Iterative technique for phrase query formation and an information retrieval system employing same

A method of selectively searching an automated data base with data processing apparatus, said data base containing a corpus of documents comprising sequences of word data stored as stop-words and non-stop-words in a memory, said method comprising the steps of:

• a) input at least one query word to said data processing apparatus;
• b) using said data processing apparatus, determining a word data search key based upon said at least one query word;
• c) using said data processing apparatus, searching said document corpus to identify all occurrences of a match between said search key and said document corpus word data;
• and
• d) using said data processing apparatus, displaying each match as a phrase containing the word data matching said search key, a single non-stop-word next adjacent to said matching word data in a distinctive form different from the display of other word data in the displayed phrases, and all intervening stop-words between said matching word data and said single next adjacent non-stop-word.

A document retrieval system in which is stored a corpus of documents comprising sequences of word data stored as stop-words and non-stop-words in a memory, an apparatus for selectively searching through the corpus of documents comprising:

• means for receiving at least one query word input by an operator of the document retrieval system;
• means for searching through said document corpus and identifying all occurrences of a match between said document corpus word data and a search key determined based upon said at least one query word;
• and
• means for displaying each match as a phrase containing the word data matching said search key, a single non-stop-word next adjacent to said matching word data, and all intervening stop-words between said matching word data and said single non-stop-word, wherein said means for displaying simultaneously displays phrases for multiple matches so that the respective non-stop-words from each phrase are aligned with each other in a common column.

A document retrieval system in which is stored a corpus of documents comprising sequences of word data stored as stop-words and non-stop-words in a memory, an apparatus for selectively searching through the corpus of documents comprising:

• means for receiving at least one query word input by an operator of the document retrieval system;
• means for searching through said document corpus and identifying all occurrences of a match between said document corpus word data and a search key determined based upon said at least one query word;
• and
• means for sequentially displaying each match as a phrase containing the word data matching said search key, at least one non-stop-word next adjacent to said matching word data in a distinctive form different from the display of other word data in the displayed phrases, and all intervening stop-words between said matching word data and said at least one non-stop-word, each phrase being displayed as no more than a single line of text.