CS297-298 Project News Feed

  -Implemented the KS test from Numerical recipes. Not sure if I can use this for all datasets because it is meant for continuous distributions.
  - Gathering results. Should be done in a week.
  - Once this is done, should finihs up the report.

Have been a while since I updated my blog. 

We finally got a version of the lattice algorithm working.

I have started writing up my report.Also, implemented a binner. 

Trying to get data and run binned data against the lattice algorithm  is the priority now.
   

1. Get comments from Eric.
2. Apriori to accomodate larger datasets
3. Fix lattice based on Eric's comments
   

1. Set up meeting with Eric.
2. Line of best fit for timing Apriori-db with Zipf data

   

1. CS298 proposal
2. Zipf distribution implementation
3. Read up on Chi-squared distribution
4. Run lattice algorithm with different datasets and time  them.
5. Graphing - log scale
   

1. Complete proposal for CS298
2. Change Zipf distribution algorithm to avoid collisions of values.
3. Run lattice algorithm with large data-set
   

Actual content of the entry.
1. Complete part 2 of deliverable 3
2. Start working on report.

1.Fix bugs for deliverable 2
2.Start on lattice algorithm (deliverable 3: phase 1)
3.Fix typos in slides.
   

1. Explore using JNI for reading a block
2. Use a data structure other than HashSet (Bag, for instance)
3.Set the size of HashSet , so they don't  resize

   

1. Use good test data
2. Get block size, buffer size as parameters for the Apriori implementation
3. Read Lin2002
4. Pretty up code
5. Search public databases for datasets

   

1. Use good test data
2. Get block size, buffer size as parameters for the Apriori implementation
3. Read Lin2002
4. Pretty up code
5. Search public databases for datasets