=Paper=
{{Paper
|id=Vol-1166/CLEF2000wn-adhoc-GollinsEt2000
|storemode=property
|title=Sheffield University: CLEF 2000 Submission (Bilingual Track - German to English)
|pdfUrl=https://ceur-ws.org/Vol-1166/CLEF2000wn-adhoc-GollinsEt2000.pdf
|volume=Vol-1166
|dblpUrl=https://dblp.org/rec/conf/clef/GollinsS00a
}}
==Sheffield University: CLEF 2000 Submission (Bilingual Track - German to English)==
https://ceur-ws.org/Vol-1166/CLEF2000wn-adhoc-GollinsEt2000.pdf
Sheffield University
CLEF 2000 Submission
(Bilingual Track - German to English)
Tim Gollins and Mark Sanderson1
(Department of Information Studies, University of Sheffield, Sheffield, South Yorkshire, UK)
Abstract
We investigated dictionary based cross language information retrieval using lexical triangulation.
Lexical triangulation combines the results of different transitive translations. Transitive translation
uses a pivot language to translate between two languages when no direct translation resource is
available. We took German queries and translated then via Spanish, or Dutch into English. We
compared the results of retrieval experiments using these queries, with other versions created by
combining the transitive translations or created by direct translation. Direct dictionary translation of a
query introduces considerable ambiguity that damages retrieval, an average precision 79% below
monolingual in this research. Transitive translation introduces more ambiguity, giving results worse
than 88% below direct translation. We have shown that lexical triangulation between two transitive
translations can eliminate much of the additional ambiguity introduced by transitive translation.
1 Introduction and Background
Cross Language Information Retrieval (CLIR) addresses the situation where the query that a user
presents to an IR system, is not in the same language as the corpus of documents they wish to search.
This situation presents a number of challenges (Grefenstette (1998)) but primary amongst these is the
problem of crossing the language barrier (Schauble & Sheridan (1997)). Almost all the approaches to this
problem require access to some form of rich translation resource to map terms in the query language (the
source) to terms in the corpus (the target). “Transitive” CLIR aims to address the situation where there
are limited direct translation resources available (Ballesteros (2000)).
A transitive CLIR system translates the source language terms by first translating the terms into
an intermediate or "pivot" language and then translating the resulting terms into the target language.
Thus, a transitive system could translate a query from German to English via either Dutch, or Spanish.
The main aim of this work is to combine translations from two different transitive routes to
discover if this can reduce the ambiguity introduced by transitive translation. Ballesteros suggested the
possibility of using this approach in the summary to her recent chapter (Ballesteros (2000)). We have
chosen to call this approach “lexical triangulation”. See figure 1 below.
Dutch Query Terms
Translate to Dutch Translate to English
Combine English Query
German Query English Terms
Translate to Spanish Translate to English
Spanish Query Terms
Figure 1 Lexical Triangulation
We have chosen to simulate a Machine-Readable Dictionary (MRD) approach to CLIR. This
follows on from the work of Ballesteros & Croft (1996, 1997, 1998), and Ballesteros (2000).
1
To whom all correspondence should be addressed. (e-mail: m.sanderson@sheffield.ac.uk )
2 The Experimental Environment
2.1 General
The underlying IR system used in the Sheffield submission was the GLASS system (Sanderson
(2000)).
The translation resources were derived from the German, Spanish, Dutch, and English
components of EuroWordNet (Vossen (1999)). The data used to lemmatise the German queries was
derived from the CELEX German databases.
2.2 EuroWordNet
Given that the intention of this work is to examine CLIR using simulated Machine Readable
Dictionaries, the choice of EuroWordNet (Vossen (1999)) as the primary translation resource may appear
a little strange. The primary basis for this choice was availability2.
The intention of the EuroWordNet project was to develop a database of WordNets for a number
of European languages similar to, and linked with, the Princeton WordNet 1.5 (Vossen (1997)). This
effectively makes English the inter lingua that all the other languages link through. One of the intended
uses of EuroWordNet was in multi-lingual information retrieval (Vossen (1997)). Gonzalo, et al. (1998)
describes a possible implementation.
By developing a series of WordNets for European languages, and linking them to the original
Princeton 1.5 WordNet for English, EuroWordNet has created a structure similar to the controlled
vocabulary thesaurus used by Salton as described by Oard & Dorr (1996). The structure is also very
similar to the structure developed by Diekema, et al. (1998). The Princeton WordNet consists of
synonyms grouped together to form “synsets”, basic semantic relationships link these together to form the
WordNet (Vossen (1997), Miller, et al. (2000)). Each synset has a unique identifier (synset-id).
In EuroWordNet, the relationships between the synsets of the various component languages and
the Princeton 1.5 WordNet synsets3 can take many forms. These include, for example, the eq_hyponym4
relation, which relates more general to more specific concepts (Vossen (1997)).
Our work used EuroWordNet to generate structures to simulate a Machine Readable Dictionary.
The only relationships used in the construction of the dictionary tables, were the eq_synonym and
eq_near_synonym relationships. These are by far the most restrictive and precise of the possible
relationships.
The eq_synonym relationship records the fact that the language synset is synonymous with the
WordNet synset. EuroWordNet introduced the eq_near_synonym relationship to record the fact that
certain terms that share a common hypernym (more general concept) are closer in meaning than others.
In this situation the co-hyponyms (more specific terms) that are closely related are close enough in
meaning that they could be used for translation purposes, but are not synonymous and are therefore not in
the same synset. This closeness is represented by linking the synsets with an eq_near_synonym
relationship (Vossen (1997)).
For each language used from EuroWordNet, two tables were generated. The first mapped
lemmas to the synset-ids of the synsets related by eq_synonym or eq_near_synonym. The second maps
synset-ids to their constituent lemmas (i.e. related by eq_synonym or eq_near_synonym). As we will
explain below, these tables are used to parameterise the translation process.
2
The Sheffield University Computer Science Department was a collaborator in the
EuroWordNet project and Wim Peters of that department kindly made extracts from EuroWordNet
available for this research.
3
In EuroWordNet terms the Inter Lingual Index or ILI.
4
The relationships in EuroWordNet have names on the form eq_relationship_name the eq_
indicates that the relationship involves some degree of “equality”.
2.3 The translation and processing of queries
Query processing was fully automatic and the queries were generated using all parts of the
topics. The queries were passed through a series of processes as follows:
• Parsing - The conversion of the topics to queries which makes use of title, description and narrative
fields.
• Normalisation - all characters were reduced to the lower case unaccented equivalents (i.e. “Ö”
reduced to “o” and “É” to “e” etc.) in order to maximise matching in both the lemmatisation and
translation processes.
• Lemmatisation - The various inflected forms of the query words were reduced to a canonical lemma
form to enable matching with the German EuroWordNet translation resources. A table derived from
the CELEX German database was used to determine the appropriate lemmata5 for a wordform.
German compound words were split using a simple algorithm. The algorithm looks for a series of
wordforms that will match with the whole compound. If such a complete match is found the
corresponding lemmata of the wordforms are returned. The algorithm takes account of the use of “s”
as “glue” in the construction of German compounds. This approach was based on the description of
the word reduction module in Sheridan & Ballerini (1996). All of the CELEX data was normalised
to unaccented lower case for matching with the query words.
• German Stop Word Removal - A stopword list, generated from the CELEX German database, was
used to remove words in the query that carried little meaning and would otherwise introduce noise to
the translation. The stop-word lists contain all of the German words marked as articles, pronouns,
prepositions, conjunctions or interjections in the CELEX database.
• Translation - The translation process used tables derived from EuroWordNet to translate between two
languages. The lemma to synset-id table for the first language and the synset to lemma table for the
second language were used to map words in the first language to words in the second. All the
possible translations through the intermediate synset-ids were returned. Three different translations
were created for each query: a direct German to English translation, a transitive translation using
Spanish as the intermediate language, and a transitive translation using Dutch as the intermediate
language.
• Merging - The results of the two transitive translation routes were merged to produce a fourth
translation, the triangulated translation. The merge process was conducted on an “original German
Lemma” by “original German Lemma” basis. The translations from each route for each lemma were
compared and only translations common to both routes were used to translate the lemma.
• Retrieval – the translation and merging process produced four different versions of the queries
translated into English, these were submitted to the GLASS IR system which had been used to index
the English corpus. The GLASS system normalised both documents and queries to lower case, and
removed any English stopwords (using a standard English stop word list). Porter stemming (Porter
(1980)) was used on both the queries and the collection. No special processing was used on the
corpus.
5
The wordform to lemma table is a many-to-many mapping as a wordform may be a valid
inflection of more than one lemma.
3 The Experimental story
3.1 Experimental Conditions
We submitted 4 official runs to the CLEF evaluation process.
• A “bilingual” run (shefbi), generated from the direct translation from German to English
• A “Spanish transitive” run (shefes), generated from the transitive translation using Spanish as the
intermediate.
• A “Dutch transitive” run (shefnl), generated from the transitive translation using Dutch as the
intermediate.
• And a “triangulated” run (sheftri), generated from the result of merging of the two transitive
translations.
Only the triangulated run (sheftri) was judged and contributed to the relevance judgement pool.
In order to provide a baseline for comparison we conducted an additional English monolingual
run using the same parsing and retrieval processes. This unofficial run is presented below to enable
comparisons to be made.
In summary then, the experimental conditions were as follows:
Experimental Variable Value for this experiment
Queries CLEF 2000 CLIR, German and English
Corpus LA Times 1994- CLEF Collection
Relevance Judgements CLEF 2000 pool
Corpus and Query Stemming Yes, Porter based
Lemmatiser Yes, including German Compound Splitting
German Stop-words removed Yes, all articles, pronouns, prepositions, conjunctions or
pre-translation interjections from the CELEX German database.
Translation Simulated Dictionary based, using lookup-tables derived from
EuroWordNet eq_synonym and eq_near_synonym relations.
Merging Strategy for Only translations common to both transitive routes,
Lexical triangulation
3.2 Results
The table below shows the average precision for the 5 runs that made up the CLEF experiment.
Only the cross language runs were submitted to the CLEF, and of those, only the triangulated run
contributed to the pooled results.
Porter, Intersection
English 0.3593
Bilingual (shefbi) 0.0856
Triangulated (sheftri) 0.0458
Spanish Transitive (shefes) 0.0098
Dutch Transitive (shefnl) 0.007
The standard 11-point recall and precision curves for the 5 runs are shown below, the second
graph shows only the 4 cross language runs.
0.9
English
0.8
0.7 Bilingual
0.6 Triangulated
0.5
Spanish
0.4 Transitive
0.3 Dutch
Transitive
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0.35
Bilingual
0.3
0.25 Triangulate
d
0.2
Spanish
0.15 Transitive
0.1 Dutch
Transitive
0.05
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
3.3 Analysis
Comparing the average precision of the monolingual run with the bilingual run we see that the
bilingual run is some 76%6 below the monolingual. This compares to the 60% below worst case reported
by Ballesteros & Croft (1996) when considering word by word dictionary based Spanish to English
CLIR.
Taking next the two transitive runs, we observe a differential of -88% in the case of the Spanish
transitive run and -92% in the case of the Dutch transitive run relative to the bilingual run. Both of these
results are statistically significant at the 0.01 level under both the sign and Wilcoxon tests. These figures
are in line with the -92% differentials reported by Ballesteros (2000) for transitive retrieval of Spanish –
French CLIR with English as the pivot compared to Spanish – French direct translation.
Comparing the triangulated run with the two transitive runs reveals the expected improvement in
performance. The differentials for the two transitive runs relative to the triangulated run are -79% for the
Spanish transitive run and -85% for the Dutch transitive. Both of these figures are statistically significant
at the 0.01 level under both the sign and Wilcoxon tests.
There is also a statistically significant differential of -47% between the triangulated run and the
bilingual in favour of the bilingual. This significance is at the 0.01level under both the sign and
Wilcoxon tests.
4 Conclusion
In summary, these results support the results of Ballesteros (2000) with respect to the behaviour
of transitive translation in CLIR. They also support the hypotheses we set out to prove that lexical
triangulation has the beneficial effect of improving the results from transitive translation in dictionary
based CLIR.
This work made use of relatively rich resources in the form of EuroWordNet. However, it
remains to be seen if these results could be repeated using the poorer quality resources that are likely to be
available for translating between less common pairs of languages.
As Samuel Johnson said “Dictionaries are like watches; the worst is better than none, and the
best cannot be expected to be quite true.” (Gendreyzig (2000))
6
Statistically significant at the 0.01 level under both the sign and Wilcoxon tests.
5 Bibliography
Ballesteros, L. & Croft, B. (1996). "Dictionary methods for cross-lingual information retrieval". In:
Database and Expert Systems Applications. 7th International Conference, DEXA '96
Proceedings. Springer-Verlag Berlin, Germany. [Online]. Available:
http://cobar.cs.umass.edu/pubfiles/ir-98.ps.gz [23/03/2000].
Ballesteros, L. & Croft, W. B. (1997). "Phrasal translation and query expansion techniques for cross-
language information retrieval". In: Proceedings of the 20th annual international ACM SIGIR
conference on Research and development in information retrieval, pp. 84 - 91. Association for
Computing Machinery. [Online]. Available:
http://www.acm.org/pubs/articles/proceedings/ir/258525/p84-ballesteros/p84-ballesteros.pdf
[29/02/2000].
Ballesteros, L. & Croft, W. B. (1998). "Resolving ambiguity for cross-language retrieval". In:
Proceedings of the 21st Annual International ACM SIGIR Conference on Research and
Development in Information Retrieval. Association for Computing Machinery. [Online].
Available: http://www.acm.org/pubs/articles/proceedings/ir/290941/p64-ballesteros/p64-
ballesteros.pdf [29/02/2000].
Ballesteros, L. A. (2000). "Cross Language Retrieval via transitive translation". In: Croft, W. B. (ed.)
Advances in Information Retrieval: Recent Research from the CIIR, pp. 203 - 234 Kulwer
Academic Publishers.
Diekema, A., Oroumchian, F., Sheridan, P. & Liddy, E. D. (1998). "TREC-7 Evaluation of Conceptual
Interlingua Document Retrieval (CINDOR) in English and French". In: Voorhees, E. M. &
Harman, D. K. (eds.), NIST Special Publication 500-242: The Seventh Text REtrieval
Conference (TREC-7). NIST. [Online]. Available:
http://trec.nist.gov/pubs/trec7/t7_proceedings.html [15/02/2000].
Gendreyzig, M. (2000). Collection of Web-Dictionaries, [Online]. LEO - Link Everything Online.
Available: http://dict.leo.org/dict/dictionaries.en.html [24/08/2000].
Gonzalo, J., Verdejo, F., Peters, C. & Calzolari, N. (1998). "Applying EuroWordNet to cross-language
text retrieval", Computers and the Humanities, 32( 2-3), pp 185-207
Grefenstette, G. (1998). "Problems and approaches to Cross Language Information Retrieval",
Proceedings of the Asis Annual Meeting, 35, pp 143-152
Miller, G. A., Chodorow , M., Fellbaum , C., Johnson-Laird, P., Tengi, R., Wakefield, P. & Ziskind, L.
(2000). WordNet - a Lexical Database for English, [Online]. Cognitive Science Laboratory,
Princeton University. Available: http://www.cogsci.princeton.edu/~wn/w3wn.html [23/08/2000].
Oard, D. W. & Dorr, B. J. (1996). A Survey of Multilingual Text Retrieval. (Report). Institute for
Advanced Computer Studies and Computer Science Department University of Maryland.
[Online]. Available: http://www.clis.umd.edu/dlrg/filter/papers/mlir.ps [09/03/2000]
Porter, M. F. (1980). "An algorithm for suffix stripping". In: Sparck Jones, K. & Willett, P. (eds.), (1997)
Readings in Information Retrieval., pp. 313 - 316. San Francisco: Morgan Kaufmann.
Sanderson, M. (2000). GLASS, [Online]. Dr Mark Sanderson. Available:
http://dis.shef.ac.uk/mark/GLASS/ [25/07/2000].
Schauble, P. & Sheridan, P. (1997). "Cross-Language Information Retrieval (CLIR) Track Overview".
In: Voorhees, E. M. & Harman, D. K. (eds.), NIST Special Publication 500-226: The Sixth Text
REtrieval Conference (TREC-6). NIST. [Online]. Available:
http://trec.nist.gov/pubs/trec6/t6_proceedings.html [15/02/2000].
Sheridan, P. & Ballerini, J. P. (1996). "Experiments in multilingual information retrieval using the
SPIDER system". In: Frei, H. P. (ed.) Proceedings of the 1996 19th Annual International ACM
SIGIR Conference on Research and Development in Information Retrieval, SIGIR 96, pp. 58 -
65. Association for Computing Machinery. [Online]. Available:
http://www.acm.org/pubs/articles/proceedings/ir/243199/p58-sheridan/p58-sheridan.pdf
[29/02/2000].
Vossen, P. (1997). "EuroWordNet: A Multilingual Database for Information Retrieval". In: THIRD
DELOS WORKSHOP Cross-Language Information Retrieval, pp. 85-94. European Research
Consortium For Informatics and Mathematics. [Online]. Available:
http://www.ercim.org/publication/ws-proceedings/DELOS3/Vossen.pdf [01/03/2000].
Vossen, P. (1999). EuroWordNet Building a multilingual database with wordnets for several European
languages., [Online]. University of Amsterdam. Available: http://www.hum.uva.nl/~ewn/
[28/02/2000].