Tools
Sorted by:
Try your query at:
 |
 |
 |
 |
 |
 |
Results 1 - 10
of
3,216
Table 6. Results for a Nonsmooth Shape Using Repeated Extrapolation These results show that repeated extrapolation can be very e ective, even when nons- mooth initial corners are present. Indeed, Table 6 suggests that an approximately second order method arises from the use of Extrapolation II. Topological mergings and breakings in three dimensional or a ne ows can also pro- duce nonsmooth corners. For this reason, we expect no better than an O( log( )) error to arise in these situations. To obtain a deeper understanding of the form of these errors, further studies are needed since an additional dependency on nonlocal properties of the interface occurs.
1996
Cited by 6
Table 2. Argument categories and understanding and confidence
"... In PAGE 3: ... The communicative effort is proportional to the size of the derivation tree that corresponds to the degree of detail in which the argument is to be presented. The degrees of understanding and confidence depend on the argument category, as sketched in Table2 . For an argument appealing to experts opinion, the degree of understanding is generally low, since a deeper understanding would require expertise.... ..."
Table 18: Raw data for the deeper inheritance hierarchy experiment
"... In PAGE 42: ... Finally, variable (20) Extra (any extra comments) is graded as: 1 nothing, 2 modi cation was straight forward, 3 OO is good to work with, and 4 other answers. The logical grouping of attributes in Table18 (the second experiment) are the same as above except for the following, variable (12) Diff (what caused the most di culty) is graded as: 1 inheritance related di culties, 2 syntax di culties and compiler errors, 3 choosing which class to specialize from, 4 part of the modi cation, and 5 other answers. Variable (13) Consume (most time consuming action) is graded as: 1 editing and making changes, 2 understanding the code, 3 removing errors, and 4 inheritance related di culties.... ..."
Table 8: Endo2m;n showing regions of object location likelyhood computed for each gridpoint m,n by super- imposing locality patterns from Endo1i;j value. and understanding (representation). It is based on the assumption that some deeper representational level or core structure might be identi ed as a common base for di erent notions of meaning developped sofar in theories of referential and situational semantics as well as some structural or stereotype semantics. For the purpose of testing semiotic processes, their situational complexity has to be reduced by abstract- ing away irrelevant constituents, hopefully without
Table 7: Subjects apos; completion times with a deeper inheritance hierarchy ability of object-oriented programs - subjects maintaining the inheritance program version will take longer than those subjects maintaining the at program version. For this hypothesis we did provide a direction because the depth being empirically investigated is within the range indicated in the results of the questionnaire survey where the majority of respondents said understanding, and hence maintenance, di culties begin to occur [12].
"... In PAGE 17: ... For this hypothesis we did provide a direction because the depth being empirically investigated is within the range indicated in the results of the questionnaire survey where the majority of respondents said understanding, and hence maintenance, di culties begin to occur [12]. Table7 presents the timing data collected for subjects to complete the maintenance task3. Table 8 presents this timing data in the same summarised form.... ..."
Table 13: Outlier data points from Figure 2 (i) Subject 26 had quite a bit of di culty dealing with syntax errors and removing them took the subject a large amount of their total time. In addition, the subject mentioned in their debrie ng questionnaire they thought there was quite a steep learning curve for object-oriented develop- ment. (ii) Subject 30 made a simple error when editing a le and this error took approximately an hour to nd. (iii) Subject 31 stated they took over half an hour to fully understand what was required for the maintenance task. In the internal replication, subject 11 stated it took them about 20 minutes to understand what was required for the maintenance task. Furthermore, the subject had quite a bit of di culty removing syntax errors. Finally, in the experiment using a deeper inheritance hierarchy, subject 29 had di culty deciding which class to specialise from and twice changed their mind about which class they should use. In addition, the subject tried to tackle the task immediately and understand the code as required. This approach may have increased their overall time.
"... In PAGE 29: ... It should be noted that all outlier data points are from boxplots which represent times for the inheritance groups. As detailed in Table13 there are three subjects times (from 27) which outlie the boxplot in the rst experimental study, one subject time (from 13) in the internal replication, and one subject time (from 15) in the experiment using the deeper inheritance hierarchy. The outlier subjects from the rst experimental study all had reasons for their longer times:... ..."
Table 11: Evaluation of iSTART with Middle School Students
"... In PAGE 8: ... In both studies, we found that those students with lower reading abilities benefited in terms of understanding the text at the textbase level of comprehension. In contrast, those with higher reading skills gained in terms of developing a deeper understanding of the text as assessed by bridging inference questions (see Table11 in Appendix A). We believe that these results reflect different benefits of training for readers with different skill sets.... ..."
Table 1. Generic relationship taxonomy
"... In PAGE 2: ... Relationship analysis helps analysts and designers develop a deeper understanding of the relationship structure of application domains, and provides useful relationships for implementations. Yoo and Bieber (2000) suggest a systematic technique for relationship analysis and also provide a generic, domain-independent relationship taxonomy ( Table1 ) based on an extensive literature review (Yoo 2000). The taxonomy is used to create a set of brainstorming questions that an analyst uses to elicit domain information from the domain expert.... ..."
Table 1: Associative responses and their frequencies according to Russel amp; Jenkins. 2. Computation of the Connective Weights We have built up a lexical net whose nodes are the set of the 100 stimulus words of the Russel amp; Jenkins study, together with the ve most frequent associative responses to each of the 100 stimuli. Altogether these are 371 di erent words. The strengths of the connective weights have been computed on the basis of two text corpora: 1. Grolier apos;s Electronic Encyclopedia which contains 10 million words approximately; 2. all abstracts of the psychological data-base PsycLIT, a total of 20 million words. 1This research was supported by the DFG (project 524/88) and by the Heinz-Nixdorf-Foundation
"... In PAGE 1: ... In this study 1000 subjects, American college students, were given a list of 100 stimulus words and they had to respond to each stimulus with the rst word they could think of. Table1 shows the ve most frequent responses for six stimulus words. The results of our study will help one to gain a deeper understanding of the automatic processes which determine verbal behavior; moreover, the simulation of human word associations can be an important component for natural language parsers and information retrieval systems.... ..."
Table 3: The types of pointer pairs verified. Of the 235 pairs that were manually inspected, we classified each verified pair by the defining scope of its members. A local pointer vari- able was always counted as a local reference, even if it could be proved to always equal a pointer argument.
"... In PAGE 7: ... We expect that a programmer with a deeper understanding of the code would be able to verify annotations almost instantaneously. Table3 shows the counts for the different relationships between the verified pointers. Not surprisingly, few pointer pairs were be- tween global references as the compiler can almost always differ- entiate the pointers in this case.... ..."
Results 1 - 10
of
3,216
