Suho Sumedha, this is a highly relevant IoT scenario and use case would
work on most time series equipment data with fault details.
Need to figure out how and where to use it in IoT story.
On Mon, Sep 19, 2016 at 6:05 PM, Supun Sethunga <sup...@wso2.com> wrote:
> What was the response variable and classes, in the classification scenario?
> On Mon, Sep 19, 2016 at 2:19 PM, Roshan Alwis <rosh...@wso2.com> wrote:
>> Hi all,
>> I am an intern who is currently working on the project "Use machine
>> learning for predictive maintenance", which is a use case of IOT. I have
>> been working with the "Turbofan Engine Degradation Simulation Data Set"
>> published by NASA as my core data set. The goal of this experiment is to
>> predict the remaining useful lifetime of turbofan engines based on their
>> sensor readings. I have built both classification and regression models for
>> this scenario.
>> Link to the data set.
>> The algorithms that I have used and their test results are,
>> *Random Forest Classification* [Classification]
>> Result : https://docs.google.com/a/wso2.com/document/d/1EiPbEA5cq4823
>> *Random Forest Regression* [Regression]
>> Result : https://docs.google.com/a/wso2.com/document/d/1HKW84QyJyAwK6
>> *H2O Deep Learning* [Regression]
>> Result : https://docs.google.com/a/wso2.com/document/d/1SncvDlVsxsSfU
>> *These results are only valid for following data sets,*
>> Training : train_FD001.txt
>> Testing : test_FD001.txt
>> Validating : RUL_FD001.txt
>> If you have any suggestions for improvements, please let me know. For
>> further information, contact through the email.
>> Best Regards,
>> Roshan Alwis
>> *Trainee Software Engineer*
>> m: 0715894672
>> a: No 3/56, Aluthgama Rd, Elpitiya.
>> e: rosh...@wso2.com
>> <https://web.facebook.com/alwisroshan> <https://twitter.com/rm_alwis>
>> Architecture mailing list
> *Supun Sethunga*
> Senior Software Engineer
> WSO2, Inc.
> lean | enterprise | middleware
> Mobile : +94 716546324
> Blog: http://supunsetunga.blogspot.com
> Architecture mailing list
Srinath Perera, Ph.D.
Architecture mailing list