Mohan, could you check when you can? On Fri, Feb 27, 2015 at 6:46 AM, Ginnaliya Gamathige, Lahiru Manananda < [email protected]> wrote:
> Hi Srinath, > > I have wrapped up the implementation and sent a pull request. Please let > me know if there are any issues with merging the pull request. > > I have attached the draft version of the report I wrote based on the > current implementation, this contains more improvements we should do to > support fully for the Hoeffding algorithm. > > As per our discussion happened in Bloomington, I had a look in to > regression analysis for data streams and there are lots of interesting > things we can do with Siddhi to do multi-dimensional regressions. > Mostly people tried implementing one parse algorithms to find patterns > with sensor data and dynamically modify the model. I hope to send a detail > email on this. > > Thanks > Lahiru > > > On Sep 23, 2014, at 3:39 AM, Ginnaliya Gamathige, Lahiru Manananda < > [email protected]> wrote: > > Hi Suho, > > I have modified Attribute class which will keep all the possible values > in attribute and can create the numeric value 2 for “rainy”. In the > incoming event we can have the possible values for each attribute. I have > the code committed to my local github[1] and once I have a working copy > with some good tests I can send a pull request. > > Once we have the tree build for these events, Siddhi should be able to > take a given event and do a prediction and determine the value of the last > attribute and based on the last attribute values we > can move event in to different event streams. > Ex: We can send 10 events with proper value for play (yes,no) after you > have the tree built you can just send the rest of the attributes in the > event and Decision tree will predict the value for play (yes or no), so the > query can be written to the play event. In the Window implementation we can > specify minimum events required to build the tree and always keep the last > attribute as the class attribute (During the initial implementation). > > So query will be, from > weather#siddhiht(minimum_events_for_tree_build,give_desired_class_values) > select * insert into play for all-events; > > For a simple scenario we can assume there is only a single class we need > to check in this case we can say play=“yes”, so query can look like below > for 10 minimum events. > > weather#siddhiht(10,play="yes") > select * insert into play for all-events; > > If events come empty play value before the tree is built we can throw an > exception or just emit the decision but it will be with low accuracy. > > Regards > Lahiru > > > [1]https://github.com/glahiru/siddhi > On Sep 23, 2014, at 1:56 AM, Sriskandarajah Suhothayan <[email protected]> > wrote: > > Looks OK. > > Can you please clarify how would you write queries in this case? (e.g If > its "rainy" how would you check?) > and internally how all of these will be processed? > > Regards > Suho > > > On Tue, Sep 23, 2014 at 10:30 AM, Srinath Perera <[email protected]> wrote: > > I think this is OK. Suho, WDYT? > > On Mon, Sep 22, 2014 at 12:15 AM, Ginnaliya Gamathige, Lahiru Manananda < > [email protected]> wrote: > > Hi Devs, > > I am developing a classification algorithm for siddhi and I have a > requirement of defining attributes as a nominal attribute. > > Nominal attribute is similar to a string attribute but during the > definition we have to give the possible values for that attribute. > One of the value from that set can come as the actual value in a > particular event. I can see in the code Siddhi doesn’t support this (Please > correct me if I am wrong), so > I am thinking of changing the Siddhi language grammar to parse nominal > attribute as below. > > Example query: define stream weather(outlook nominal{sunny, overcast, > rainy}, temperature nominal{hot, mild, cool}, humidity nominal{high, > normal},windy nominal{TRUE, FALSE},play nominal{yes, no}). > > If there is a better way to do this, please provide me some feedback. > > During the classification algorithm it will use numeric values which > maps to these possible values(Ex: sunny=0,overcase=1,rainy=2) and do the > calculation using numeric values. Initial phase of the algorithm > I am implementing[1], I hope to process nominal values (which is easy to > implement), then algorithm should support processing numeric data too. > > If you have any feedback please let me know. > > [1]http://homes.cs.washington.edu/~pedrod/papers/kdd00.pdf > > Regards > Lahiru > > _______________________________________________ > Architecture mailing list > [email protected] > https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture > > > > > -- > ============================ > Srinath Perera, Ph.D. > http://people.apache.org/~hemapani/ > http://srinathsview.blogspot.com/ > > > > > -- > > *S. Suhothayan * > Technical Lead & Team Lead of WSO2 Complex Event Processor > *WSO2 Inc. *http://wso2.com > * <http://wso2.com/>* > lean . enterprise . middleware > > > *cell: (+94) 779 756 757 <%28%2B94%29%20779%20756%20757> | blog: > http://suhothayan.blogspot.com/ <http://suhothayan.blogspot.com/> twitter: > http://twitter.com/suhothayan <http://twitter.com/suhothayan> | linked-in: > http://lk.linkedin.com/in/suhothayan <http://lk.linkedin.com/in/suhothayan>* > _______________________________________________ > Architecture mailing list > [email protected] > https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture > > > _______________________________________________ > Architecture mailing list > [email protected] > https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture > > > > _______________________________________________ > Architecture mailing list > [email protected] > https://mail.wso2.org/cgi-bin/mailman/listinfo/architecture > > -- ============================ 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
