Week-Feb 15

[vivekaxl]

DOING:

• Work on rank based prediction paper

DONE:

• Collected data for 2 scenarios of CLOUDSIM -- The performance can be predicted with low MMRE
• Band Generation for CLOUDSIM -- The performance can be predicted with low MMRE
• Band Generation for other SE models -- The performance can be predicted with low MMRE
• Band Generation for Artificial Model -- Useful for our rank-based technique
• Rank-based diagrams for Artificial Model

TODO:

• Try it on multi-objective datasets. How to extend the rank-based technique to a multi-objective setting
• Set up PRISM

roadblocks:

None

admin:

None

[timm]

the thing that makes this publishable is a demonstration that it steps up from other work. e.g.

• previously all prior studies were limited by A, which we can remove.
• previously, they explored X but found it too hard to collect data
• previously they explored scenario Y yet with out new fast collection methods, we can explore Y1,Y2... Y20 so we can be more through than before
• or...

but some statement that shows demonstrable improvement from the current lits