subset of the plurality of changes made to the lexicon after the predetermined time variable, locating all information objects in the information database that are affected by the plurality of changes to the lexicon, reindexing the portions of the plurality of indices representative of the information objects affected by the changes to the lexicon to reflect the changes in the lexicon, and then updating the predetermined time variable to indicate changes to the lexicon have been processed. The foregoing process is repeated until all changes to the lexicon after the predetermined time have been applied to the plurality of indices.

Method and apparatus for integrating a dynamic lexicon into a full-text

A method and apparatus that translates a natural language query into a Boolean expression to be used to search a database. The Boolean expression is displayed on a screen so that the user can alter the Boolean expression using a mouse or similar input device and re-execute the search. The manipulations performed by the user include moving terms of the query, changing the order in which query terms are evaluated, adding terms, deleting terms, and selecting alternate versions of terms.

Growing community of inventors

Marlboro, MA, United States of America

Peter G Anick