Collaborative Filtering on Skewed Datasets


 
Thread Tools Search this Thread
Special Forums News, Links, Events and Announcements UNIX and Linux RSS News Collaborative Filtering on Skewed Datasets
# 1  
Old 05-22-2008
Collaborative Filtering on Skewed Datasets

HPL-2008-50 Collaborative Filtering on Skewed Datasets - Banerjee, Somnath; Ramanathan, Krishnan
Keyword(s): collaborative filtering, skewed dataset, pLSA
Abstract: Many real life datasets have skewed distributions of events when the probability of observing few events far exceeds the others. In this paper, we observed that in skewed datasets the state of the art collaborative filtering methods perform worse than a simple probabilistic model. Our test bench inc ...
Full Report

More...
Login or Register to Ask a Question

Previous Thread | Next Thread

5 More Discussions You Might Find Interesting

1. UNIX for Dummies Questions & Answers

Who are all opening my datasets,?

Hi, I need a command/script, who opened my dataset, consider a situation like, if a user has opened the dataset few days back then, that command/script should list his/her id. I don't want audit on my dataset, i need only list of users who are using my dataset. Thank you. (10 Replies)
Discussion started by: subbarao12
10 Replies

2. AIX

Problem in ftpying the datasets containing comp values to AIX from mainframe

Hi, When i am trying to ftp COBOL generated data sets which contain comp values to AIX in ASCII mode. the comp values are getting corrupted. If i ftp the data set in binary mode it is working properly, but for this i have to change some compiler options in the COBOL. Also if i want to use the... (5 Replies)
Discussion started by: sekhar gajjala
5 Replies

3. UNIX for Dummies Questions & Answers

ls command showing skewed listing

Hello, I'm running the ls command on an HP-UX 11i platform and am getting skewed listings. In other words, I see 3 columns of perfectly aligned file names, except 1 file is shifted by 2 or 3 bytes. The file to the immediate left of it seems to be causing the problem, for when I do an ls on... (1 Reply)
Discussion started by: bsp18974
1 Replies

4. Solaris

Copy data from zfs datasets

I 've few data sets in my zfs pool which has been exported to the non global zones and i want to copy data on those datasets/file systems to my datasets in new pool mounted on global zone, how can i do that ? (2 Replies)
Discussion started by: fugitive
2 Replies

5. UNIX for Advanced & Expert Users

Planning on writing a Guide to Working with Large Datasets

In a recent research experiment I was handling, I faced this task of managing huge amounts of data to the order of Terabytes and with the help of many people here, I managed to learn quite a lot of things in the whole process. I am sure that many people will keep facing these situations quite often... (2 Replies)
Discussion started by: Legend986
2 Replies
Login or Register to Ask a Question
drslib()																  drslib()

NAME
drs_tool - Command-line interface to DRSLIB SYNPOSIS
DESCRIPTION
USAGE
Usage Usage: drs_tool [command] [options] [drs-pattern] command: list list publication-level datasets todo show file operations pending for the next version upgrade make changes to the selected datasets to upgrade to the next version mapfile make a mapfile of the selected dataset history list all versions of the selected dataset init initialise CMIP5 product detection data OPTIONS
--help display this help message and exit --version output version information and exit -h, --help show this help message and exit -R ROOT, --root=ROOT Root directory of the DRS tree -I INCOMING, --incoming=INCOMING Incoming directory for DRS files. Defaults to <root>/output -a ACTIVITY, --activity=ACTIVITY Set DRS attribute activity for dataset discovery -p PRODUCT, --product=PRODUCT Set DRS attribute product for dataset discovery -i INSTITUTE, --institute=INSTITUTE Set DRS attribute institute for dataset discovery -m MODEL, --model=MODEL Set DRS attribute model for dataset discovery -e EXPERIMENT, --experiment=EXPERIMENT Set DRS attribute experiment for dataset discovery -f FREQUENCY, --frequency=FREQUENCY Set DRS attribute frequency for dataset discovery -r REALM, --realm=REALM Set DRS attribute realm for dataset discovery -v VERSION, --version=VERSION Force version upgrades to this version -P FILE, --profile=FILE Profile the script exectuion into FILE --detect-product Automatically detect the DRS product of incoming data REQUIRES
FILES
drs_tool uses metaconfig to configure and store configuration information. Its configuration may be found then in a metaconfig.conf file, in /etc/metaconfig/metaconfig.conf on Debian systems, which may be overriden by $HOME/.metaconfig.conf or ./metaconfig.conf ENVIRONMENT VARIABLES
The variable METACONF_CONF VERSION
0.3.0a3 BUGS
To query and report bugs in drs_tool, please use reportbug on Debian systems. SEE ALSO
http://esgf.org/esgf-drslib-site/index.html AUTHOR
DRSLIB was written by Stephen Pascoe, STFC.SC.UK. This manual page was written by Alastair McKinstry. drslib()