Hi sanjiva, Sorry forgot to update that part. Will change and send.
--srinath On 19 Aug 2014 12:36, "Sanjiva Weerawarana" <[email protected]> wrote: > Srinath why did u put ML Wizard on top of BAM? I see ML as a dark colored > product box like CEP, BAM etc. and the wizard is just part of that. A more > detailed picture could show the different Hadoop clusters too but its not > just Hadoop of course. > > > On Tue, Aug 19, 2014 at 8:48 AM, Srinath Perera <[email protected]> wrote: > >> Sanjva and myself chatted about how BAM/CEP etc fit within the ML work we >> are doing. Following picture shows out thoughts. >> >> [image: MLArchitecture.png] >> >> >> 1. Data may be received via Data receiver or already in storage >> 2. We use ML algos to create models from that data (via ML Wizard we >> are discussing in this thread) >> 3. Those models are stored in registry/ store and can be used from >> CEP or BAM for decision processes. (Model is a decision function) >> 4. Models may also used within business logic via ESB or integrated >> within products (e.g. IS to detect suspicious logins etc). Following are >> some examples. >> - >> - Operators in BAM and CEP >> - Recommendations in Store >> - DOS attack avoidance >> - Throttling >> - Autoscaling >> - Time Series tool box >> - Fraud toolbox >> - IS - account alerts, suspicious activity >> >> >> >> On Fri, Aug 15, 2014 at 10:20 PM, Upul Bandara <[email protected]> wrote: >> >>> >>> wso2_ml_workflow_version_2.pdf >>> <https://docs.google.com/a/wso2.com/file/d/0B1KWJ16Pv0V0U3BvcVZuZUR4WE0/edit?usp=drive_web> >>> >>> Attached the document. >>> >>> Regards, >>> Upul >>> >>> >>> On Fri, Aug 15, 2014 at 9:22 PM, Srinath Perera <[email protected]> >>> wrote: >>> >>>> Upul attach the doc. >>>> >>>> Sanjiva, I will schedule. >>>> On 15 Aug 2014 18:22, "Sanjiva Weerawarana" <[email protected]> wrote: >>>> >>>>> Guys this doc is not readable? >>>>> >>>>> I'd like to brainstorm the overall model of how ML fits to BAM/CEP >>>>> stuff - can we schedule another discussion? >>>>> >>>>> Sanjiva. >>>>> >>>>> >>>>> On Fri, Aug 15, 2014 at 11:39 AM, Srinath Perera <[email protected]> >>>>> wrote: >>>>> >>>>>> pls add arch@ >>>>>> >>>>>> >>>>>> On Fri, Aug 15, 2014 at 11:37 AM, Upul Bandara <[email protected]> wrote: >>>>>> >>>>>>> Hi Srinath, >>>>>>> >>>>>>> Attached, please find the second version of the WSO2 ML workflow. >>>>>>> >>>>>>> >>>>>>> wso2_ml_workflow_version_2.pdf >>>>>>> <https://docs.google.com/a/wso2.com/file/d/0B1KWJ16Pv0V0U3BvcVZuZUR4WE0/edit?usp=drive_web> >>>>>>> >>>>>>> Regards, >>>>>>> Upul >>>>>>> >>>>>>> >>>>>>> On Wed, Aug 13, 2014 at 12:04 PM, Supun Sethunga <[email protected]> >>>>>>> wrote: >>>>>>> >>>>>>>> Hi Anuruddha, >>>>>>>> >>>>>>>> Usually normalization is just a one technique used for data >>>>>>>> transformations in "Preprocessing" step. (Here we have referred to it >>>>>>>> as >>>>>>>> "transformations"). >>>>>>>> >>>>>>>> And for model selection, for now, its going to be user dependent. >>>>>>>> Once we moved to Deep Learning, this can have the option of >>>>>>>> automating, to >>>>>>>> find the best algorithm. >>>>>>>> >>>>>>>> About the name, eventhough the training of the model is supervised, >>>>>>>> here we use some unsupervised techniques as well (e.g. PCA for feature >>>>>>>> selection), thus it would be better to keep it general :) >>>>>>>> >>>>>>>> Thanks, >>>>>>>> Supun >>>>>>>> >>>>>>>> >>>>>>>> On Wed, Aug 13, 2014 at 11:35 AM, Anuruddha Premalal < >>>>>>>> [email protected]> wrote: >>>>>>>> >>>>>>>>> Hi Supun, >>>>>>>>> >>>>>>>>> In a general supervised ML work flow we have to perform feature >>>>>>>>> normalization [1] (This stage is after the feature extraction stage). >>>>>>>>> This >>>>>>>>> step is very important to reduce the convergence time of the training >>>>>>>>> algorithm. >>>>>>>>> >>>>>>>>> How is the model selection going to happen in this approach? is it >>>>>>>>> a user input or is it based on a modal selection algorithm? >>>>>>>>> >>>>>>>>> Since this doesn't cover the unsupervised learning isn't it good >>>>>>>>> to name this as supervised ML workflow? >>>>>>>>> >>>>>>>>> [1] Feature scaling : http://en.wikipedia.org/wiki/Feature_scaling >>>>>>>>> >>>>>>>>> Regards, >>>>>>>>> Anuruddha. >>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> On Tue, Aug 12, 2014 at 8:55 PM, Supun Sethunga <[email protected]> >>>>>>>>> wrote: >>>>>>>>> >>>>>>>>>> [adding Lochana] >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> On Tue, Aug 12, 2014 at 6:05 PM, Supun Sethunga <[email protected]> >>>>>>>>>> wrote: >>>>>>>>>> >>>>>>>>>>> Hi Srinath, >>>>>>>>>>> >>>>>>>>>>> Attached is a proposed work-flow for the ML design came up by >>>>>>>>>>> Lochana, Upul and me. >>>>>>>>>>> >>>>>>>>>>> Appreciate any comments. >>>>>>>>>>> >>>>>>>>>>> Regards, >>>>>>>>>>> Supun >>>>>>>>>>> >>>>>>>>>>> -- >>>>>>>>>>> *Supun Sethunga* >>>>>>>>>>> Software Engineer >>>>>>>>>>> WSO2, Inc. >>>>>>>>>>> lean | enterprise | middleware >>>>>>>>>>> Mobile : +94 716546324 >>>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>>> -- >>>>>>>>>> *Supun Sethunga* >>>>>>>>>> Software Engineer >>>>>>>>>> WSO2, Inc. >>>>>>>>>> lean | enterprise | middleware >>>>>>>>>> Mobile : +94 716546324 >>>>>>>>>> >>>>>>>>>> _______________________________________________ >>>>>>>>>> Architecture mailing list >>>>>>>>>> [email protected] >>>>>>>>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture >>>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>>> >>>>>>>>> -- >>>>>>>>> *Anuruddha Premalal* >>>>>>>>> Software Eng. | WSO2 Inc. >>>>>>>>> Mobile : +94710461070 >>>>>>>>> Web site : www.regilandvalley.com >>>>>>>>> >>>>>>>>> >>>>>>>>> _______________________________________________ >>>>>>>>> Architecture mailing list >>>>>>>>> [email protected] >>>>>>>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture >>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> -- >>>>>>>> *Supun Sethunga* >>>>>>>> Software Engineer >>>>>>>> WSO2, Inc. >>>>>>>> lean | enterprise | middleware >>>>>>>> Mobile : +94 716546324 >>>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> -- >>>>>>> Upul Bandara, >>>>>>> Associate Technical Lead, WSO2, Inc., >>>>>>> Mob: +94 715 468 345. >>>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> ============================ >>>>>> Director, Research, WSO2 Inc. >>>>>> Visiting Faculty, University of Moratuwa >>>>>> Member, Apache Software Foundation >>>>>> Research Scientist, Lanka Software Foundation >>>>>> Blog: http://srinathsview.blogspot.com twitter:@srinath_perera >>>>>> Site: http://people.apache.org/~hemapani/ >>>>>> Photos: http://www.flickr.com/photos/hemapani/ >>>>>> Phone: 0772360902 >>>>>> >>>>>> _______________________________________________ >>>>>> Architecture mailing list >>>>>> [email protected] >>>>>> https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture >>>>>> >>>>>> >>>>> >>>>> >>>>> -- >>>>> Sanjiva Weerawarana, Ph.D. >>>>> Founder, Chairman & CEO; WSO2, Inc.; http://wso2.com/ >>>>> email: [email protected]; office: (+1 650 745 4499 | +94 11 214 5345) >>>>> x5700; cell: +94 77 787 6880 | +1 408 466 5099; voip: +1 650 265 8311 >>>>> blog: http://sanjiva.weerawarana.org/; twitter: @sanjiva >>>>> Lean . Enterprise . Middleware >>>>> >>>> >>> >>> >>> -- >>> Upul Bandara, >>> Associate Technical Lead, WSO2, Inc., >>> Mob: +94 715 468 345. >>> >> >> >> >> -- >> ============================ >> Director, Research, WSO2 Inc. >> Visiting Faculty, University of Moratuwa >> Member, Apache Software Foundation >> Research Scientist, Lanka Software Foundation >> Blog: http://srinathsview.blogspot.com twitter:@srinath_perera >> Site: http://people.apache.org/~hemapani/ >> Photos: http://www.flickr.com/photos/hemapani/ >> Phone: 0772360902 >> > > > > -- > Sanjiva Weerawarana, Ph.D. > Founder, Chairman & CEO; WSO2, Inc.; http://wso2.com/ > email: [email protected]; office: (+1 650 745 4499 | +94 11 214 5345) > x5700; cell: +94 77 787 6880 | +1 408 466 5099; voip: +1 650 265 8311 > blog: http://sanjiva.weerawarana.org/; twitter: @sanjiva > Lean . Enterprise . Middleware >
_______________________________________________ Architecture mailing list [email protected] https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture
