Re: Meeting next week

Hi!

 Yup, I did take a look!

 My motivation to join the group was to push for better standards in aspect-based sentiment analysis. As I wrote earlier, current benchmarks like SEMEVAL 2015 are pretty bad for evaluating different approaches. Do you think this publication can accomodate this?

Best regards,
Pavel

> On 07 Jul 2015, at 11:30, J. Fernando Sánchez <jfernando@dit.upm.es> wrote:
> 
> Hello Pavel,
> 
> This is the document: https://docs.google.com/document/d/1-nzaTg2wTXU37uE_14czRAieDaVYvXfZRVSkX_CUN-w <https://docs.google.com/document/d/1-nzaTg2wTXU37uE_14czRAieDaVYvXfZRVSkX_CUN-w>
> 
> As you can see, it is a draft for a publication on the state of the art of LD models for sentiment and emotion analysis. We have assigned different tasks to each contributor. Feel free to let us know what you think and whether you would like to participate.
> 
> Cheers!
> 
> 
> 
> On Mon, Jul 6, 2015 at 1:57 PM, Pavel Velikhov <pavel.velikhov@gmail.com <mailto:pavel.velikhov@gmail.com>> wrote:
> Hi,
> 
>   What Google doc should we add contributions to?
> 
> Best regards,
> Pavel Velikhov
> 
>> On 06 Jul 2015, at 14:48, J. Fernando Sánchez <jfernando@dit.upm.es <mailto:jfernando@dit.upm.es>> wrote:
>> 
>> Dear all,
>> 
>> This is a kind reminder of our meeting tomorrow at 13.00 CEST.
>> 
>> Please, remember to add your contributions to the google doc throughout the day.
>> 
>> Regards.
>> 
>> On Thu, Jun 25, 2015 at 4:35 PM, J. Fernando Sánchez <jfernando@dit.upm.es <mailto:jfernando@dit.upm.es>> wrote:
>> Dear all,
>> 
>> You can find the minutes of the meeting here: https://www.w3.org/community/sentiment/wiki/Meetings <https://www.w3.org/community/sentiment/wiki/Meetings>
>> 
>> To those of you who joined: thank you. Feel free to add anything I might have missed and remove anything that should not be public.
>> 
>> The summary is that we need a better review the state of the art for a comprehensive survey. We defined some action points, which we will check in our next meeting on 7th July, 13.00 CEST.
>> 
>> @Pavel, welcome to the community group!. I believe aspect-based sentiment analysis is an important use case for us, and it is also in our interest to have more quality datasets. A common LD model should improve the accessibility/interoperability of different datasets and ease the development of a common toolset.
>> 
>> Regards,
>> 
>> On Thu, Jun 25, 2015 at 2:01 PM, Pavel Velikhov <pavel.velikhov@gmail.com <mailto:pavel.velikhov@gmail.com>> wrote:
>> Hi, I was late for the call today, looked up CEST incorrectly. And at the end of the call something happened to my Skype, seems like nobody could hear me…
>> 
>> So I’m a bit in the dark about the next steps of the group.
>> 
>> My interest is quite simple and I think is shared by a number of companies in the industry: we develop technologies for aspect-based sentiment analysis, but both the training data available and benchmark results need a lot more work to be useful. So that’s my main interest and I’d be happy to share all the details of what I think we’re missing now.
>> 
> 
> С уважением,
> Павел Велихов
> pavel.velikhov@gmail.com <mailto:pavel.velikhov@gmail.com>
> 

С уважением,
Павел Велихов
pavel.velikhov@gmail.com