1. Quick Start Guide PDF
   
   or UCINET A google doc
   
2. A simple tutorial
   
   and Entering Data Using the DL Editor
   
3. Matrix Algebra help
   
   and Network analyses via matrix algebra commands
   

1. UCINET_6_Users_Guide.docx (96.48 KB)
   
2. Entering Data Using the DL Editor.docx (432.31 KB)
   

1. Configure network and attribute data for standard software packages
   
2. Apply elementary transformations to matrix data
   
3. Extract and reconfigure network and attribute data
   

1. how to format network data for import into a network analysis software package,
   
2. how to transform network data to make it suitable for different analyses, and
   
3. how to export network data and results for use in other programs.
   
   

• matrix format
  
• list format
  
  • Nodelist
    
  • Edgelist
    

• matrix format takes a lot of space in terms of memory and disk.
  
• Indicating the only relations among the network nodes may be more efficient.
  
  

[JPG image (69.46 KB)]

DL
N=5
FORMAT = FULLMATRIX DIAGONAL PRESENT
SYMMETRIC = YES
LABELS embedded
DATA:
"Beill Smith" "Carrie Jones" "Doug Johnson" "Erric Morrison"
"Erric Morrison" "Finn Cobb"
"Doug Johnson" "Finn Cobb" "Erric Morrison"
"Carrie Jones" "Finn Cobb"

[JPG image (66.1 KB)]

[JPG image (80.2 KB)]

5-3_NodeListData2.txt (255 Bytes) OR 5-3_NodeListData.txt (749 Bytes)

Two mode data

5-6_TwoModeNodelistDlFile.txt (176 Bytes)

DL nr=3, nc=5
format=nodelist2
row labels embedded
column labels embedded
data:
George, Darts, Pool, Dancing
Sue, Dancing, Volleyball
Sally, Dancing, Darts, Basketball

• Be careful:
  
  replicated actor
  missing actor
  
• Data characteristics
  
  had lunch with . . . . data should be symmetric.
  What do you do about it?
  
• Take a look at isolates.
  
• Ethnographic hunch . . . Who's gonna be the center . . . Make a quick picture of the data.
  
  
  
  

• transposing matrices,
  
• symmetrizing,
  
• dichotomizing,
  
• imputing missing values,
  
• combining relations,
  
• combining nodes,
  
• extracting subgraphs, and many more.
  

• Permute: Re-order rows, columns or matrices in a dataset according to a user specified list.
  
• bipartite: Convert a 2-mode dataset into a 1-mode adjacency matrix
  
  • Any 2-mode incidence matrix can be thought of as a bipartite graph. If the 2-modes are actors and events then the bipartite graph consists of the union of the actors and events as vertices with the edges only connecting actors with events (ie no connections between actors or between events). This routine takes a 2-mode incidence matrix and converts it to a 1-mode adjacency matrix of a bipartite graph. If the incidence matrix had n rows and m columns then the resultant adjacency matrix would be a square matrix of dimension m+n.
    
    
    

• To interchange its rows with its columns
  
• When the matrix is non-symmetric, it makes relationships reversed. (Transposing an adjacency matrix is equivalent to reversing the direction of the arrows.)
  
• by suvey's nature, it becomes very useful.
  
  • 'who do you seek advice from?' : the constructed data should be transposed:
    
  • food webs, too.
    
    • which species eat which species?
      
    • whose energy flow to whom? = The matrix is transposed by ecologist.
      

• row x col x layers
  
  • row - col
    
  • row - layer
    
  • col - layer
    
• Padgett dataset: marriage and business ties among Florentine families during the Renaissance period.
  
  
  

  DL
  N=16 NM=2
  FORMAT = FULLMATRIX DIAGONAL PRESENT
  ROW LABELS:
  "ACCIAIUOL"
  "ALBIZZI"
  "BARBADORI"
  "BISCHERI"
  "CASTELLAN"
  "GINORI"
  "GUADAGNI"
  "LAMBERTES"
  "MEDICI"
  "PAZZI"
  "PERUZZI"
  "PUCCI"
  "RIDOLFI"
  "SALVIATI"
  "STROZZI"
  "TORNABUON"
  COLUMN LABELS:
  "ACCIAIUOL"
  "ALBIZZI"
  "BARBADORI"
  "BISCHERI"
  "CASTELLAN"
  "GINORI"
  "GUADAGNI"
  "LAMBERTES"
  "MEDICI"
  "PAZZI"
  "PERUZZI"
  "PUCCI"
  "RIDOLFI"
  "SALVIATI"
  "STROZZI"
  "TORNABUON"
  LEVEL LABELS:
  "PADGM"
  "PADGB"
  DATA:
   0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
   0 0 0 0 0 1 1 0 1 0 0 0 0 0 0 0
   0 0 0 0 1 0 0 0 1 0 0 0 0 0 0 0
   0 0 0 0 0 0 1 0 0 0 1 0 0 0 1 0
   0 0 1 0 0 0 0 0 0 0 1 0 0 0 1 0
   0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0
   0 1 0 1 0 0 0 1 0 0 0 0 0 0 0 1
   0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0
   1 1 1 0 0 0 0 0 0 0 0 0 1 1 0 1
   0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0
   0 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0
   0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
   0 0 0 0 0 0 0 0 1 0 0 0 0 0 1 1
   0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0
   0 0 0 1 1 0 0 0 0 0 1 0 1 0 0 0
   0 0 0 0 0 0 1 0 1 0 0 0 1 0 0 0
   0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
   0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
   0 0 0 0 1 1 0 0 1 0 1 0 0 0 0 0
   0 0 0 0 0 0 1 1 0 0 1 0 0 0 0 0
   0 0 1 0 0 0 0 1 0 0 1 0 0 0 0 0
   0 0 1 0 0 0 0 0 1 0 0 0 0 0 0 0
   0 0 0 1 0 0 0 1 0 0 0 0 0 0 0 0
   0 0 0 1 1 0 1 0 0 0 1 0 0 0 0 0
   0 0 1 0 0 1 0 0 0 1 0 0 0 1 0 1
   0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
   0 0 1 1 1 0 0 1 0 0 0 0 0 0 0 0
   0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
   0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
   0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
   0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
   0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0
  

  
  
• centrality measure
  
  
  

  MULTIPLE CENTRALITY MEASURES
  --------------------------------------------------------------------------------
  
  Input dataset: PADGETT (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch05\PADGETT)
  Output dataset: PADGETT-cent (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch05\PADGETT-cent)
  Treat data as:                          Auto-detect
  Type of scores to output:               Raw scores
  Undefined dist in closeness:            replace with max dist + 1
  
  Network PADGM is directed? NO
  
  Value of Beta was:                      0.305579943942584
  Network PADGB is directed? NO
  
  Value of Beta was:                      0.300228499969756
  Centrality Measures
  
  PADGM
  
                       1       2       3       4       5       6       7       8       9
                  Degree  2local  BonPwr   2Step     ARD Closene Eigenve Between 2StepBe
                 ------- ------- ------- ------- ------- ------- ------- ------- -------
    1 ACCIAIUOL    1.000   6.000 301.290   6.000   5.917  44.000   0.132   0.000   0.000
    2   ALBIZZI    3.000  11.000 557.416  10.000   7.833  35.000   0.244  19.333   2.500
    3 BARBADORI    2.000   9.000 482.693   9.000   7.083  38.000   0.212   8.500   1.000
    4  BISCHERI    3.000  11.000 644.632   8.000   7.200  41.000   0.283   9.500   2.000
    5 CASTELLAN    3.000   9.000 590.363   6.000   6.917  42.000   0.259   5.000   2.000
    6    GINORI    1.000   3.000 171.335   3.000   5.333  48.000   0.075   0.000   0.000
    7  GUADAGNI    4.000  10.000 660.285   9.000   8.083  36.000   0.289  23.167   5.500
    8 LAMBERTES    1.000   4.000 202.770   4.000   5.367  49.000   0.089   0.000   0.000
    9    MEDICI    6.000  14.000 982.689  11.000   9.500  31.000   0.430  47.500  13.500
   10     PAZZI    1.000   2.000 102.991   2.000   4.767  55.000   0.045   0.000   0.000
   11   PERUZZI    3.000  10.000 628.266   6.000   6.783  44.000   0.276   2.000   0.500
   12     PUCCI    0.000   0.000  -0.000   0.000   0.000  90.000   0.000   0.000   0.000
   13   RIDOLFI    3.000  13.000 778.167  11.000   8.000  34.000   0.342  10.333   2.000
   14  SALVIATI    2.000   7.000 333.762   7.000   6.583  42.000   0.146  13.000   1.000
   15   STROZZI    4.000  12.000 811.167   8.000   7.833  38.000   0.356   9.333   3.500
   16 TORNABUON    3.000  13.000 742.852  10.000   7.833  35.000   0.326   8.333   1.500
  
  
  PADGB
  
                       1       2       3       4       5       6       7       8       9
                  Degree  2local  BonPwr   2Step     ARD Closene Eigenve Between 2StepBe
                 ------- ------- ------- ------- ------- ------- ------- ------- -------
    1 ACCIAIUOL    0.000   0.000   0.000   0.000   0.000  90.000   0.000   0.000   0.000
    2   ALBIZZI    0.000   0.000   0.000   0.000   0.000  90.000   0.000   0.000   0.000
    3 BARBADORI    4.000  14.000 756.242   9.000   6.833  47.000   0.390  25.000   4.000
    4  BISCHERI    3.000  10.000 663.468   5.000   5.417  55.000   0.344   2.500   0.500
    5 CASTELLAN    3.000  12.000 754.534   7.000   6.000  50.000   0.391   5.000   0.500
    6    GINORI    2.000   9.000 370.702   7.000   5.417  52.000   0.191   0.000   0.000
    7  GUADAGNI    2.000   7.000 452.922   4.000   4.433  62.000   0.235   0.000   0.000
    8 LAMBERTES    4.000  12.000 838.463   5.000   5.917  54.000   0.435   6.000   2.000
    9    MEDICI    5.000   9.000 471.830   7.000   6.917  49.000   0.241  24.000   9.000
   10     PAZZI    1.000   5.000 142.657   5.000   4.367  58.000   0.073   0.000   0.000
   11   PERUZZI    4.000  14.000 908.501   7.000   6.500  49.000   0.471  13.500   2.000
   12     PUCCI    0.000   0.000   0.000   0.000   0.000  90.000   0.000   0.000   0.000
   13   RIDOLFI    0.000   0.000   0.000   0.000   0.000  90.000   0.000   0.000   0.000
   14  SALVIATI    1.000   5.000 142.657   5.000   4.367  58.000   0.073   0.000   0.000
   15   STROZZI    0.000   0.000   0.000   0.000   0.000  90.000   0.000   0.000   0.000
   16 TORNABUON    1.000   5.000 142.657   5.000   4.367  58.000   0.073   0.000   0.000
  
  
  ----------------------------------------
  Running time:  00:00:01
  Output generated:  29 9 14 16:41:40
  UCINET 6.511 Copyright (c) 1992-2012 Analytic Technologies
  
  

  
  

• Getting rid of the actor by deleting both row and the corresponding column)
  
• In case of symmetric or undirected relations.
  

• Who do you communicate with whom?
  
• asymmetric matrix . . .
  
• by union rule (or OR rule)
  

• For very practical reason
  
  • some methods, especially graph-theoretic methods, are only applicable to binary data.
    
  • Also, dichotomizing with a high cut-off can serve to reduce the density of the network, which is useful in handling large networks
    
• it is usually advisable to dichotomize at different levels ???
  
  

• who do you attend sports events with?
  
• who do you go to the theatre with?, and
  
• who do you go out to dinner with?
  
• combine them into a more general, analytically defined, relation, such as
  
  • who socialized with whom?
    
• Sampson dataset
  
  • Transform - Matrix Operation - Aggregation
    
  • Use add method.
    
    
  • Correlations between relations
    
    
    

    SIMILARITIES
    --------------------------------------------------------------------------------
    
    Measure:                                CORRELATION
    Variables are:                          MATRICES
    Diagonal:                               TREATED AS MISSING
    Input dataset:                          sampson (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch05\Sampson-Samplk\sampson)
    Similarity matrix:                      sampson-Sim (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch05\Sampson-Samplk\sampson-Sim)
    
    
                      1      2      3      4      5      6      7      8      9     10
                 SAMPLK SAMPLK SAMPLK SAMPDL SAMPES SAMPDE SAMPIN SAMPNI SAMPPR SAMPNP
                 ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
      1 SAMPLK1   1.000  0.645  0.638 -0.103  0.510 -0.146  0.530 -0.149  0.384 -0.011
      2 SAMPLK2   0.645  1.000  0.768 -0.149  0.633 -0.167  0.616 -0.166  0.508 -0.087
      3 SAMPLK3   0.638  0.768  1.000 -0.168  0.653 -0.176  0.651 -0.165  0.554 -0.055
      4 SAMPDLK  -0.103 -0.149 -0.168  1.000 -0.145  0.689 -0.113  0.522 -0.136  0.340
      5  SAMPES   0.510  0.633  0.653 -0.145  1.000 -0.186  0.797 -0.171  0.629 -0.141
      6 SAMPDES  -0.146 -0.167 -0.176  0.689 -0.186  1.000 -0.142  0.660 -0.148  0.425
      7  SAMPIN   0.530  0.616  0.651 -0.113  0.797 -0.142  1.000 -0.170  0.642 -0.151
      8 SAMPNIN  -0.149 -0.166 -0.165  0.522 -0.171  0.660 -0.170  1.000 -0.141  0.355
      9  SAMPPR   0.384  0.508  0.554 -0.136  0.629 -0.148  0.642 -0.141  1.000 -0.125
     10 SAMPNPR  -0.011 -0.087 -0.055  0.340 -0.141  0.425 -0.151  0.355 -0.125  1.000
    
    Cronbach's Alpha = 0.710
    
    Similarity matrix saved as dataset sampson-Sim
    
    ----------------------------------------
    Running time:  00:00:01
    Output generated:  29 9 14 17:38:15
    UCINET 6.511 Copyright (c) 1992-2012 Analytic Technologies
    

    
    
  • Then do factor analysis
    
    
    

    FACTOR ANALYSIS
    --------------------------------------------------------------------------------
    
    Method of extraction:                   PRINCIPAL COMPONENTS
    Method of rotation:                     VARIMAX
    Minimum eigenvalue to retain:           1
    Data type:                              1-mode correlation matrix
    Input dataset:                          sampson-Sim (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch05\Sampson-Samplk\sampson-Sim)
    
    
    EIGENVALUES
    
      FACTOR    VALUE PERCENT   CUM %   RATIO
     ------- -------- ------- ------- -------
          1:  4.33185    43.3    43.3   1.913
          2:  2.26442    22.6    66.0   2.722
          3:  0.83177     8.3    74.3   1.310
          4:  0.63496     6.3    80.6   1.301
          5:  0.48813     4.9    85.5   1.249
          6:  0.39076     3.9    89.4   1.067
          7:  0.36615     3.7    93.1   1.356
          8:  0.26992     2.7    95.8   1.181
          9:  0.22847     2.3    98.1   1.180
         10:  0.19356     1.9   100.0
     ======= ======== ======= ======= =======
             10.00000   100.0
    
    
    Unrotated Factor Loadings
    
                         1      2
                    ------ ------
        1  SAMPLK1   0.710  0.231
        2  SAMPLK2   0.817  0.227
        3  SAMPLK3   0.836  0.235
        4  SAMPDLK  -0.373  0.732
        5   SAMPES   0.835  0.206
        6  SAMPDES  -0.422  0.790
        7   SAMPIN   0.830  0.233
        8  SAMPNIN  -0.400  0.706
        9   SAMPPR   0.724  0.179
       10  SAMPNPR  -0.272  0.564
    
    
    Rotated Factor Loadings
    
                         1      2
                    ------ ------
        1  SAMPLK1   0.745 -0.046
        2  SAMPLK2   0.844 -0.089
        3  SAMPLK3   0.864 -0.088
        4  SAMPDLK  -0.078  0.818
        5   SAMPES   0.852 -0.115
        6  SAMPDES  -0.102  0.889
        7   SAMPIN   0.858 -0.088
        8  SAMPNIN  -0.113  0.804
        9   SAMPPR   0.739 -0.099
       10  SAMPNPR  -0.046  0.624
    
    Eigenvalues saved as dataset            Eigen (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch05\Sampson-Samplk\Eigen)
    Factor loadings saved as dataset        Loadings (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch05\Sampson-Samplk\Loadings)
    
    ----------------------------------------
    Running time:  00:00:01
    Output generated:  29 9 14 17:37:04
    UCINET 6.511 Copyright (c) 1992-2012 Analytic Technologies
    

    
    

    Rotated Factor Loadings
    
                         1      2
                    ------ ------
        3  SAMPLK3   0.864 -0.088
        7   SAMPIN   0.858 -0.088
        5   SAMPES   0.852 -0.115
        2  SAMPLK2   0.844 -0.089
        1  SAMPLK1   0.745 -0.046
        9   SAMPPR   0.739 -0.099
        -------------------------
       10  SAMPNPR  -0.046  0.624
        4  SAMPDLK  -0.078  0.818
        6  SAMPDES  -0.102  0.889
        8  SAMPNIN  -0.113  0.804
    

• Aggregating them accordingly (two different matrix, positive and negative). Then analyze them accordingly.
  
• . . . .
  

• who collaborates with whom in an organization, but what we are really interested in is the pattern of ties between departments . . .
  
• But, how?????
  

• We do not want to analyze the whole network.
  

• 'row-dominated locally aggregated structure' (RLAS)
  
  • TRUST network
    
  • (i, i, j) i's as in a row word taken as truth.
    
• Intersection method (Intersected locally aggregated structure (ILAS)
  
  • (i, i, j) =\= (j, i, j)
    
  • (i, i, j) = 1 | (j, i, j) = 1 ==> (i, j) = 1
    
• Consensus type
  
  • R = (k, i, j) = ? When, a certain portion agrees, R has a value.
    
    

1. Represent one- and two-mode data in a two-dimensional map
   
2. Cluster data into groups using hierarchical clustering
   
3. Correctly interpret the information contained in the clusters and maps
   

METRIC MULTIDIMENSIONAL SCALING
--------------------------------------------------------------------------------

Starting config:                        GOWER'S PRINCIPAL COORDINATES

Type of Data:                           Dissimilarities
Input dataset:                          CITIES (D:\Users\Hyo\Cs-Ds\CS\Classes\2014-fall\NewMediaStudies-grad\Borgatti_Data\Ch06\CITIES)

9 items
Initial Stress = 0.019696305
Final Stress = 0.014 after 11 iterations.


Metric MDS coordinates (stress = 0.014)

                     1      2
                ------ ------
    1   BOSTON   0.410 -0.144
    2       NY   0.367 -0.095
    3       DC   0.331 -0.031
    4    MIAMI   0.374  0.304
    5  CHICAGO   0.132 -0.039
    6  SEATTLE  -0.486 -0.191
    7       SF  -0.520  0.056
    8       LA  -0.450  0.156
    9   DENVER  -0.160  0.016

Coordinates saved as dataset MetricMdsCoord

----------------------------------------
Running time:  00:00:01
Output generated:  23 9 14 15:34:48
UCINET 6.511 Copyright (c) 1992-2012 Analytic Technologies

• distortion = stress (not fitting into the two-dimensional space)
  
• 0.2 = acceptance level for metric
  
• 0.12 = for non-metric (ordinal)
  
• the above case = .014
  
  The first is that the axes are, in themselves, meaningless, and the second is that the orientation of the picture is arbitrary.
  
  . . . two things to look for in interpreting an MDS picture: clusters and dimensions.
  

1. Start by assigning each item to its own cluster, so that if you have N items, you now have N clusters, each containing just one item. Let the distances between the clusters equal the distances between the items they contain.
   
2. Find the closest pair of clusters and merge them into a single cluster, so that now you have one less cluster.
   
3. Compute distances between the new cluster and each of the old clusters.'''
   
4. Repeat steps 2 and 3 until all items are clustered into a single cluster of size N.
   
   
   Step 3 can be done in different ways, which is what distinguishes
   
5. single-link from
   
6. complete-link (dia-meter, maximum) and
   
7. average-link clustering.
   

1. In single-link clustering (also called the nearest neighbor, the connectedness method and the minimum method), we consider the distance between one cluster and another cluster to be equal to the shortest distance from any member of one cluster to any member of the other cluster.
   
2. In complete-link clustering (also called the diameter or maximum method), we consider the distance between one cluster and another cluster to be equal to the longest distance from any member of one cluster to any member of the other cluster.
   
3. In average-link clustering, we consider the distance between one cluster and another cluster to be equal to the average distance from any member of one cluster to any member of the other cluster. More recently, Newman (2004) has introduced another criterion, which joins the pair of clusters that would maximize Q modularity, a measure of clustering quality.
   

QAP CORRELATION
--------------------------------------------------------------------------------

Data Matrices:                          PADGB
                                        PADGM
# of Permutations:                      5000
Random seed:                            1
Method:                                 Fast: no missing values allowed


QAP results for PADGM * PADGB (5000 permutations)

                                 1         2         3         4         5         6         7         8
                         Obs Value Significa   Average   Std Dev   Minimum   Maximum Prop >= O Prop <= O
                         --------- --------- --------- --------- --------- --------- --------- ---------
    Pearson Correlation     0.3719    0.0008   -0.0004    0.0935   -0.1690    0.4395    0.0008    0.9998



QAP Correlations

               1     2
           PADGB PADGM
           ----- -----
  1 PADGB  1.000 0.372
  2 PADGM  0.372 1.000


QAP P-Values

               1     2
           PADGB PADGM
           ----- -----
  1 PADGB  0.000 0.001
  2 PADGM  0.001 0.000

QAP statistics saved as datafile QAP Correlation Results

----------------------------------------
Running time:  00:00:01
Output generated:  07 10 14 15:15:41
UCINET 6.511 Copyright (c) 1992-2012 Analytic Technologies

DL
N=17
FORMAT = FULLMATRIX DIAGONAL PRESENT
LEVEL LABELS:
"NEWC0"
DATA:
 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 1
 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 1 0
 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1
 0 1 0 0 0 0 1 0 0 0 0 0 0 0 0 0 1
 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1
 0 0 0 1 0 0 0 1 0 0 0 0 1 0 0 0 0
 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 1
 0 0 0 0 0 1 0 0 0 1 1 0 0 0 0 0 0
 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1
 1 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0
 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1
 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 0 1
 1 0 0 0 0 1 0 0 0 0 0 0 0 0 1 0 0
 0 0 0 0 0 0 1 0 1 1 0 0 0 0 0 0 0
 0 0 0 0 1 0 0 0 0 1 1 0 0 0 0 0 0
 0 0 0 1 0 0 0 0 1 0 1 0 0 0 0 0 0
 0 0 0 1 0 0 0 0 1 0 0 1 0 0 0 0 0

Tools -> Similarities & Distances 

input: davis
output: davis-Mat-C

Matches . . . . Cols

input: davis
output: davis-Mat-R

Matches . . . . Rows

Network -> Cohesion -> Density with Group 
input: davis
row partition: davis-Mat-R
columns partion: davis-Mat-C