|This page is a proposal for a new Wikimedia Foundation Sister Project.|
|What is the proposed name for the project?||WikiSandy|
|Proposed project tagline||Be Informed...|
What is the project purpose? What will be its scope? How would it benefit to be part of Wikimedia?
|Provide a Wikipedia search service that indexes Wikipedia data semantically, based on sentence structure; subject, subject complement, or direct object, etc. versus just key words. Recognize information that is not directly communicated by the author, by relating acronyms, abbreviations, and compound nouns to appropriate subject matter within an article. Results will be ordered and prioritized by the strength of the correlation of search term to the sentences returned. Results will provide full sentences where possible, with deep links to those sentences, making it possible for users to jump directly to those sentences of interest. Such a tool will improve the search experience within Wikipedia and increase the value of the Wikipedia data.|
|How many wikis?
Will there be many language versions or just on one multilingual wiki?
|Just English, for now.|
|How many languages?
Is the project going to be in one language or in many?
|Just English, for now.|
|Proposed project website address||TBD|
|Proposed logo for the project||TBD|
If the project requires any new features that the MediaWiki software currently doesn't have, please describe in detail. Are additional MediaWiki extensions needed for the project?
|We don’t believe that any MediaWiki extensions are required|
Project will involve creating a Wikipedia Contextual Search Engine (“WCSE”) that can parse through the complete Wikipedia website, extracting knowledge contained within. Such knowledge will be ascertained by analyzing sentences to determine sentence subjects, subject complements, and direct objects, etc. to relate and comprehend knowledge. The project will then involve developing a query system that will present search results in order of their sentence strength and topic correlation to the query term. The results where possible should present the actual results in sentence form and provide links that can bring the user directly to that sentence within that Wikipedia article. An alpha version of this WCSE has already been created and can be tried at www.wikisandy.org.
To allow access to the Wikipedia Contextual Search Engine, the project will require creating a link on the Wikipedia website. Initially the link could bring the user to www.wikisandy.org which would need to be rebranded as Wikipedia’s search engine. Eventually the WCSE could be integrated into Wikipedia’s website.
Since Wikipedia is constantly being updated with new articles as well as changes to existing articles, the project would involve creating a queue of articles that have been added or updated, and then feeding those articles into the WCSE for processing to keep the WCSE’s knowledge base current.
Much of this project has already been developed, and is usable at www.wikisandy.org. The part that would need development would be the queueing of new and changed articles for the WCSE to keep up to date. The technology to update the knowledge base is complete, but there would need to be integration into the Wikipedia core engine. Additionally, computing resources would need to be obtained as usage of the WCSE increases.
Thomas H. Cowley https://meta.wikimedia.org/wiki/User:Sandypondfarm email@example.com www.spfllab.com
Wiki research, Wiki deep knowledge search, Wiki Contextual Search
www.wikisandy.org (live today), www.wikiresearch.org (not live today)