Hi Randika, Thank you for showing interest for this project.
I've checked the SPMF library and what this library supports is sequential pattern mining which is quite different from machine learning algorithms used in WSO2 ML. What this project intends to achieve is to leverage the existing algorithms to support streaming data. As an initiative, first you can get an idea about the architecture of WSO2 ML[1]. CEP event streams[2] / publishers[3] maybe used for feeding data streams in to ML. Since ML is using Apache Spark mllib[4] for its' algorithms, you might want to read about that. To get an idea about an architecture, try to understand how Spark streaming[5] (see examples) handles input data streams. Also, have a look in the streaming algorithms[6][7] supported. In order to use these algorithms, you may have to use Scala APIs(Since Spark does not have Java implementations yet). There are two approaches indicated in the project proposals page. These streaming algorithms can be directly used in the first approach. For the other approach, the architecture should contain a procedure to create mini batches from streaming data with relevant sizes (i.e. a moving window) and do periodic retraining of the same algorithm. BTW, watching the video referenced in the proposal (reference: 5) will help you getting a better idea about machine learning algorithms with streaming data. Let us know if you need any help with these. Best regards [1] https://docs.wso2.com/display/ML110/Architecture [2] https://docs.wso2.com/display/CEP400/Understanding+Event+Streams [3] https://docs.wso2.com/display/CEP400/HTTP+Event+Publisher [4] https://spark.apache.org/docs/1.4.1/mllib-guide.html [5] https://spark.apache.org/docs/1.4.1/streaming-programming-guide.html [6] https://spark.apache.org/docs/1.4.1/mllib-linear-methods.html#streaming-linear-regression [7] https://spark.apache.org/docs/1.4.1/mllib-clustering.html#streaming-k-means On Thu, Feb 25, 2016 at 10:40 AM, Randika Navagamuwa < randika...@cse.mrt.ac.lk> wrote: > Hi, > I'm a 3rd year undergraduate from Department of Computer Science and > Engineering, University of Moratuwa. I went through the project proposals > and I want to clarify some things regarding this project. > > - I've seen two approaches are mentioned, but other than those two > methods can the objectives be achieved using this approach > - SPMF[1] library can be used for pattern analysis. > - Then if a data set has a same pattern as a previously modeled > data set same algorithm can be used. > > According to the deliverables, first step is to come with an architecture. > Is there any online material to refer before starting this project. > > [1]http://www.philippe-fournier-viger.com/spmf/ > > > *Best Regards* > > *Randika Navagamuwa,* > > *Department of Computer Science & Engineering,* > > *University of Moratuwa,* > *Sri Lanka.* > > *www.rnavagamuwa.com <http://www.rnavagamuwa.com>*[image: > lk.linkedin.com/in/rnavagamuwa/] <http://lk.linkedin.com/in/rnavagamuwa/> > [image: > https://www.facebook.com/rnavagamuwa] > <https://www.facebook.com/rnavagamuwa> [image: > https://twitter.com/rnavagamuwa] <https://twitter.com/rnavagamuwa> [image: > https://plus.google.com/+RandikaNavagamuwa/] > <https://plus.google.com/+RandikaNavagamuwa/> > -- Pruthuvi Maheshakya Wijewardena mahesha...@wso2.com +94711228855
_______________________________________________ Dev mailing list Dev@wso2.org http://wso2.org/cgi-bin/mailman/listinfo/dev
