A MIND MAP QUERY IN INFORMATION RETRIEVAL

A MIND MAP QUERY IN INFORMATION RETRIEVAL

Ms. Sunayana R. GawdeM.Tech Part I

14109

Original Paper by:

Rihab AyedFarah HarrathiM. Mohsen GammoudiMahran Farhat

Presented in:

Fourth International Conference on Computer Science, Engineering and Applications

(ICCSEA 2014), July 26 – 27, 2014 in Chennai, India

Traditional Approach:Bag of Words-misinterpretation of user’s

need user cannot express clearly the main words of

his idea of search and the less important words.

idea of the user is packaged in a linear form which does not express the nodes in the network of the user’s mind.

Related Work:Pearl trees system:Storing interests in Mind MapsSearch:

Bag of WordsSelecting a Pearl

Problem:Semantic ClusteringEx. “popular terms Information Retrieval”

Proposed Approach

Mind Map Query:

Mind Map query formulation by the user andAn internal representation of the Mind Map

query by the IR system.

Why only Mind Maps?The association aspect: The relative importance of terms:

Example Query:“The definition of a concept in a thesaurus or

by standard norms”Bag of Words: Equal Weighting for all terms: “definition of concept thesaurus standard norms”

Mind Map Query Approach:

Example Query: “a good java course which describes

inheritance and polymorphism, but also contains other notions of java. This course should contain exercises”

Bag Of Words: “good java course inheritance polymorphism

exercises”

Mind Map Query Approach

Example Query:“the definition of a semantic resource, for

example thesaurus, ontology”

Bag Of Words: semantic resource definition

Mind Map Query Approach:

Internal Semantic for Mind Map Queries

Where= The power of importance between levels = The weight attributed to the leafs of the

graph The depth of the node in the query graph h = The height of the query graph

Example Query:“Documents about Precision measure in Information Retrieval, for example: GMAP, MAP”

Calculation:W1=Term Weight of PrecisionW1=W2=W3=W4=

Term Precision was twice more important than other terms.

Experimentation:On medical corpus (CLEF 2009, (Cross

Language Evaluation Forum)74’902 images from 20’000 English journal

articles in Radiology.25 queries in the collection test.

IRS based QueryMind Map Query

Results:

Future Work:

Guessing of Central Idea by the System.

Academic recommender System.

ReferencesRihab Ayed, Farah Harrathi, M. Mohsen Gammoudi and

Mahran Farhat(2014) A Mind Map Query in Information Retrieval: The ‘User Query Idea’ concept and preliminary results. Fourth International Conference on Computer Science, Engineering and Applications (ICCSEA 2014), July 26 – 27, 2014 in Chennai, India

Kamvar, M., Kellar, M., Patel, R. and Xu, Y. (2009) Computers and iPhones and Mobile Phones, a logs based comparison of search users on different devices. In Proceedings of the 18th International Conference on World Wide Web (Madrid, Spain, April 20-24, 2009). WWW'09. ACM, New York,NY,801-810.

Thank You!!