J. M. D. Hill. Personal Communication, September 1997. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Algorithmic and Practical Considerations for Dense Matrix .. - Alexandros Gerbessiotis (Correct)
....megaflop performance of the IBM SP2, for the problem instances and the implementations examined, seems to be extremely good and higher than that of the SGI. Finally in parts of the presentation below we shall require the BSP parameters of the machine configurations tested. These were obtained from [17, 18] and can be expressed in tuple form as follows. The SGI Power Challenge of the experiments is reported to have approximately (p; L; g) 4; 2000f lops; 20f lops=double) 17] For the IBM SP2 [17] the corresponding values are (p; L; g) 4; 3600f lops; 16f lops=double) For the CRAY T3D [18] ....
....[17, 18] and can be expressed in tuple form as follows. The SGI Power Challenge of the experiments is reported to have approximately (p; L; g) 4; 2000f lops; 20f lops=double) 17] For the IBM SP2 [17] the corresponding values are (p; L; g) 4; 3600f lops; 16f lops=double) For the CRAY T3D [18], they are (4; 430; 1:6f lops=double) 16; 842; 2:0f lops=double) and (64; 1000; 3:4f lops=double) 5.2 Matrix Multiplication We first implemented algorithms Mult A and Mult B. In the multiplication of two matrices, say A and B to form the product A ThetaB, the operations performed may be ....
J. M. D. Hill. Personal Communication, September 1997.
Learning Automated Product Recommendations Without Observable.. - Lee, Moore (1995) (Correct)
....be extended, and discuss related work. 2 Our Approach A number of groups have begun developing distributed systems to collect and predict consumer preferences, including [ Shardanand and Maes, 1995 ] for the music domain, Resnick et al. 1994 ] and [ Lang, 1994 ] for the Usenet domain, and [ Hill, 1994 ] for the video domain. Some of these groups have put significant effort into implementation issues to do with user interfaces, and the gathering and communicating of data via Internet and Usenet. The ability to analyze and predict consumer preferences will be increasingly valuable in coming ....
....newsgroup an article came from, and achieved up to 73 accuracy with a training set of articles drawn from twenty newsgroups. Lang is currently working to extend these results to predicting people s ratings of Usenet articles. Will Hill at Bellcore is working on a video recommendation service, Hill, 1994 ] and has an Internet site that receives ratings from people. But to our knowledge he has yet to publish results. 7 Conclusions Practical tools to model consumer preferences would be of great benefit in a wide range of domains. With more and more electronic data available on which videos ....
W. Hill. Personal Communication. , 1994.
Algorithmic and Practical Considerations for Dense Matrix.. - Gerbessiotis (1997) (1 citation) (Correct)
....problem instances and the implementations examined, seems to be extremely good and higher than that of the SGI. Finally in parts of the presentation below we shall require the BSP parameters of the machine configurations tested. In an earlier version of this work these values were obtained from [18, 19]. Those values, however, were not reliable and in some cases significantly underestimated the true performance of the version of BSPlib used in these experiments. The underestimation of the BSP parameters coupled with the overestimation of communication time under the BSP cost model led to a good ....
J. M. D. Hill. Personal Communication, September 1997.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC