|
Title:
|
SOLVING CROSSWORD PUZZLES VIA THE GOOGLE API |
|
Author(s):
|
David E. Goldschmidt , Mukkai Krishnamoorthy |
|
ISBN:
|
972-99353-0-0 |
|
Editors:
|
Pedro Isaías and Nitya Karmakar |
|
Year:
|
2004 |
|
Edition:
|
1 |
|
Keywords:
|
Crossword Puzzles, Google, RDF, Searching, Semantic Web, XML. |
|
Type:
|
Full Paper |
|
First Page:
|
382 |
|
Last Page:
|
389 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
The Google API enables software agents to query and use search results from the large collections of data available via the ever-popular Google search engine. Web searches using Google are exposed to over 4 billion pages, many of which are cached within Google. While the Google API may be used to produce customized user interfaces to Google, the API also provides direct programmatic access to the subset of the Web covered by Google. In this paper, we present a fresh approach to solving crossword puzzles by making use of the Google API. Our system, the Google CruciVerbalist (GCV), reads XML-encoded crossword puzzles, derives answers to clues via the Google API, and uses a refined depth-first search algorithm to populate the crossword grid. GCV has successfully solved smaller puzzles, especially ones containing pop-culture and fill-in-the-blank types of clues. Based on this ongoing work, limitations of current search technologies are identified. To overcome these limitations, we look ahead to semantic queries via the emerging Semantic Web, including techniques using RDF that augment the Google search engine with semantic information, enabling semantically rich queries beyond the current capabilities of Google. |
|
|
|
|
|
|
