Tools
Sorted by:
Try your query at:
 |
 |
 |
 |
 |
 |
Results 1 - 10
of
10,963
Table 1. MAE Results for four filtering algorithms on EachMovie. Four algorithms are pure content-based filtering (CBF), pure collaborative filtering (CF), unified filtering by collaborative ensemble learning (CEL) and unified filtering by combining mixture model and exponential model (UFME).
Table 1. MAE Results for four filtering algorithms on EachMovie. Four algorithms are pure content-based filtering (CBF), pure collaborative filtering (CF), unified filtering by collaborative ensemble learning (CEL) and unified filtering by combining mixture model and exponential model (UFME).
Table 5 Dependence of the performance of steganalyzer on audio content Methods Speech records Pure instrument records Music records
"... In PAGE 9: ....4. The dependence of the steganalyzer on the cover material We investigated the performance dependence of the steganalyzer on the type of document in which data hiding takes place, that is, speech and audio. Table5 presents comparatively the steganalysis performances for different sources (speech, bass, soprano, rock, etc.).... ..."
Table 5 Dependence of the performance of steganalyzer on audio content Methods Speech records Pure instrument records Music records
2005
"... In PAGE 9: ....4. The dependence of the steganalyzer on the cover material We investigated the performance dependence of the steganalyzer on the type of document in which data hiding takes place, that is, speech and audio. Table5 presents comparatively the steganalysis performances for different sources (speech, bass, soprano, rock, etc.).... ..."
Table B1: Table shows the content of the first five packets recorded during a test run with the pure LZ method acting on xfig1.
Table B1: Table shows the content of the first five packets recorded during a test run with the pure LZ method acting on xfig1.
Table 4 shows that performance in double buffering and MRPSD-2 MCR is very similar for all cases. The memory saving factor is very large in all circumstances, including the service of very heterogeneous contents (set F). Its particular value is related to the number of clients served in the system (see set E). Pure CBR contents (sets D and E) need less memory than VBR sets because in this case the estimation made for static reservation is closer to the real requirements.
Table 3: Pure LRU, separate disks
1993
"... In PAGE 15: ...f a page. We also insure that memory is the only resource where the two classes compete to any signi cant degree. This is accomplished by segregating the data referenced by each class onto separate disks and by setting the CPU speed such that processor utilizations are 10% or below (50 MIPS). Table3 shows the resulting response times and memory allocations for this rst experiment. Examining the rst two rows in Table 3, we see the impact of adding queries to a transaction workload: transaction response times double { even when those queries are allocated the absolute minimum amount of working storage.... In PAGE 15: ... Table 3 shows the resulting response times and memory allocations for this rst experiment. Examining the rst two rows in Table3 , we see the impact of adding queries to a transaction workload: transaction response times double { even when those queries are allocated the absolute minimum amount of working storage. Since there is no signi cant contention at the disk or CPU in this experiment, the only reason for the change is a drop in transaction bu er hit rates when queries are added, resulting in an increase in the average I/Os per transaction from about 2 to about 4.... In PAGE 15: ... This hit rate decrease is due to the inability of pure global LRU to distinguish the more frequently accessed transaction pages from the less valuable pages accessed by the queries. The second and third rows of Table3 show the e ects of adding fragment fencing to a pure global LRU memory manager. While the 60 msec goal is not achievable for the transactions, their average response time of 71 msecs under fragment fencing approaches their stand-alone performance of 64.... In PAGE 16: ... Table 4 shows the results of this experiment. If we examine the rst two rows of this table and compare them to the previous experiment ( Table3 ), we can see that adding hints on page type and reference patterns signi cantly reduces the impact of adding queries to a transaction workload relative to the pure LRU case. This is largely because sequential ooding4 is eliminated via the hints.... ..."
Cited by 42
Table 2. Water and ash content of chips.
"... In PAGE 3: ... In this study, the Morbark 1200 tub grinder was able to produce chips that were equal to those produced by the two chippers included in this study. The moisture contents of the logging residues are presented in Table2 . The values obtained were typical for green logging residues.... In PAGE 3: ... The values obtained were typical for green logging residues. The ash content values (also Table2 ) are typical for pure logging residues; there was no sand or other impurities in the material. Machine Productivity and Cost of Chips Produced at the Roadside The gross effective time (E15-h) distributions of the machines are presented in Figures 2, 3, and 4.... ..."
Table 1: Comparison of the classi cation approach and the abstraction approach. The image preprocessing workload in CLA is heavier than in ABA. In both cases, the images are segmented, and a feature vector is computed for each connected component. In CLA, this is followed by a classi cation process resulting in a more time-consuming preprocessing step in CLA compared to ABA. In terms of the time to insert the images into the database after the preprocessing is complete, ABA is slower than CLA. In terms of user interaction during the image interpretation process, in ABA there is no need for human interaction. On the other hand, in CLA the user must assist in creating the training set which is used to automatically classify the images. Retrieving queries purely on the basis of content (with no spatial constraint, e.g., example query Q1) is slightly slower in ABA than in CLA. Our results varied from a retrieval time that was greater by a factor of 1.2 for the small data set, to a retrieval time that was greater by a factor of 2.4 for the large data set. A similar observation was made for hybrid queries (e.g., query Q2). In terms of exibility, ABA has a few advantages. The rst advantage is that it is applicable for a larger number of applications. CLA is only applicable when all the classes of the application are 17
"... In PAGE 18: ...Table1... ..."
Results 1 - 10
of
10,963
