Phd Meeting - What've I Done So Far..

Compiled a ppt presentation (Models for Query Matching using Lattice Mining Techniques) with links to what i've done so far :

1)  FCA tools and example
2) FCA & Automatic Grading
3) IslamGrid (Questions & Answers) - An Experiment
4) Lattice Mining & Graph Theory (proof of concept based on Bae et al's method)
5) Lattice Mining & FCA  (a proposed method using Graph Theory & FCA)

Still to explore :
1)  How to compare two graphs/lattices using Matlab
2)  Need to find supports from literature regarding the proposed method in 5)

To do :  Write an extended abstract regarding 4)

PhD Meeting - Updates

Done with FTA (actually i'm not done yet but still have to figure out the significance of FTA (hopefully there is!) to my current research.  I still believe that it lies somewhere around the significance/importance of a node to the similarity/distance measure in lattices  (how I wish... i pray hard for it anyway..)

Discussion :  Focusing on the paper by Bae et. al (2007)

Among the ideas/suggestions were :

1)  Lattice Mining - similar concept to Data/Text/Process Mining except that the information is in the form of lattices

2) Lattice Mining vs. Lattice Warehouse  :
i)  How to extract information from lattice  -  use Graph Theory, ie. Adjacency Matrix to represent the data

ii) Example of a scenario - How to mine similar lattices?
Possible solution :  Compare two lattices (use the Distance Measure  method proposed by Bae et. al)

PhD Meeting - Updates

Discussion on Fault Tree Analysis (FTA) :

1)   How to obtain Minimal Cut Set equation from a FTA diagram

2)  How to calculate the criticality of each node

Questions :

1)  How to link/any relationship between FTA & lattice theory?

2)  WHAT'S NEXT?  (I think it has something to do with similarity, or distance between the nodes)

Came across this paper - Development of Distance Measures for Process Mining, Discovery, and Integration by Bae, et.al (2007).

Anything interesting?

Suggested direction  -  look into graph theory & how to mine information from lattice

Idea :  Lattice Mining via Graph Theory

A New Postgraduate Student @ FST, UMT

Registered as a PG student with UMT today. 

Summary of today's discussion - 27th August 2013, UMT Library , 11 am


1.   3 areas to consider based on the idea of Fault Tree Analysis (FTA) : 
   i.  FTA diagram = FCA lattice

   ii. Corresponding table (based on the FTA diagram) to calculate the criticality  of a node = a table that represents the relationship / dependency among nodes in FCA lattices

  iii. Importance Measure = probability / relevance of each node in the lattice structure (possibility of linking  to degree of similarity among FCA nodes)


2.  Other points to consider :
    i.  Graph Theory -  Convert FCA lattice into its corresponding matrices, then produce 1.ii

   ii.  Get to know PetriNet (input, output & possible use in this study)
Question : How to model FCA  lattices using PetriNet


3.  Writing  (Idea : Formalization of FCA Lattice Using Graph Theory)
Keywords :
   i.   FCA
   ii.   IR
  iii.   Graph Theory

PhD Research Plan

PHASES	TASKS	EXPECTED OUTCOMES
Sem. 1 Sept 2013 - Feb 2014	1.      Review of Literature on – IR, Semantic search, FCA, Graph Theory 2.      Study laws / theories / models & FCA tools – Zipf Law, ConExp, In-Close2, FCABedrock, IR measures	Review ppr (FCA, IR & Graph Theory)
Sem. 2 March  – Aug 2014	1.    More lit. review  – IR, FCA lattice, matrices 2.      Prototype 1 (System#1)	1.   2 conference pprs (1 national & 1 international) 2.      1 journal
Sem. 3 Sept 2014 – Aug 2015	1.      More lit. review  – IR, FCA lattice, matrices 2.      Protoype 2 – an enhancement  (System#2)	1.      2 conference pprs (1 national & 1 international) 2.      1 journal
Sem. 4 Sep 2015 – Feb 2016	1.      More lit. review  – IR, Semantic search, FCA 2.      Protoype 3 – an enhancement to System#2 (System#3)	1.      2 conference pprs (1 national & 1 international) 2.      1 journal
Sem. 5 & 6 March 2016 –  Aug 2017	Testing, evaluation, writing	Thesis writing :  Title : ?? Chap1 - Introduction Chap2 – Lit. Survey Chap3 - Methodology Chap4 - Design of system (basic) Chap5 - Design System#1 – an enhancement to the basic system Chap6 - Design System#2 – an enhancement to System#2 Chap7 – Conclusions, Limitations of the research & Future work References