1 / 16

Multi Order Matching & Knowledge Bridge: Techniques for Post-Processing Search Results

Multi Order Matching & Knowledge Bridge: Techniques for Post-Processing Search Results. CS598CXZ – Spring 2005 Project ID: HLE Presenter: Hieu Le (hieule2@uiuc.edu). Introduction. Task: Re-organizing search results so that minimize the effort of users to examine

Download Presentation

Multi Order Matching & Knowledge Bridge: Techniques for Post-Processing Search Results

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Multi Order Matching & Knowledge Bridge:Techniques for Post-Processing Search Results CS598CXZ – Spring 2005 Project ID: HLE Presenter: Hieu Le (hieule2@uiuc.edu)

  2. Introduction • Task: Re-organizing search results so that minimize the effort of users to examine • A lot of similar works have done. What’s new here?  We care more about helping users examine clusters • We care more about the cohesion of the cluster  We care more about longer phrases, not only single words

  3. Step 1: Clustering • Used technique: Multi Order Matching (MOM) • Consider different lengths of segments • Consider importance of segments to documents

  4. Step 1: Clustering

  5. Step 2: Ordering inside clusters • Used technique: Knowledge Bridge (KB) • Minimize users’ effort to walk through result in a cluster • Minimize knowledge gap inside a cluster

  6. Step 2: Ordering inside clusters

  7. Step 2: Ordering inside clusters ρ(A, C) + ρ(B, C) < ρ(A, B)

  8. Step 3: Ordering clusters • Ranked score of each item in a cluster will contribute to general ranked score of it. • Cohesion of items in a cluster also contribute to general ranked score.

  9. What’ve done so far • Propose 3 steps of re-organizing search results • Developing and implementing MOM • Developing KB • Designing and Implementing user interface • Implement caching function for Google API to avoid limitation of 1000 queries a day.

  10. Remained works • Implementing KB • Developing method for step 3, implementing it • Conducting experiment

  11. Issues • How to quantify the method systematically?

  12. Thanks

More Related