The back end consists of the following components: search, browse, clusters, and query logs. The search component consists of a text index for classical information retrieval queries and a classification index for categorizing the documents. The text index type provides a rich set of text search capabilities for finding content, without returning pages of spurious results. The classification index type is the inverse of the usual text index. The index is created on a table of queries, where the queries define the classification or routing criteria.

The user interface is based on a two-view model where the left side shows structure and the right content. Yapa presents the search results in a two-view model too adding extra features. The left view presents the results in the following structures: a) list, b) categories, and c) clusters. The right view shows content and document operations. The operations are: document with highlighted query terms, document themes, document summary, and document gist.

Yapa is a south american word that in spanish means "a little bit more for free".

You can see some screenshots here.

Download 9i version
Download 10g version
Download 10g version for Java. Sample code for Kuassi Mensah's book: "Programming the Oracle Database using Java & Web Services".
Download the images

Installation and Notes

9i version
10g version


Q: Where are the visualization metaphors?
A: The StretchViewer information visualization metaphor is not part of the download. It is only available on the 10g CD in the following directory: $ORACLE_HOME/ctx/sample/stretch.

There is Java visualization package available too that you can use to visualize categories/clusters and so on. Please check it out here.

Q: How do I get documents?
A: Yapa doesn't have a crawler at the moment. Assumes you already have content. There are several free crawlers out there like wget. If you want to you can use the utl_http package for writing your crawler or fetching procedure.

Q: Do you have any scripts for dropping stuff?
A: The script dropall.sql drops tables and indices in case you want to start over. The script dropix.sql drops indexes only.

Q: I don't have content in English. What do I do?
A:For multilingual content you should change the settings in the yapaci.sql script.

I would like to hear your comments about this demo. Please send email to: Omar Alonso.

Collaborate 16

Oracle Database Cloud