TopCells: Supporting Keyword Search in Text Cube
Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
seminar class
Active In SP
**

Posts: 5,361
Joined: Feb 2011
#1
16-02-2011, 10:33 AM


TopCells: Supporting Keyword Search in Text Cube
TopCells: Keyword Search

 Keyword Search
– Simple but popular (as we can see in Zhai’s course)
– A lot of studies in IR community
 Recently, Keyword Search in DB (DB+IR)
– Finding the close connection between tuples
Motivation
Search a light weighted powerful laptop from customer reviews
Text Cube
 Aggregating Multi-dimensional Text Data
TopCells: Keyword Query
 Simple Keyword Query
– A set of keywords
 Extended Keyword Query
– With dimension constraints
TopCells: Ranking Cells
 Given a keyword query q = {t1, …, tm}
How to rank the cells?
 Relevance of a cell C to the given query
– Average the relevance of documents in a cell
– s(d, q) could be ANY IR scoring formula, like Okapi
Algorithm 1 (one-scan of inverted index)
 Naïve one-scan inverted-index-based algorithm
– Compute s(d, q) for all documents d’s
– For each d, update rel(C, q) for all cells C’s containing d
(There are 2dim cells containing d)
– Output top-k cells C’s with highest rel(C, q)’s
– Deficiency
– Given a query q, equivalent to scanning all cells once
– When the dimensionality is high, the number of cells is huge
– Solution (search space ordering)
– Explore as small number of cells as possible
Algorithm 2 (search-space ordering)
 Monotonicity of relevance score
– Used to estimate upper bounds
– Search-Space-Ordering algorithm
– Starting from single documents
– Bottom-to-up computation: aggregating cells of small sizes into big ones
– Aggregating cells with higher scores first
– Partially-aggregated cell
– Output C, when C is fully-aggregated and rel(C, q) is no less than the upper bound of rel(C’, q) for all partially-aggregated cells C’


download full report
https://netfiles.uiuc.edu/bding3/www/pap...090508.ppt
Reply

Important Note..!

If you are not satisfied with above reply ,..Please

ASK HERE

So that we will collect data for you and will made reply to the request....OR try below "QUICK REPLY" box to add a reply to this page

Quick Reply
Message
Type your reply to this message here.


Image Verification
Please enter the text contained within the image into the text box below it. This process is used to prevent automated spam bots.
Image Verification
(case insensitive)

Possibly Related Threads...
Thread Author Replies Views Last Post
  A Character Segmentation Algorithm for Printed Kannada Text Document uploader 1 1,508 10-01-2015, 12:52 PM
Last Post: zcfqmbrtb
  A MACHINE LEARNING APPROACH FOR IDENTIFYING DISEASE-TREATMENT RELATIONS IN SHORT TEXT project uploader 2 1,791 10-01-2014, 04:56 PM
Last Post: seminar project topic
  Search Engines Work ppt study tips 0 297 22-08-2013, 03:34 PM
Last Post: study tips
  GRAPH SEARCH ALGORITHMS PPT study tips 0 338 16-08-2013, 04:47 PM
Last Post: study tips
  EFFICIENT MULTI-DIMENSIONAL FUZZY SEARCH FOR PERSONAL INFORMATION MANAGEMENT REPORT study tips 0 302 30-07-2013, 02:14 PM
Last Post: study tips
  Enabling Secure and Efficient Ranked Keyword Search over Outsourced Cloud Data pdf study tips 0 459 09-07-2013, 03:46 PM
Last Post: study tips
  BREADTH FIRST SEARCH ALGORITHM ppt study tips 0 342 02-07-2013, 02:52 PM
Last Post: study tips
  supporting efficient and scalable multicasting over mobile report study tips 0 374 01-07-2013, 04:15 PM
Last Post: study tips
  A Novel Method of Significant Words Identification in Text Summarization pdf study tips 0 300 18-06-2013, 01:00 PM
Last Post: study tips
  On Optimizing Overlay Topologies for Search in Unstructured Peer-to-Peer Networks pdf study tips 0 317 01-06-2013, 04:46 PM
Last Post: study tips