Entry Leveling:2010:SWI from talip.bib
Last update: Sun Oct 15 02:55:04 MDT 2017
Top |
Symbols |
Numbers |
Math |
A |
B |
C |
D |
E |
F |
G |
H |
I |
J |
K |
L |
M |
N |
O |
P |
Q |
R |
S |
T |
U |
V |
W |
X |
Y |
Z
BibTeX entry
@Article{Leveling:2010:SWI,
author = "Johannes Leveling and Gareth J. F. Jones",
title = "Sub-Word Indexing and Blind Relevance Feedback for
{English}, {Bengali}, {Hindi}, and {Marathi} {IR}",
journal = j-TALIP,
volume = "9",
number = "3",
pages = "12:1--12:??",
month = sep,
year = "2010",
CODEN = "????",
DOI = "https://doi.org/10.1145/1838745.1838749",
ISSN = "1530-0226 (print), 1558-3430 (electronic)",
ISSN-L = "1530-0226",
bibdate = "Sat Sep 18 15:58:58 MDT 2010",
bibsource = "http://portal.acm.org/;
http://www.math.utah.edu/pub/tex/bib/talip.bib",
abstract = "The Forum for Information Retrieval Evaluation (FIRE)
provides document collections, topics, and relevance
assessments for information retrieval (IR) experiments
on Indian languages. Several research questions are
explored in this article: (1) How to create a simple,
language-independent corpus-based stemmer, (2) How to
identify sub-words and which types of sub-words are
suitable as indexing units, and (3) How to apply blind
relevance feedback on sub-words and how feedback term
selection is affected by the type of the indexing unit.
More than 140 IR experiments are conducted using the
BM25 retrieval model on the topic titles and
descriptions (TD) for the FIRE 2008 English, Bengali,
Hindi, and Marathi document collections.\par
The major findings are: The corpus-based stemming
approach is effective as a knowledge-light term
conflation step and useful in the case of few
language-specific resources. For English, the
corpus-based stemmer performs nearly as well as the
Porter stemmer and significantly better than the
baseline of indexing words when combined with query
expansion. In combination with blind relevance
feedback, it also performs significantly better than
the baseline for Bengali and Marathi IR.\par
Sub-words such as consonant-vowel sequences and word
prefixes can yield similar or better performance in
comparison to word indexing. There is no best
performing method for all languages. For English,
indexing using the Porter stemmer performs best, for
Bengali and Marathi, overlapping 3-grams obtain the
best result, and for Hindi, 4-prefixes yield the
highest MAP. However, in combination with blind
relevance feedback using 10 documents and 20 terms,
6-prefixes for English and 4-prefixes for Bengali,
Hindi, and Marathi IR yield the highest
MAP.\par
Sub-word identification is a general case of
decompounding. It results in one or more index terms
for a single word form and increases the number of
index terms but decreases their average length. The
corresponding retrieval experiments show that relevance
feedback on sub-words benefits from selecting a larger
number of index terms in comparison with retrieval on
word forms. Similarly, selecting the number of
relevance feedback terms depending on the ratio of word
vocabulary size to sub-word vocabulary size almost
always slightly increases information retrieval
effectiveness compared to using a fixed number of terms
for different languages.",
acknowledgement = ack-nhfb,
articleno = "12",
fjournal = "ACM Transactions on Asian Language Information
Processing",
journal-URL = "http://portal.acm.org/browse_dl.cfm?&idx=J820",
keywords = "blind relevance feedback; evaluation; FIRE;
Information retrieval; stemming; sub-word indexing",
}
Related entries
- all,
6(4)2,
7(1)1,
7(2)7,
7(3)8,
7(4)11,
7(4)12,
8(2)7,
8(4)16,
8(4)17,
9(2)5,
9(3)11,
10(2)9,
10(3)15,
11(2)6,
11(2)7,
11(4)18,
12(2)5,
12(3)9,
13(1)2,
13(2)6,
13(4)16
- almost,
6(2)6,
7(3)9,
9(3)11,
11(1)1,
13(1)1
- always,
7(3)10,
9(3)11,
10(1)6
- apply,
7(3)8,
8(1)4,
8(4)18,
8(4)19,
11(4)13,
12(2)7,
12(4)15
- article,
3(4)227,
4(3)321,
5(2)121,
6(2)6,
6(2)7,
6(2)8,
6(4)3,
7(1)1,
7(1)3,
7(2)5,
7(2)6,
7(2)7,
7(3)8,
7(3)9,
7(4)11,
7(4)12,
7(4)13,
8(1)2,
8(1)3,
8(1)4,
8(2)6,
8(2)8,
8(2)9,
8(3)10,
8(3)11,
8(3)12,
8(4)14,
8(4)16,
8(4)17,
8(4)18,
9(1)2,
9(1)4,
9(2)6,
9(3)10,
9(3)11,
9(4)13,
9(4)14,
10(1)3,
10(1)5,
10(1)6,
10(2)7,
10(2)9,
10(2)10,
10(3)12,
10(3)13,
10(3)14,
10(3)15,
10(4)17,
10(4)18,
10(4)20,
10(4)21,
11(1)1,
11(2)4,
11(2)5,
11(2)7,
11(3)8,
11(3)10,
11(3)11,
11(4)13,
11(4)14,
11(4)15,
11(4)16,
11(4)17,
11(4)18,
12(1)1,
12(1)3,
12(1)4,
12(2)5,
12(2)6,
12(2)7,
12(3)9,
12(3)10,
12(3)11,
12(3)12,
12(4)14,
13(1)1,
13(1)2,
13(1)3,
13(1)4,
13(2)6,
13(2)7,
13(2)8,
13(2)9,
13(3)12,
13(3)13,
13(4)16
- assessment,
7(3)10
- average,
7(3)10,
9(3)11,
11(3)10,
13(2)6
- based, corpus-,
1(3)207,
5(2)89,
9(2)5,
9(4)14,
13(1)4
- baseline,
7(1)1,
7(2)5,
7(2)6,
7(3)8,
8(1)4,
8(3)10,
9(1)3,
9(1)4,
9(2)6,
9(2)7,
11(4)13,
11(4)14
- benefit,
7(2)7,
7(4)13,
8(4)15,
8(4)16,
12(1)2,
12(4)15
- Bengali,
9(3)11,
10(2)9,
11(1)1,
13(3)14
- best,
5(2)89,
5(3)183,
7(2)7,
7(4)13,
8(2)6,
9(1)3,
9(2)7,
9(3)11,
11(3)8,
11(4)13,
12(1)2,
12(2)5,
12(2)7,
12(3)9,
12(4)14,
13(1)4,
13(4)16
- better,
6(2)8,
7(2)7,
7(4)13,
8(1)4,
8(3)11,
8(4)16,
9(1)1,
9(2)7,
9(3)11,
11(3)8,
11(4)13,
11(4)15,
12(2)7
- blind,
13(3)13
- case,
8(2)8,
8(4)14,
9(2)5,
10(3)12,
10(4)18,
12(1)1,
12(1)2,
13(2)8
- collections,
7(2)5,
8(1)2,
9(3)10,
9(3)11,
13(3)13
- combination,
3(2)113,
4(3)340,
6(4)3,
8(1)4,
8(2)6,
8(4)16,
10(3)12,
10(3)15,
10(4)19,
11(4)17,
12(2)5,
12(2)7,
12(4)14,
13(2)9
- combined,
8(2)6,
8(4)16,
9(1)4,
10(1)6,
12(2)7,
12(4)15
- compared,
5(2)121,
6(3)9,
6(3)11,
8(4)16,
9(1)1,
9(3)11,
10(4)17,
11(2)6,
12(1)2,
12(4)17,
13(1)2,
13(1)4,
13(2)6,
13(3)12
- comparison,
1(2)145,
1(3)225,
2(3)193,
6(3)9,
8(1)2,
9(3)11,
11(2)6,
13(1)3
- conducted,
6(2)8,
7(4)12,
8(1)2,
9(2)7,
9(3)10,
11(2)4,
11(2)7,
13(1)3,
13(4)18
- corpus-based,
1(3)207,
5(2)89,
9(2)5,
9(4)14,
13(1)4
- corresponding,
5(2)89,
7(3)10,
8(2)7,
9(2)7,
10(2)7,
11(2)5,
11(3)8
- create,
7(2)6,
7(4)11,
9(3)10,
10(1)4
- description,
11(4)13
- different,
5(2)89,
6(3)9,
6(4)3,
7(2)7,
7(3)8,
7(4)13,
8(1)2,
8(2)7,
8(2)8,
8(3)11,
8(4)16,
8(4)17,
9(1)1,
9(1)4,
9(2)5,
9(2)6,
10(1)4,
10(1)5,
10(3)12,
10(4)17,
10(4)19,
11(3)8,
11(3)11,
11(4)16,
11(4)17,
11(4)18,
12(1)2,
12(3)11,
12(4)17,
13(2)6,
13(3)11,
13(4)16
- document,
1(3)225,
2(1)1,
3(2)128,
4(3)340,
5(2)121,
6(4)2,
7(1)3,
7(2)7,
7(3)8,
8(1)1,
8(1)2,
8(1)3,
8(3)11,
8(3)12,
8(4)19,
11(2)5,
12(1)3,
13(1)1,
13(2)6,
13(2)7,
13(2)9,
13(3)13,
13(4)17
- effective,
4(2)78,
6(2)7,
6(3)11,
7(4)12,
7(4)13,
8(1)2,
8(3)10,
9(2)5,
10(2)10,
10(3)14,
10(4)17,
10(4)18,
11(2)4,
11(2)7,
11(4)18,
12(2)7,
12(4)14,
13(3)13,
13(4)16
- effectiveness,
6(2)6,
6(2)8,
6(4)2,
7(2)5,
7(3)10,
8(2)7,
8(4)17,
9(2)5,
9(3)11,
10(3)15,
11(2)5,
11(4)18,
12(4)16,
12(4)17,
13(2)6,
13(3)13,
13(4)18
- English,
2(3)245,
4(2)135,
5(2)89,
5(2)121,
5(3)245,
6(2)6,
6(2)7,
6(3)11,
6(4)2,
7(1)1,
7(4)11,
8(2)9,
8(4)15,
8(4)16,
8(4)17,
9(1)1,
9(1)3,
9(2)7,
9(4)14,
9(4)15,
10(1)2,
10(1)4,
10(2)8,
10(3)14,
10(3)15,
10(4)17,
11(2)4,
11(2)5,
11(2)6,
11(3)8,
11(3)11,
12(2)5,
12(3)12,
12(4)14,
12(4)17,
13(1)1,
13(4)16
- evaluation,
3(4)243,
5(2)165,
6(1)z-2,
7(2)7,
7(4)13,
8(2)6,
8(3)12,
8(4)15,
8(4)16,
9(1)3,
9(2)6,
9(3)10,
9(3)11,
9(4)14,
10(3)15,
11(2)5,
11(4)13,
11(4)18,
12(2)6,
12(4)15,
13(1)1
- expansion,
4(2)57,
8(1)4,
9(1)1,
12(2)7,
13(3)13
- experiment,
2(2)101,
2(2)143,
5(2)146,
5(2)165,
5(3)245,
6(2)7,
6(2)8,
7(1)1,
7(1)2,
7(1)3,
7(2)5,
7(4)11,
8(1)2,
8(2)6,
8(3)11,
8(4)17,
9(2)6,
9(2)7,
9(3)11,
10(4)20,
11(2)4,
11(2)5,
11(2)7,
11(3)10,
11(3)11,
11(4)15,
11(4)17,
12(1)2,
12(2)5,
12(2)7,
12(3)9,
12(3)11,
12(3)12,
12(4)16,
12(4)17,
13(1)2,
13(1)3,
13(2)6,
13(2)8,
13(3)13,
13(4)18
- explored,
8(1)3
- feedback,
4(2)111,
6(2)8,
12(1)2,
12(1)4,
13(3)13
- few,
8(2)9,
8(4)18,
9(2)7,
11(1)1,
13(2)6,
13(2)8
- finding,
10(1)6,
10(2)10,
10(3)14,
11(4)13,
12(1)3
- FIRE,
9(3)10,
9(3)11
- fixed,
9(2)7
- form,
6(3)9,
7(1)3,
8(1)3,
8(4)18,
9(1)3,
9(4)13,
10(2)8,
11(2)7,
11(4)13,
12(1)4,
12(3)11
- Forum,
9(3)10
- general,
8(4)14,
10(3)12,
11(2)6,
11(2)7,
11(3)10,
12(2)6,
12(4)14,
12(4)15
- highest,
12(1)2
- Hindi,
2(2)85,
2(2)118,
2(2)130,
2(2)164,
2(3)169,
2(3)193,
2(3)219,
2(3)245,
2(3)270,
2(3)290,
9(3)11,
9(4)14,
10(2)9,
13(2)8
- how,
5(2)146,
6(4)3,
7(1)2,
7(3)9,
7(4)12,
8(4)14,
8(4)16,
9(2)5,
11(3)8,
12(4)14,
13(2)9,
13(2)10
- however,
6(2)8,
7(2)7,
7(3)8,
7(3)10,
7(4)11,
7(4)12,
7(4)13,
8(3)10,
9(1)1,
10(3)12,
10(4)18,
11(2)6,
11(2)7,
12(1)2,
12(2)7,
12(3)10,
12(3)11,
12(4)14,
13(3)11,
13(4)17
- identification,
4(4)417,
8(3)11,
9(1)2,
10(2)10,
11(1)2
- identify,
5(2)146,
6(2)6,
8(3)11,
10(1)5,
11(2)6,
11(4)18,
12(3)10
- increase,
7(2)5,
7(3)9,
8(2)7,
8(4)15,
9(4)15,
10(2)8,
11(2)6,
11(4)18,
12(2)6,
12(3)11
- independent, language-,
5(2)89,
9(3)11
- index,
6(2)8,
7(4)11,
13(4)18
- indexing,
1(3)225,
6(2)8,
7(4)11,
9(3)11,
11(3)11
- Indian,
8(2)8,
9(3)9,
9(3)10,
10(2)8,
10(2)9,
11(1)1,
13(2)8,
13(3)14
- information,
1(1)65,
1(1)83,
1(4)281,
2(3)245,
2(3)295,
3(1)1,
3(4)227,
4(2)57,
4(2)78,
4(3)243,
4(3)357,
4(4)375,
4(4)475,
5(1)44,
5(2)89,
5(3)264,
5(4)291,
5(4)296,
5(4)323,
6(2)7,
6(4)2,
7(1)2,
7(1)3,
7(2)5,
7(3)8,
7(4)12,
8(1)2,
8(1)3,
8(3)10,
8(3)11,
8(4)15,
8(4)16,
8(4)17,
9(1)1,
9(2)7,
9(3)9,
9(3)10,
9(3)11,
9(4)13,
9(4)14,
9(4)15,
10(2)8,
10(2)10,
10(3)15,
10(3)16,
10(4)19,
10(4)20,
10(4)21,
11(1)2,
11(2)6,
11(2)7,
11(4)15,
11(4)18,
12(2)5,
12(3)11,
12(4)16,
13(2)7,
13(3)13
- IR,
4(3)340,
7(3)8,
8(1)2,
9(3)10,
9(3)11,
13(2)7
- language-independent,
5(2)89,
9(3)11
- language-specific,
5(2)121,
9(1)3
- larger,
8(4)18,
12(1)2
- length,
7(1)3,
8(4)19,
10(4)19,
12(4)14
- major,
5(2)89,
10(4)17,
11(1)1
- Marathi,
9(3)11
- more,
5(2)146,
6(2)7,
6(3)10,
7(3)9,
7(4)13,
8(1)4,
8(2)7,
8(3)12,
8(4)14,
8(4)16,
9(1)2,
9(2)6,
9(3)11,
10(1)4,
10(4)19,
11(2)4,
11(2)7,
12(1)1,
12(1)2,
12(3)9,
13(1)1,
13(1)3,
13(1)4,
13(2)8,
13(4)18
- nearly,
7(4)12,
11(4)13
- number,
6(2)8,
6(3)9,
6(3)11,
7(1)3,
7(3)9,
7(4)11,
8(1)3,
8(4)18,
10(1)5,
10(2)8,
10(4)21,
11(1)2,
11(3)8,
11(3)9,
12(1)2,
12(2)7,
12(3)9,
13(2)7,
13(3)12,
13(3)13
- obtain,
7(2)7,
7(3)8,
7(3)9,
8(4)15,
9(2)5,
10(3)12,
11(3)8,
12(3)10,
12(4)17,
13(3)14
- one,
5(2)89,
5(2)121,
6(2)6,
6(3)9,
6(4)3,
7(3)8,
7(3)9,
7(4)11,
7(4)13,
8(2)9,
8(4)16,
8(4)17,
9(1)1,
9(2)5,
9(2)7,
9(4)14,
10(1)5,
10(3)12,
10(3)13,
10(4)19,
11(2)4,
11(2)6,
11(2)7,
11(4)14,
12(1)1,
12(1)2,
12(2)5,
12(2)7,
12(3)11,
12(4)16,
13(1)4,
13(2)10,
13(4)17,
13(4)18
- overlapping,
8(3)11,
12(1)2
- perform,
7(1)1,
9(4)14,
11(2)4,
11(3)11,
11(4)16,
12(1)2,
13(4)17
- performance,
5(2)121,
5(2)165,
6(2)8,
6(3)9,
6(4)1,
6(4)3,
7(1)1,
7(1)2,
7(2)5,
7(2)6,
7(2)7,
7(3)9,
7(3)10,
7(4)13,
8(1)2,
8(1)3,
8(2)7,
8(2)8,
8(2)9,
8(3)10,
8(4)16,
8(4)17,
8(4)18,
9(1)2,
9(1)4,
9(2)5,
9(2)6,
9(3)11,
9(4)14,
10(2)8,
10(3)13,
10(3)14,
11(2)7,
11(3)10,
11(3)11,
11(4)14,
11(4)15,
11(4)17,
12(1)2,
12(3)9,
12(3)11,
12(4)14,
12(4)15,
12(4)16,
13(1)3,
13(1)4,
13(2)7,
13(2)9,
13(4)16,
13(4)17
- performing,
6(2)7,
7(4)12
- prefix,
6(4)2,
8(4)16,
12(4)15
- provide,
7(1)2,
8(2)9,
8(3)10,
8(3)12,
9(2)5,
9(3)11,
9(4)13,
12(1)2,
12(1)3,
12(4)14
- query,
1(2)123,
1(2)159,
4(2)57,
5(3)245,
5(4)323,
5(4)360,
6(2)8,
6(4)2,
7(2)5,
9(1)1,
11(2)6,
11(3)11,
11(4)17,
13(2)6,
13(3)13
- question,
4(3)243,
4(3)263,
6(1)z-2,
7(2)6,
7(4)12,
8(3)10,
10(4)21,
11(1)2,
11(4)13
- ratio,
8(1)3,
11(4)13
- relevance,
6(2)8,
8(1)3,
10(4)21
- research,
2(2)101,
5(2)89,
7(4)12,
8(1)2,
8(1)4,
8(4)18,
9(2)7,
10(1)4,
10(3)14,
10(4)19,
10(4)20,
10(4)21,
11(1)1,
11(1)2,
11(3)10,
12(1)3,
12(4)14,
13(1)1,
13(2)7,
13(2)10
- resource,
2(2)101,
2(2)124,
4(2)135,
5(2)89,
7(3)9,
7(4)11,
8(2)9,
8(3)12,
8(4)17,
9(2)5,
9(4)14,
10(2)8,
10(3)12,
10(3)16,
10(4)19,
11(4)14,
11(4)16,
11(4)18,
12(1)1,
12(4)14,
12(4)16,
13(1)1,
13(1)3,
13(2)7
- result,
4(2)135,
5(2)121,
5(2)146,
5(2)165,
6(2)6,
6(2)7,
6(3)9,
6(3)11,
6(4)3,
7(1)2,
7(2)5,
7(2)6,
7(2)7,
7(3)8,
7(3)10,
7(4)11,
7(4)12,
7(4)13,
8(1)2,
8(1)3,
8(1)4,
8(2)6,
8(2)9,
8(3)10,
8(3)12,
8(4)14,
8(4)15,
8(4)16,
8(4)17,
8(4)18,
8(4)19,
9(1)1,
9(1)2,
9(2)5,
9(2)6,
9(2)7,
9(3)11,
9(4)14,
10(1)2,
10(2)7,
11(2)4,
11(2)5,
11(3)8,
11(3)9,
11(3)11,
11(4)13,
11(4)14,
11(4)15,
12(1)3,
12(1)4,
12(2)5,
12(2)7,
12(3)9,
12(3)10,
12(3)11,
12(4)14,
12(4)16,
13(1)1,
13(1)4,
13(2)6,
13(2)7,
13(2)9,
13(3)11,
13(3)12,
13(3)14
- retrieval,
1(3)225,
1(4)281,
2(1)1,
2(2)164,
3(2)128,
4(1)1,
4(2)57,
4(2)78,
4(2)135,
4(2)186,
4(3)340,
4(3)357,
4(4)377,
4(4)475,
5(2)89,
5(3)264,
5(4)296,
5(4)323,
6(2)8,
6(4)2,
7(2)5,
7(3)8,
7(4)12,
8(1)1,
8(1)2,
9(1)1,
9(3)9,
9(3)10,
9(3)11,
9(4)13,
9(4)14,
10(2)8,
10(4)20,
11(1)2,
11(3)11,
13(2)6,
13(2)7,
13(3)13,
13(4)18
- selecting,
7(4)13
- selection,
1(1)34,
7(1)3,
8(1)3,
8(2)8,
8(4)15,
8(4)19,
9(2)5,
12(1)1,
13(3)13
- sequence,
5(2)165,
6(2)7,
6(3)10,
7(3)9,
8(1)2,
8(1)3,
9(2)7,
9(4)14,
10(1)4,
12(1)2,
12(1)3,
12(3)9,
12(3)10,
12(4)17,
13(1)2
- several,
6(2)6,
6(2)7,
6(4)3,
7(2)5,
7(2)7,
7(3)10,
8(3)10,
8(4)16,
8(4)17,
8(4)18,
11(2)6,
11(4)13,
11(4)16,
12(1)2,
13(3)12,
13(3)14
- show,
5(2)89,
5(2)146,
7(1)1,
7(1)2,
7(1)3,
7(4)11,
7(4)12,
7(4)13,
8(1)4,
8(2)7,
8(2)9,
8(3)12,
8(4)16,
8(4)17,
9(1)1,
9(1)2,
9(1)3,
9(2)5,
9(2)6,
9(2)7,
9(3)11,
9(4)14,
10(1)3,
10(3)15,
11(2)4,
11(2)5,
11(2)7,
11(3)8,
11(3)11,
11(4)14,
11(4)15,
11(4)17,
11(4)18,
12(1)2,
12(1)4,
12(2)5,
12(2)7,
12(3)9,
12(3)10,
12(3)11,
12(4)15,
12(4)16,
13(1)3,
13(2)6,
13(2)7,
13(2)9,
13(3)14
- significantly,
7(1)1,
7(4)13,
8(2)6,
8(4)15,
8(4)18,
9(1)2,
9(2)6,
11(2)6,
11(4)13,
11(4)17,
12(4)16,
13(2)6,
13(4)17
- similar,
5(2)146,
7(3)8,
7(4)11,
8(3)12,
9(3)11,
10(1)2,
10(2)10,
11(2)6,
11(3)11,
12(1)2,
12(4)17,
13(2)8
- simple,
9(2)5,
10(4)18,
12(1)2,
12(3)9,
13(1)4
- single,
7(3)8,
8(3)11,
11(2)5,
13(3)14,
13(4)17
- size,
5(3)245,
6(3)9,
9(2)7
- specific, language-,
5(2)121,
9(1)3
- stemmer,
9(3)11,
10(2)8,
13(2)6,
13(2)7
- stemming,
6(4)2,
9(3)11,
10(2)8,
13(2)7,
13(3)14
- step,
7(3)8,
8(2)8,
8(4)16,
10(1)5,
10(3)12,
11(2)6,
12(1)2,
12(2)7,
13(3)14,
13(4)17
- such,
7(2)7,
7(3)8,
7(3)10,
7(4)12,
8(2)8,
8(3)10,
8(3)11,
8(3)12,
8(4)14,
8(4)16,
8(4)17,
9(1)1,
9(4)13,
9(4)15,
10(1)5,
10(2)8,
10(3)12,
10(4)21,
11(1)2,
11(2)5,
11(2)7,
11(3)8,
11(3)10,
11(3)11,
11(4)13,
11(4)16,
11(4)17,
11(4)18,
12(1)1,
12(1)2,
12(2)6,
12(3)10,
12(3)11,
12(4)14,
12(4)17,
13(1)1,
13(3)12,
13(4)17
- suitable,
7(2)5,
7(3)9,
10(3)16,
11(2)4
- term,
1(1)34,
4(3)357,
5(1)22,
7(2)5,
7(2)7,
7(4)12,
8(2)9,
9(1)1,
10(4)17,
11(2)4,
11(3)9,
12(2)5,
12(4)15,
13(3)13
- than,
5(2)146,
6(3)10,
7(3)8,
7(4)11,
7(4)13,
8(2)8,
8(4)16,
9(1)2,
9(2)7,
9(3)11,
10(1)3,
10(1)4,
10(3)14,
11(2)4,
11(2)7,
11(3)8,
11(3)9,
11(4)13,
11(4)15,
12(2)7,
12(3)10,
12(4)16,
13(1)1,
13(4)17
- there,
7(3)9,
7(4)11,
8(2)7,
8(3)12,
8(4)17,
9(2)5,
9(4)15,
10(1)2,
10(3)14,
12(1)2,
13(1)1,
13(2)8,
13(3)14
- topic,
2(2)85,
3(4)227,
5(4)388,
6(3)11,
8(1)2,
8(3)12,
9(2)7,
9(4)15,
11(2)4,
11(2)5
- type,
7(4)11,
8(4)17,
9(2)6,
9(3)11,
10(4)19,
11(1)2,
11(1)3,
11(4)18,
12(3)10,
12(3)11,
13(1)3,
13(2)6,
13(3)11
- unit,
6(3)9,
10(1)6,
11(2)5,
11(2)7,
11(4)16,
11(4)18,
13(2)9
- useful,
4(3)243,
5(2)146,
6(4)1,
7(2)6,
8(4)19,
10(4)20,
12(4)16,
13(3)13
- vocabulary,
6(3)9,
8(1)2,
11(2)7,
12(4)14
- well,
5(2)121,
6(2)6,
6(3)11,
7(3)8,
7(3)9,
7(3)10,
8(1)2,
8(4)18,
9(1)2,
10(3)15,
11(1)2,
11(2)4,
11(3)11,
12(3)9,
13(1)2
- when,
5(2)89,
6(3)9,
6(3)11,
7(2)6,
7(4)11,
7(4)13,
8(1)3,
8(4)16,
8(4)17,
9(1)3,
9(2)5,
9(3)11,
9(4)13,
11(2)4,
12(1)2,
12(4)14,
13(2)8,
13(3)12
- which,
5(2)89,
5(2)121,
6(2)6,
6(2)8,
6(3)10,
6(4)1,
7(1)3,
7(2)7,
7(3)10,
7(4)13,
8(1)2,
8(1)4,
8(2)9,
8(3)10,
8(3)12,
8(4)14,
8(4)16,
8(4)18,
9(1)1,
9(2)5,
9(4)14,
9(4)15,
10(1)4,
10(1)6,
10(2)7,
10(2)8,
10(2)9,
10(3)12,
10(3)13,
10(3)14,
10(3)15,
10(4)17,
10(4)19,
10(4)20,
11(2)6,
11(3)8,
11(3)9,
11(4)13,
11(4)14,
11(4)15,
11(4)16,
11(4)18,
12(1)1,
12(1)3,
12(1)4,
12(2)5,
12(2)6,
12(2)7,
12(3)10,
12(3)11,
12(4)15,
12(4)17,
13(1)1,
13(2)8,
13(3)13,
13(4)17,
13(4)18
- yield,
6(2)7,
8(3)12,
12(2)7