Fichas
Sin título
Fichas del conjunto (13)
- 1Complete Linkage Algorithm1) Compute distance matrix (euclidean distance)
2) Find the two points that have the min distance out of all distances
3) Merge points
4) Create row with MAX points from each
5) Repeat until no more
6) Create dendogram - 2Single Linkage Algorithm1) Compute distance matrix (euclidean distance)
2) Find the two points that have the min distance out of all distances
3) Merge points
4) Create row with MIN points from each
5) Repeat until no more
6) Create dendogram - 3Group Linkage Algorithmx
- 4Bisecting K-Means Algorithmx
- 5DBScan Algorithmc
- 6Cosine SimilarityA = (0, 1, 3)
B = (1, 2, 0)
1) Calculate Dot Product
= (0*1)+(1*2)+(3*0)
2) Calculate ||A|| and ||B||
||A|| = sqrt(0^2+1^2+3^2)
3) Put into formula
A*B/||A||*||B|| - 7PageRank Algorithm (Start & End)1) Create vector for each node in network. To create the vector, we only include '1' for the outbound link. If there's more than 1 we then divide by number of outbound links
2) Create table with the following headings:
x, O1, P_0(x)
Where x = node
O1 = Num of outbound links
P_x(X) is the value at iteration x for node X
END) We stop when the values stabilise - 8PageRank Algorithm (Formula)
- 9Degree CentralityImportant nodes are those that are heavily linked to other nodes
Undirected = d(i)/n-1
DIrected = d_o(i)/n-1 - 10Closeness Centralityd
- 11Betweenness Centralityf
- 12Degree PrestigeUses In-Links
Directed = d_i(i)/n-1 - 13Proximity PrestigeWe look at the number of nodes that can reach Node i, then divide by nuymber that can
Distances/total
Información
- Descripción
- Fichas sobre Sin título, creado por Simon Hutchinson el 02/05/2024.
- Creado por
- Simon Hutchinson
- Visitas
- 0
- Valoración