Changes between Version 4 and Version 5 of MatrixFactorization


Ignore:
Timestamp:
05/06/11 22:14:53 (3 years ago)
Author:
MarinkaZitnik
Comment:

Legend:

Unmodified
Added
Removed
Modified
  • MatrixFactorization

    v4 v5  
    1  
    21= Matrix Factorization Techniques for Data Mining = 
    32 
     
    3433Reference: (Salakhutdinov, 2008). 
    3534 
    36 === nnma === 
    37 Nonnegative matrix approximation. Method for dimensionality reduction with respect on the nonnegativity of input data. Multiplicative iterative scheme.[[BR]] 
    38 Reference: (Sra, 2006). 
    39  
    4035=== psmf === 
    4136Probabilistic Sparse MF. PSMF allows for varying levels of sensor noise, uncertainty in the hidden prototypes used to explain the data and uncertainty as to the prototypes selected to explain each data vector.[[BR]] 
     
    4338 
    4439=== bd === 
    45 Bayesian decomposition. A Bayesian treatment of NMF, based on a normal likelihood and exponential priors, Gibbs sampler to approximate the posterior density.[[BR]] 
     40Bayesian decomposition. A Bayesian treatment of NMF, based on a normal likelihood and exponential priors, MCMC sampling method (Gibbs) to approximate the posterior density.[[BR]] 
    4641Reference: (Schmidt, 2003). 
    4742 
    48 === bfrm === 
    49 Bayesian factor regression model. Markov chain Monte Carlo technique.[[BR]] 
     43=== icm === 
     44Bayesian model - Iterated conditional modes (ICM) algorithm. The modes of conditional densities have closed form expressions and ICM algorithm has a low computational cost per iteration. [[BR]] 
    5045Reference: (Schmidt, 2003). 
    51  
    52 === i-nmf === 
    53 Interval-valued NMF.[[BR]] 
    54 Reference: (Shen, 2010). 
    55  
    56 === i-pmf === 
    57 Interval-valued PMF.[[BR]] 
    58 Reference: (Shen, 2010). 
    59  
    6046 
    6147== Timeline == 
     
    8268 
    8369'''June 18 -- July 5''' 
    84  * Implementing family of NMF techniques: sNMF, lNMF, NNMA, PMF. 
     70 * Implementing family of NMF techniques: sNMF, lNMF, PMF. 
    8571 
    8672'''July 5 -- July 15''' 
     
    9379 
    9480'''July 15 -- July 25''' 
    95  * Implement Bayesian methods. Bayesian decomposition using Gibbs sampler, MCMC techniques such Bayesian factor regression modeling (BD, BFRM). 
     81 * Implement Bayesian methods. Bayesian decomposition using Gibbs sampler (BD). 
    9682 * Handling PMF on large, sparse and unbalanced datasets (algorithm for probabilistic sparse matrix factorization (PSMF)). 
    9783 
    9884'''July 25 -- July 31''' 
    99  * Adapt PMF model to the interval-valued matrices and implement Interval-valued PMF (I-PMF) and Interval-valued NMF (I-NMF). 
     85 * Implement Bayesian model - Iterated conditional modes algorithm (ICM). 
    10086 * Improve efficieny of the code, bug removal, exception handling, additional testing. 
    10187 
     
    11298 
    11399 * Extend Bayesian methods (variational BD, linearly constrained BD). 
     100 * Adapt PMF model to the interval-valued matrices and implement Interval-valued PMF (I-PMF) and Interval-valued NMF (I-NMF) (Shen, 2010). 
     101 * Nonnegative matrix approximation. Method for dimensionality reduction with respect on the nonnegativity of input data. Multiplicative iterative scheme (Sra, 2006). 
    114102 
    115103----