Курсы английского
<<  Natural Language Processing for the Web Create the Internet of Your Things: The Microsoft Vision for IoT  >>
Картинок нет
Картинки из презентации «Estimating the Selectivity of XML Path Expressions for Internet Scale Applications» к уроку английского языка на тему «Курсы английского»

Автор: Ashraf Aboulnaga. Чтобы познакомиться с картинкой полного размера, нажмите на её эскиз. Чтобы можно было использовать все картинки для урока английского языка, скачайте бесплатно презентацию «Estimating the Selectivity of XML Path Expressions for Internet Scale Applications.ppt» со всеми картинками в zip-архиве размером 333 КБ.

Estimating the Selectivity of XML Path Expressions for Internet Scale Applications

содержание презентации «Estimating the Selectivity of XML Path Expressions for Internet Scale Applications.ppt»
Сл Текст Сл Текст
1Estimating the Selectivity of XML Path 38Suffix-* Summarization. Path. Freq.
Expressions for Internet Scale Path. Freq. A. 1. AC. 6. B. 11. AD. 4. C.
Applications. Ashraf Aboulnaga Alaa R. 15. BC. 9. D. 19. BD. 7. AB. 11. CD. 8.
Alameldeen Jeffrey F. Naughton Computer 39Suffix-* Summarization. Path. Freq.
Sciences Department University of Path. Freq. A. 1. AC. 6. B. 11. AD. 4. C.
Wisconsin - Madison. 15. BC. 9. D. 19. BD. 7. AB. 11. CD. 8. *.
2Motivation. XML enables Internet scale 0. **. 0.
applications that query data from many 40Suffix-* Summarization. Path. Freq.
sources Niagara, Xyleme, … Queries over Path. Freq. A. 1. AC. 6. B. 11. AD. 4. C.
XML data use path expressions Optimizing 15. BC. 9. D. 19. BD. 7. AB. 11. CD. 8. *.
these queries requires estimating the 0. **. 0.
selectivity of the path expressions Focus 41Suffix-* Summarization. Path. Freq.
of this talk: Building statistics for XML Path. Freq. AC. 6. B. 11. AD. 4. C. 15.
data and using them for estimating the BC. 9. D. 19. BD. 7. AB. 11. CD. 8. *.
selectivity of simple path expressions. f=1,n=1. **. 0.
3What is XML? <readings> 42Suffix-* Summarization. Path. Freq.
<play> Path. Freq. AC. 6. B. 11. AD. 4. C. 15.
<title>Pygmalion</title> BC. 9. D. 19. BD. 7. AB. 11. CD. 8. *.
<author>Bernard Shaw</author> f=1,n=1. **. 0. SD= { }.
</play> <novel> 43Suffix-* Summarization. Path. Freq.
<title>David Path. Freq. AC. 6. B. 11. C. 15. BC. 9. D.
Copperfield</title> 19. BD. 7. AB. 11. CD. 8. *. f=1,n=1. **.
<author>Charles 0. SD= { (AD,4) }.
Dickens</author> </novel> 44Suffix-* Summarization. Path. Freq.
</readings> Path. Freq. AC. 6. B. 11. C. 15. BC. 9. D.
4Querying XML. FOR $n_auth IN 19. BD. 7. AB. 11. CD. 8. *. f=1,n=1. **.
document("*")//novel/author 0. SD= { (AD,4) }.
$p_auth IN 45Suffix-* Summarization. Path. Freq.
document("*")//play/author WHERE Path. Freq. AC. 6. B. 11. C. 15. BC. 9. D.
$n_auth/text() = $p_auth/text() RETURN 19. BD. 7. AB. 11. CD. 8. *. f=1,n=1. **.
$n_auth Optimizing this query requires 0. SD= { (AD,4) }.
estimating the selectivity of the path 46Suffix-* Summarization. Path. Freq.
expressions This requires information Path. Freq. A*. f=10,n=2. B. 11. C. 15.
about the structure of the XML data. BC. 9. D. 19. BD. 7. AB. 11. CD. 8. *.
5Goal of this Work. Build database f=1,n=1. **. 0. SD= { }.
statistics that capture the structure of 47Suffix-* Summarization. Path. Freq.
XML data Ensure that the statistics fit in Path. Freq. A*. f=10,n=2. B. 11. C. 15.
a small amount of memory For efficient BC. 9. D. 19. BD. 7. AB. 11. CD. 8. *.
query optimization Important for Internet f=1,n=1. **. 0. SD= { }.
scale applications Use the statistics to 48Suffix-* Summarization. Path. Freq.
estimate the selectivity of simple XML Path. Freq. A*. f=10,n=2. B. 11. C. 15.
path expressions //t1/t2/…/tn. BC. 9. D. 19. AB. 11. CD. 8. *. f=1,n=1.
6Outline of Presentation. Introduction **. 0. SD= { (BD,7) }.
Path Trees Markov Tables Performance 49Suffix-* Summarization. Path. Freq.
Evaluation Conclusions. Path. Freq. A*. f=10,n=2. B. 11. C. 15.
7Path Trees. <A> <B> BC. 9. D. 19. AB. 11. CD. 8. *. f=1,n=1.
</B> <B> <D> </D> **. 0. SD= { (BD,7) }.
</B> <C> <D> </D> 50Suffix-* Summarization. Path. Freq.
<E> </E> <E> </E> Path. Freq. A*. f=10,n=2. B. 11. C. 15.
<E> </E> </C> </A> BC. 9. D. 19. AB. 11. *. f=1,n=1. **. 0.
8Summarizing Path Trees. Path trees SD= { (BD,7), (CD,8) }.
contain all the information needed for 51Suffix-* Summarization. Path. Freq.
selectivity estimation Problem: May not Path. Freq. A*. f=10,n=2. B. 11. C. 15.
fit in available memory Small available BC. 9. D. 19. AB. 11. *. f=1,n=1. **. 0.
memory Internet scale Remove low frequency SD= { (BD,7), (CD,8) }.
nodes Removed nodes replaced with *-nodes 52Suffix-* Summarization. Path. Freq.
Tag name: * meaning "any tag" Path. Freq. A*. f=10,n=2. B. 11. C. 15.
Frequency: Average frequency of replaced BC. 9. D. 19. AB. 11. *. f=1,n=1. **. 0.
nodes Sibling-*, Level-*, Global-*, No-*. SD= { (BD,7), (CD,8) }.
9Sibling-* Summarization. 53Suffix-* Summarization. Path. Freq.
10Sibling-* Summarization. A. 1. Path. Freq. A*. f=10,n=2. B. 11. C. 15.
11Sibling-* Summarization. A. 1. I. 2. B*. f=16,n=2. D. 19. AB. 11. *. f=1,n=1.
12Sibling-* Summarization. A. 1. I. J. **. 0. SD= { (CD,8) }.
2. 4. 54Suffix-* Summarization. Path. Freq.
13Sibling-* Summarization. *-nodes Path. Freq. A*. f=10,n=2. B. 11. C. 15.
represent deleted sibling nodes Memory B*. f=16,n=2. D. 19. AB. 11. *. f=1,n=1.
saved by coalescing nodes. A. 1. **. 0. SD= { (CD,8) }.
14Sibling-* Summarization. A. 1. E. 5. 55Suffix-* Summarization. Path. Freq.
*. f=6 n=2. Path. Freq. B. 11. C. 15. B*. f=16,n=2. D.
15Sibling-* Summarization. A. 1. E. H. 19. AB. 11. *. f=1,n=1. **. f=10,n=2. SD=
5. 6. *. f=6 n=2. { (CD,8) }.
16Sibling-* Summarization. A. 1. D. E. 56Suffix-* Summarization. Path. Freq.
H. 7. 5. 6. *. f=6 n=2. Path. Freq. B. 11. C. 15. B*. 8. D. 19.
17Sibling-* Summarization. A. 1. *. f=6 AB. 11. *. 1. **. 6. SD= { }.
n=2. 57Global-*, No-* Summarization. Global-*
18Sibling-* Summarization. A. 1. C. 9. Two *-paths, * and ** Deletes fewer paths
*. f=12 n=2. H. 6. *. f=6 n=2. than suffix-* to summarize the Markov
19Sibling-* Summarization. A. 1. C. 9. table No-* No *-paths Conservatively
*. f=12 n=2. G. H. 10. 6. *. f=6 n=2. assumes that paths not in the Markov table
20Sibling-* Summarization. A. 1. C. 9. do not exist in the data.
*. f=12 n=2. *. f=6 n=2. 58Outline. Introduction Path Trees
21Sibling-* Summarization. A. 1. C. 9. Markov Tables Performance Evaluation
*. f=12 n=2. *. f=16 n=2. *. f=6 n=2. K. Conclusions.
f=23 n=2. 59Data Sets for Experiments. Synthetic
22Sibling-* Summarization. A. 1. C. 9. data set 100,000 XML elements Path tree:
*. *. 6. 8. *. K. f=23 n=2. 3. 3197 nodes, 6 levels, 38 KB Element
23Original Path Tree. frequencies: Zipfian (z=1) DBLP data set
24Sibling-* Summarization. Try to retain 1,399,765 XML elements Path tree: 5883
as much information as possible about the nodes, 6 levels, 69 KB.
deleted nodes. A. 1. C. 9. *. *. 6. 8. *. 60Query Workloads. 1,000 paths of length
K. f=23 n=2. 3. between 1 and 4 Random paths All query
25Level-* Summarization. paths exist in the data Random tags Most
26Level-* Summarization. A. 1. D. E. H. query paths of length 2 or more do not
7. 5. 6. I. J. 2. 4. exist in the data Available memory between
27Level-* Summarization. Less 5 and 50 KB.
information about deleted nodes than 61Best Summarization Methods. Path trees
sibling-* Deletes fewer nodes than Query paths in data: Global-* Query paths
sibling-*. A. 1. not in data: No-* Markov tables m = 2 is
28Global-* Summarization. best Query paths in data: Suffix-* Query
29Global-* Summarization. A. 1. D. E. H. paths not in data: No-*.
7. 5. 6. I. J. 2. 4. 62Path Trees vs. Markov Tables. When to
30Global-* Summarization. 3. *. D. H. 7. use path trees and when to use Markov
6. tables? Also compared against Pruned
31No-* Summarization. Suffix Trees (PSTs) [Chen et al, ICDE
32No-* Summarization. A. 1. D. E. H. 7. 2001] Can handle branching path
5. 6. I. J. 2. 4. expressions Can handle conditions on
33No-* Summarization. Memory savings element values.
similar to global-* Conservative 63Synthetic Data – Random Paths.
assumption about deleted nodes. D. E. H. 64Synthetic Data – Random Tags.
7. 5. 6. 65DBLP Data – Random Paths.
34Outline. Introduction Path Trees 66DBLP Data – Random Tags.
Markov Tables Performance Evaluation 67When are Markov Tables Better? DBLP
Conclusions. Repeated sub-structures effectively
35Markov Tables. A table of all distinct captured by Markov tables. <sigmod>
paths of length up to m and their <inproceedings>
frequencies For paths of length greater <author>…</author> …
than m, combine paths from the Markov </inproceedings> … </sigmod>
table Example: Uses "short <vldb> <inproceedings>
memory" or "Markov" <author>…</author> …
property. </inproceedings> … </vldb>
36Markov Tables. Path. Freq. Path. Freq. 68Conclusions. Novel statistics for
A. 1. AC. 6. B. 11. AD. 4. C. 15. BC. 9. estimating the selectivity of XML path
D. 19. BD. 7. AB. 11. CD. 8. expressions Scale to "all the XML
37Summarizing Markov Tables. Exact data on the Internet" More accurate
selectivities for paths of length up to m than best previously known alternative
Approximate selectivities for paths longer Repeated sub-structures: Markov tables No
than m Problem: May not fit in available repeated sub-structures: Path trees Query
memory Remove low frequency paths Discard paths exist in the data: Global-*,
removed paths of length > 2 Replace Suffix-* Query paths do not exist in the
removed paths of length 1 or 2 with data: No-* To appear in VLDB 2001.
*-paths Suffix-*, Global-*, No-*.
Estimating the Selectivity of XML Path Expressions for Internet Scale Applications.ppt
http://900igr.net/kartinka/anglijskij-jazyk/estimating-the-selectivity-of-xml-path-expressions-for-internet-scale-applications-238688.html
cсылка на страницу

Estimating the Selectivity of XML Path Expressions for Internet Scale Applications

другие презентации на тему «Estimating the Selectivity of XML Path Expressions for Internet Scale Applications»

«English for you» - Твои школьные учебники. Ты научишься правильно строить предложение. «Английский для тебя» (курс английского языка с применением ИКТ). Что ты научишься делать и узнаешь. ENGLISH FOR YOU. You are welcome! Викторины Задания Игры Игровые упражнения Ты можешь выбрать уровень сложности. Слова и выражения по темам.

«The green movement» - The main objective — to achieve the decision of global environmental problems, including by attraction to them of attention of the public and the authorities. "Green" movement in the world. The countries in which there are offices Greenpeace. One of the largest victories гринписовцев in the given campaign can name refusal of flooding of an oil platform brent spar as it contained many toxic substances.

«The english-speaking countries» - Disneyland. The English-speaking countries. Scotland. Australia. USA. Great Britain.

«Женщина the woman» - 9 семантических подгрупп, характеризующих женщин по: От нашего ребра нам не ждать добра; Холостому помогай боже, а женатому хозяйка поможет. Пословицы. Значение понятия «женщина» в семье. The wife is the key to the house. « Der mann»- нем. As great a pity to see a woman cry as a goose go barefoot. Муж - голова, жена- шея; куда хочу- туда верчу.

«The animals» - BEAR. LION. FOX. PENGUIN. The animals which live in a SAVANNA. The animals which live in the OCEAN. KANGAROO. FLAMINGO. SCORPIO. HIPPO. SEA-HORSE. WOMBAT. TIGER. DOLPHIN. WHALE. SEAL. REINDEER. The animals which live in Australia. STARFISH. The animals which live in the desert. PANDA. EMU. POLAR BEAR.

«Работа в Internet» - Существует сеть серверов IRC. Разговор. ICQ выполняется в фоновом режиме, используя минимум памяти и ресурсов сети. Видео. Позволяет в любой момент времени видеть, находятся ли ваши друзья в сети. Позволяет одновременно находиться в онлайне любому количеству пользователей с одинаковыми именами; Звук.

Курсы английского

25 презентаций о курсах английского
Урок

Английский язык

29 тем
Картинки
900igr.net > Презентации по английскому языку > Курсы английского > Estimating the Selectivity of XML Path Expressions for Internet Scale Applications