This shows you the differences between two versions of the page.

Both sides previous revision Previous revision Next revision | Previous revision | ||

cs190c:project4c_09 [2009/04/21 14:40] tang |
cs190c:project4c_09 [2009/04/21 15:19] (current) |
||
---|---|---|---|

Line 25: | Line 25: | ||

A good graph clustering will result in clusters of proteins with high values of similarity metrics within a cluster (which means values based on the good edges should be large) and low values between the clusters (which means the values based on bad edges should be small). | A good graph clustering will result in clusters of proteins with high values of similarity metrics within a cluster (which means values based on the good edges should be large) and low values between the clusters (which means the values based on bad edges should be small). | ||

+ | |||

====== Submission instructions ====== | ====== Submission instructions ====== | ||

- | The page we [[provide]] for sample graphs also contains four matrices, two for synthetic data and two for the biological data sets. For a graph having associated matrices, use function ''import_matrix'' to create each graph GW, use function ''mcl_clustering'' to determine a clustering of the nodes in graph GC, and use graph GW to evaluate the clusters. **Note: If mcl_clustering is taking too long to run on your computer, the clustering is provided for you on the page with the graphs.** | + | The page we [[provide]] for sample graphs also contains four matrices, two for synthetic data and two for the biological data sets. For a graph having associated matrices, use function ''import_matrix'' to create each graph GW, use function ''mcl_clustering'' to determine a clustering of the nodes in graph GC, and use graph GW to evaluate the clusters. |

+ | | ||

+ | **Note: If mcl_clustering is taking too long to run on your computer, the clustering is provided for you on the page with the graphs.** | ||

Summarizing what is to be done for each data set: | Summarizing what is to be done for each data set: |