24小时热门版块排行榜

>论坛更新日志 (3395)
>虫友互识 (431)
>文献求助 (376)
>导师招生 (257)
>硕博家园 (176)
>招聘信息布告栏 (171)
>博后之家 (119)
>休闲灌水 (105)
>论文投稿 (102)
>考博 (76)
>基金申请 (69)
>教师之家 (63)
>金融投资 (44)
>找工作 (42)
>考研 (39)
>论文道贺祈福 (36)

返回列表

糖果车站

捐助贵宾 (初入文坛)

应助: 0 (幼儿园)
金币: 70.8
帖子: 11
在线: 9小时
虫号: 3510405

[交流] 人工智能国际顶会 IJCAI 2017 Workshop征稿

IJCAI  2017  Workshop:  Abuse  Preventive  Data  Mining  2017

URL:  https://www.apdm2017.conferences.academy
Submission  System:  https://easychair.org/conferences/?conf=apdm2017

Workshop  paper  will  be  invited  for  extension  for  the  special  issue  of  an  international
journal  of  Intelligent  Data  Analysis  (SCI)

(https://www.iospress.nl/journal/intelligent-data-analysis/)

==========================================================================================

-----~*******~--------
Important  Dates
-----~*******~--------

Paper  Submission  Due: 07/05/2017
Author  Notification  Due: 07/06/2017
Conference  Days: 19-21/08/2017

-----~*******~--------
APDM  2017  Workshop  Scope
-----~*******~--------

This  workshop  is  with  the  Twenty-sixth  International  Joint  Conference  on  Artificial  Intelligence
(IJCAI-17)  will  be  held  in  Melbourne,  Australia,  August  19-25,  2017.

We  solicit  contributions  on  the  advanced  techniques  for  Abuse  Preventive  Data  Mining.

Data  mining  critically  relies  on  the  information  (data  and  domain  knowledge)  disclosure
from  the  data  curator,  and  the  information  accessibility  from  the  data  miners.  This  fully
or  partial  transfer  of  information  ownership,  if  not  done  properly,  may  lead  to  information
abuse.  Here,  information  abuse  is  referred  to  the  disclosure  of  important  information  for
which  the  data  owners  are  not  willing  to  disclose,  including  but  not  limited  to  the  private
information  of  users,  the  trade  secret  of  businesses,  etc.

Abuse  Preventive  Data  Mining  (APDM)  aims  to  curb  the  potential  information  abuse  across
different  steps  of  data  mining.  There  are  various  related  studies  such  as  privacy-preserving
data  mining,  data  security,  data  propriety  maintenance,  distributed  learning,  etc.
These  efforts,  however,  are  disparate  in  different  domains.  Now  it  is  the  time  to  revisit
from  a  unified  perspective,  especially  when  considering  the  fact  that  most  related  studies
can  be  categorized  based  on  their  levels  of  information  ownership  transferring:

-  Full  Access  to  Data:  data  will  be  processed  before  releasing.
-  Partial  Access  to  Data:  distributed  data  access.
-  No  Access  to  Data:  access  to  intermediate  result.

Advances  in  abuse  protective  data  mining  will  result  in  safer  collaboration  and  trusted
information  sharing.  We  believe  that  it  is  a  good  time  to  cover  these  topics  in  the  workshop,
which  offers  a  timely  forum  for  researchers  and  industry  partners  to  present  and  discuss
latest  advances  in  abuse  preventive  data  mining.

回复此楼