Evaluation of Algorithms using Weka

I am looking for somebody with a strong background in machine learning to assist us in our project. We have built a small simulation program that generates us .arff files that should be used for predictive analytics. Your task would be to: 1. define criteria by which to evaluate different machine learning algorithms. this is dependent upon the context of our project. more information about it upon request 2. select ~5 different machine learning algorithms to evaluate 3. generate the necessary data sets using our simulation (English manual will be prepared and Java source code is available) 4. evaluate the performance of the selected algorithms according to the criteria you defined 5. summarize the work in written form. no need for long scientific text, but we need to be able to understand: a) which criteria did you choose and why in the context of our project b) which algorithms did you select and why c) how did you do the performance evaluation. here a lot of details are necessary. e.g. did you use n-fold cross validation, if so, why? d) what are the results, e.g. which algorithm is best in which scenario, what factors can be used to improve the performance, etc. Your notes need to be sufficient for us to base around 20 pages of semi-scientific paper on such as this: [login to view URL] The evaluation has to be done with Weka 3 and your final report must include Weka-generated console exports and graphics! Ideally you can start within the next 7 days and need no longer than 10 days for this. Final note: this is not deep scientific research. If you have a good background in data mining and are familiar with how evaluations works, this should be a pretty simple task. For any further question, don't hesitate to contact me.