Iterative Aggregation/Disaggregation(IAD)

Iterative Aggregation/Disaggregation(IAD)

Iterative Aggregation/Disaggregation(IA D) By: Jesse Ehlert Dustin Wells Li Zhang Introduction What are we trying to do? We are trying to find a more efficient way than the power method to compute the pagerank vector.

How are we going to do this? We are going to use an IAD from the theory of Markov Chains to compute the pagerank vector. We are going to apply the power method to Markov Chains We will represent the web by a Markov chain. Markov chain is a stochastic process describing a chain of events.

Consist of a set of states S = {s 1, , sn} Web pages will be the states Probability to move from state s i to state sj in one step is pij. We can represent this by a stochastic matrix with entries pij Probabilistic vector v is a stationary distribution if: vT = v T G This means that the PageRank vector is also a stationary distribution vector of the Markov chain represented by the matrix G

Aggregation/Disaggregation Approach Main idea to compute the pagerank vector v is to block the matrix G so the size of the problem is reduced to about the size of one of the diagonal blocks. In fact (I G11) is non singular. Then we define: and S to be

Aggregation/Disaggregation Approach Cont. From the previous slide we can show that I G = LDU Thus, because U is nonsingular, we have: From the last equation, we can get v2T = v2TS which implies that V2 is a

stationary distribution of S. If u2 is the unique stationary distribution of S with then we have: Aggregation/Disaggregation Approach Cont. We need to find an expression for v1 Let A be the aggregated matrix associated to G, defined as: What we want to do now is find the stationary

distribution of A. From vTLD = 0, we can get: v1T(I G11) v2TG21 = 0 If we rearrange things, we can get Aggregation/Disaggregation Approach Cont. From v2T = v2TS, we also have:

From the previous three statements we can get an expression for v1. Theorem 3.20 (Exact aggregation/disaggregation) Theorem 3.20 Theorem 3.20 Cont. Instead of finding the stationary distribution of

G, we have broken it down to find the stationary distribution of two smaller matrices. ProblemForming the matrix S and computing its stationary distribution u2 is very expensive and not very efficient. Solution: Use an approximation This leads us to Approximate Aggregation

Matrix Approximate Aggregation Matrix We now define the approximate aggregation matrix as: The only difference between this matrix and the previous aggregation matrix is the last row where we use an arbitrary probabilistic vector

that plays the role of the exact stationary distribution u2. In general this approach does not give a very good approximation to the stationary distribution of the original matrix G. To improve the accuracy, we add a power method step. Approximate Aggregation Matrix

Typically, we will have so that the actual algorithm to be implemented consists of repeated applications of the algorithm above. This gives us an iterative aggregation/disaggregation algorithm (IAD) Aggregation/Disaggregation Algorithm (IAD) using Power Method

As you can see from above, we still need to compute the stationary distribution of , IAD Cont. First, we write so that we get rid of G 22 We then let

From we have: IAD Cont. Now we will try to get some sparsity out of G. We will write G like we did before: G = H + auT + (1 )euT . From the blocking of G, we will block the matrices H, auT and euT

for some matrices A, B, C, D, E, F, J, K. From here you can see IAD Cont. We now take G11, G12 and G21 and plug them into We get: For the iterative process of power method within IAD, we give an arbitrary initial guess and iterate according to the formulas above for the next approximation

until our tolerance is reached. Combine Linear Systems and IAD Before, we had This can be written as Combine Linear Systems and IAD Cont.

The problem with this is the matrices G 11, G12 and G21 are full matrices which means the computations at each step are generally very expensive Combine Linear Systems and IAD Cont. We will return to the original matrix H to find some sparsity.

From this equation, we can look at G11 in more depth to get: We will use the fact that equation Note: we used Note: we used to simplify the to get:

Using Dangling Nodes We can reorder H by dangling nodes so that H21 is a matrix of zeros Then our equation from before reduces to: We approximate as: We can show that:

Linear Systems and IAD Process Combined Now, we combine ideas from IAD and linear systems, with H arranged by dangling nodes, to get the process below: Conclusion Instead of finding the stationary distribution of G directly, we have broken it down to find the stationary distribution of smaller matrices, S and

A, which gives us the stationary distribution of G The problem with this is that it was very inefficient. So we found the approximation of the stationary distribution and used power method techniques to improve accuracy. Then we used linear systems along with our iterative aggregation/disaggregation algorithm to find another solution to the pagerank vector.

Recently Viewed Presentations

  • Automatic Speaker Recognition: Recent Progress, Current ...

    Automatic Speaker Recognition: Recent Progress, Current ...

    This is required for non-embedded training. RES: The Source Code Projects related to programs required for speech recognition Initiali This project initializes the HMM models. HMM model parameters are evaluated according to a clustering procedure training This project re-estimates HMM...
  • TEORI SOSIOLOGI - Weebly

    TEORI SOSIOLOGI - Weebly

    Realisme - Nominalisme (Durkheim - M Weber) b. Konflik - Solidaritas (K Marx - E Durkheim) 3. Pola berfikir Simmel dipengaruhi oleh: a. Herbert Spencer => Evolusi (diferensiasi dan heterogenesi masyarakat) b. Immanuel Kant => Empirisme & rasionalisme c. Hegel...
  • Class 8: Swiss Reformation

    Class 8: Swiss Reformation

    Treaty of Augsburg, 1555, included Swiss Cantons; each Canton can chose its own religion. Treaty of Westphalia, 1648, recognized Swiss Confederation. Swiss Federal government officially takes stand of neutrality in European politics and wars. True today
  • Introduction to Travel and Tourism Better by the

    Introduction to Travel and Tourism Better by the

    Arial MS Pゴシック Calibri Blank Presentation 1_Blank Presentation Introduction to Travel and Tourism Better by the Bunch: The Tour Industry Today Common Kinds of Tours PowerPoint Presentation Other Kinds of Tours The Tour Manager Pros & Cons of Tour Managing...
  • LIVD FHIR Mapping

    LIVD FHIR Mapping

    profiling composition. let's ignore this box - we cannot figure out how it maps into the LIVD document - may need as part of FHIR organization so then we use it profiling catalogEntry. We understand that Entry is needed to...
  • DASA-CE Training INTRODUCTION TO COST MANAGEMENT Cost Management

    DASA-CE Training INTRODUCTION TO COST MANAGEMENT Cost Management

    Life Cycle Cost Estimate: ... Enterprise Resource Planning, or ERP, is an enabler to Cost Management "Culture" by providing the necessary technology. The Army Cost Model is being designed into GFEBS which utilizes the SAP ERP application.
  • Smoothing and analyzing 1D signals

    Smoothing and analyzing 1D signals

    Today. Clinic this evening (here), Greg on hashing. Associative arrays. Efficiency: Asymptotic analysis, effects of locality. Hashing. Additional requirements for cryptographic hashing
  • Logic 101 - Literary Cudas

    Logic 101 - Literary Cudas

    The Toulmin Method of Argument. Toulmin's basic understanding of argument includes several elements: A claim. Groundsthat state the reason for the claim. Qualifications that identify possible exceptions to the claim. Based on evidence of some sort. A warrant that explains...