Hi,

I'm working on the project proposal and there were several questions
emerged as follows,

   1. In offline machine learning we normally preprocess data by doing
   operations such as *scaling*, before training the model. So in the
   streaming algorithm is there any way to do that?

   2. Does this algorithms we are going to implement should support *non
   numeric* input such as text data?

   3. If it's so,  do we need to implement a *word2vec* model to convert
   inputs into numeric vectors?

   4. If our streaming algorithms supports sentiment analysis on the Social
   Networks,  how are we going to address this problem with streaming data?
   I researched about this matter and found some approach such as *Modified
   Balanced Winnow *for streaming algorithms instead of batch mode learning
   approach.
   http://file.scirp.org/pdf/JDAIP_2014053015541008.pdf

I appreciate your opinion about this problems and it would be very helpful
in creating the proposal for me.

Thanks and Regards,

*Kasun Bandara*
Faculty of Information Technology, University of Moratuwa
0715787004 | kasun...@gmail.com
<http://www.linkedin.com/in/kasunweerakoon>
<http://github.com/KasunWeerakoon>  <http://devkasun.blogspot.com/>

On Thu, Mar 23, 2017 at 11:19 AM, Kasun Weerakoon <kasun...@gmail.com>
wrote:

> Hi,
>
> I started working on the project proposal as you mentioned and came up
> with a sample perceptron classifier algorithm.
>
> Please find the github repo in following link - https://github.com/
> KasunWeerakoon/GSoC_WSO2
>
> It would be great if you can provide me some feedback about it.
>
> Thanks and regards,
>
> On Wed, 15 Mar 2017, 07:25 Upul Bandara, <u...@wso2.com> wrote:
>
>> Hi Kasun,
>>
>> Sounds good. Also, please start working on the project proposal as well.
>> The proposal should be included following high-level sections and please
>> feel free to add sub-sections as needed.
>>
>> 1. Description
>> 2. Deliverables
>> 3. Timeline
>> 4. Implementation
>>
>>
>> Thanks,
>> Upul
>>
>> On Tue, Mar 14, 2017 at 9:54 AM, Kasun Weerakoon <kasun...@gmail.com>
>> wrote:
>>
>> Hi,
>>
>> Sorry for being late to communicate my progress with you, because of my
>> academic works. From now on I will be able to work on this project with my
>> full potential.
>>
>> I went through the CEP quick start guide and created  event flow,
>> received events via HTTP transport, processed events with an execution plan
>> and published events to dashboard successfully.
>>
>> I also went through the Apache SAMOA getting start guide and did some
>> tutorials.
>>
>> I'm currently reading Chapter 12 of the Mining of Massive Datasets book
>> as you mentioned and hope your guidance to carry out the work.
>>
>> Thanks and Regards,
>>
>> *Kasun Bandara*
>> Faculty of Information Technology, University of Moratuwa
>> 0715787004 | kasun...@gmail.com
>> <http://www.linkedin.com/in/kasunweerakoon>
>> <http://github.com/KasunWeerakoon>  <http://devkasun.blogspot.com/>
>>
>> On Tue, Feb 28, 2017 at 12:06 PM, Kasun Weerakoon <kasun...@gmail.com>
>> wrote:
>>
>> Hi,
>>
>> According to the rules mentioned,  I think I'm eligible to the program.
>>
>> Thanks and Regards,
>>
>> *Kasun Bandara*
>> Faculty of Information Technology, University of Moratuwa
>> 0715787004 | kasun...@gmail.com
>> <http://www.linkedin.com/in/kasunweerakoon>
>> <http://github.com/KasunWeerakoon>  <http://devkasun.blogspot.com/>
>>
>> On Tue, Feb 28, 2017 at 10:59 AM, Upul Bandara <u...@wso2.com> wrote:
>>
>> Hi Kasun,
>>
>> Great!
>> Also, please verify your eligibility as per the GSoC rules:
>> https://summerofcode.withgoogle.com/rules/
>>
>> Thanks,
>> Upul
>>
>> On Tue, Feb 21, 2017 at 9:52 AM, Kasun Weerakoon <kasun...@gmail.com>
>> wrote:
>>
>> Hi Upul,
>>
>> Thanks you for your reply. I'll go through above mentioned guides and get
>> back to you later.
>>
>> Regards,
>> Kasun Weerakoon.
>>
>> On Tue, 21 Feb 2017, 08:39 Upul Bandara, <u...@wso2.com> wrote:
>>
>> Hi Kasun,
>>
>> We are glad to see your interest!
>>
>> The main objective of this project is to develop a real-time machine
>> learning toolkit for Siddhi [1]. Siddhi CEP is a lightweight, easy-to-use
>> Open Source Complex Event Processing Engine (CEP) released as a Java
>> Library under Apache Software License v2.0. The readme page of the Siddhi's
>> Github repository contains a good overview of the Siddhi library. Also,
>> SiddhiQL Guide [2] will be helpful you to understand Siddhi Query Language.
>>
>> Most probably, we will be writing this toolkit as a Siddhi extension.
>> Therefore, please go through Siddhi Extension Guide [3] to get a quick
>> overview regarding how to write Siddhi extensions.
>>
>> In addition to that, if you are not familiar with general machine
>> learning concepts, I think Chapter 12 of the Mining of Massive Datasets
>> book (link is provided in the GSco project page) will be really helpful.
>> Also, if you have some free time please look at other streaming machine
>> learning libraries such as Apache Samoa[4].
>>
>> Please go through these Siddhi/ML tutorials/guides. They will be helpful
>> when it comes to writing your project proposal.
>>
>> Thanks,
>> Upul
>>
>> [1]. https://github.com/wso2/siddhi
>> [2]. https://docs.wso2.com/display/CEP420/SiddhiQL+Guide+3.1
>> [3] https://docs.wso2.com/display/CEP420/Writing+Extensions+to+Siddhi
>> [4] https://samoa.incubator.apache.org/
>>
>> On Mon, Feb 20, 2017 at 7:00 PM, Kasun Weerakoon <kasun...@gmail.com>
>> wrote:
>>
>> Hi all,
>>
>> I'm Kasun Weerakoon, final year undergraduate of University of Moratuwa,
>> Sri Lanka. I'm interested about above mentioned project idea as I'm
>> passionate about machine learning and deep learning techniques. I have used
>> ML in my final year project and have followed the Coursera machine learning
>> course by Stanford University as well as a ML course module in the
>> university. I've recently participated to the Deep Learning meetup held in
>> your organization. I use Java as my major programming language and
>> currently  I'm following tutorials on Siddhi. I believe I'm a good
>> candidate for this project and hope your guidance for this project.
>>
>> Regards,
>> Kasun Weerakoon.
>>
>> _______________________________________________
>> Dev mailing list
>> Dev@wso2.org
>> http://wso2.org/cgi-bin/mailman/listinfo/dev
>>
>>
>>
>>
>> --
>> Upul Bandara,
>> Mob: +94 715 468 345 <+94%2071%20546%208345>.
>>
>>
>>
>>
>> --
>> Upul Bandara,
>> Associate Technical Lead, WSO2, Inc.,
>> Mob: +94 715 468 345 <071%20546%208345>.
>>
>>
>>
>>
>>
>>
>> --
>> Upul Bandara,
>> Associate Technical Lead, WSO2, Inc.,
>> Mob: +94 715 468 345 <+94%2071%20546%208345>.
>>
>
_______________________________________________
Dev mailing list
Dev@wso2.org
http://wso2.org/cgi-bin/mailman/listinfo/dev

Reply via email to