Skip to content

kangqiwang/Information_Retrieval

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

19 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Information_Retrieval

image Retrieval

background

it implements the retrieval the relevant images based on the controlled vocabulary. it will rank the image and output the similarity based on Cosine Similarity. the programming also implements the spelling error correction.

input: the controlled vocabulary

output: retrieval the relevant images

  1. the function:
  2. the string-string correction - using Damerau-Levenshtein distance
  3. describe the concept of vocabulary - using tree structure
  4. the Synonyms - build our own dictionary, accuatually can get information from Intenet, BUT give up due to the low accuracy.
  5. the image retrieval - similarity : Cosine or Jaccard
  6. the object distance in image - similarity : Cosine or Jaccard

the result:

alt text

output: the value of similarity and the corresponding image

the main idea :

  1. use the inputed vocabulary compare with each known image
  2. get the value of similarity
  3. ranking
  4. output