A SURVEY ON CROSS LANGUAGE INFORMATION RETRIEVAL

International Journal on Cybernetics & Informatics (IJCI) Vol. 5, No. 1, February 2016
DOI: 10.5121/ijci.2016.5113 127
A SURVEY ON CROSS LANGUAGE INFORMATION
RETRIEVAL
Prof. Sanjay K.Dwivedi1
and Ganesh Chandra2
1-2
Department of Computer Science, BBAU (A Central University), Lucknow
ABSTRACT
Now a days, number of Web Users accessing information over Internet is increasing day by day. A huge
amount of information on Internet is available in different language that can be access by anybody at any
time. Information Retrieval (IR) deals with finding useful information from a large collection of
unstructured, structured and semi-structured data. Information Retrieval can be classified into different
classes such as monolingual information retrieval, cross language information retrieval and multilingual
information retrieval (MLIR) etc. In the current scenario, the diversity of information and language
barriers are the serious issues for communication and cultural exchange across the world. To solve such
barriers, cross language information retrieval (CLIR) system, are nowadays in strong demand. CLIR refers
to the information retrieval activities in which the query or documents may appear in different languages.
This paper takes an overview of the new application areas of CLIR and reviews the approaches used in the
process of CLIR research for query and document translation. Further, based on available literature, a
number of challenges and issues in CLIR have been identified and discussed.
KEYWORDS
Cross language information retrieval, query translation, document translation, corpora based translation,
dictionary based translation, machine translation
1. INTRODUCTION
Information retrieval (IR) generally refers to the process where the users search for required
information from a large number of documents. Traditional IR systems are implemented mainly
for monolingual documents. However, with rapid development of the Internet, the demand for
searching information from multi-lingual documents is increasing, which results in the great
challenge of how to match the users query written in one language with the documents written in
other languages. Therefore, suitable techniques are required to enhance the performance of IR,
CLIR & MLIR. The Cross Language Information Retrieval (CLIR) [1] provides a convenient
way that can solve the problems of language boundaries, where users can submit queries written
in their own language and retrieve documents in another language [2]. For example, a query in
Hindi will return the related information description in Chinese language, as given in figure1.
Figure.1 Cross Language Information Retrieval System

128
With the rapid development of Internet technology, globalization of information structure caused
the urgent demand to CLIR, because CLIR allows the usage of information exchanges between
different languages, remove linguistic disparity between the queries that are submitted and
documents that are retrieved using resource over the network, and also decreases the
communication cost [3]. The research on information retrieval, came into existence since early
1970s where as experiments for retrieving information across languages were first initiated by
Salton in 1973[67]. However, most of the modern research on CLIR started in 90‟s, and now a
day‟s it has become one of the most important research topic in the area of information retrieval.
An ever active research field, a vast number of researches and studies have been published on
CLIR and various issues are addressed in several evaluation forums such as TREC [64], NTCIR
[65], and CLEF [66] while each of them cover different languages: TREC includes Spanish,
Chinese, German, French, Arabic and Italian; NTCIR includes Japanese, Chinese and Korean and
CLEF includes French, German, Italian, Spanish, Dutch, Finnish, Swedish and Russian [4]. The
most effective way to solve the problem of language barriers may be achieved through CLIR by
using query translation approach, document translation approach, or by using both query and
document translation approach.
This paper is organised as follows: section 2, 3, 4 consists of query translation, document
translation and dual translation respectively. Section 5, 6, 7 describes the comparative study of the
three approaches, challenges of CLIR and CLIR tools respectively. Section 8 describes the
application areas of CLIR, section 9 describes CLIR researches in Foreign & Indian languages
and section 10 describes the conclusion and future of CLIR.
2. QUERY TRANSLATION APPROACH
A major challenge in CLIR is to bridge the language gap between query and documents. Query
translation is now serving as a major cross-lingual mechanism in current CLIR systems [5], [40],
[70] as shown in figure2. CLIR search engines enable users to retrieve content in a language
different from language used to formulate the query. Translation of query has the advantage that
the computational effort i.e. time and space, is less as compared with other methods. Query
translation has following disadvantages:
(i) Usually a query does not provide enough contexts to automatically find the intended meaning
of each term in the query.
(ii) Translation errors affect retrieval performance sensibly.
(iii) In case of searching a multilingual database, query has to be translated into each one of
the languages of database.
In CLIR query translation play an important role that can be achieved by following approaches:
dictionary based translation approach, corpora based translation approach and machine translation
based approach.
2.1. Dictionary Based Translation Approach
In dictionary-based query translation, the query will be processed linguistically and only
keywords are translating using Machine Readable Dictionaries (MRD), given in figure3. MRDs
are electronic versions of printed dictionaries, either in general domain or specific domain. The
use of existing linguistics resources, especially the MRDs, is a natural approach to cross-lingual
IR. Translating the query using the dictionaries is much faster and simpler than translating the
documents [6], [7], [41]. Some common problems associated with dictionary-based translation [8]
are:

129
(i) Untranslatable words (like new compound words, proper names, spelling variants, and
special terms): Not every form of words used in query is always found in dictionary. Some time
problem occurs in translating different compound words (formed by combination of new words)
due to the unavailability of their proper translation in dictionary [42].
Figure 2: Query Translation
(ii) Processing of inflected words: Inflected word forms are usually not found in dictionaries [43].
(iii) Lexical ambiguity in source and target languages: Relevant forms of lexical meaning for
information retrieval are: 1) homonymous and 2) polysemous words. Two words are
homonymous; if they have at least two different meanings and senses of words are unrelated e.g.
bank (river bank) and bank (financial institution). Polysemous words should have related senses
e.g. star in the sky and star. Due to ambiguity in the search keys, matching for retrieving relevant
documents may not be successful [46].
Figure3.Dictionary Based Translation
2.2. Corpora Based Translation Approach
Query translation using corpora requires single corpus or many corpuses. Corpora, (plural of
corpus) are the systematic collection of naturally occurring language material, such as texts,
paragraphs and sentences from one or many languages. In corpus-based methods [9], [12] queries
are translated on the basis of multilingual terms extracted from parallel or comparable document
collections. A parallel corpus has been used since the early 1990‟s for translation of given word.
A parallel corpus is a collection of texts, each of which is translated into one or more languages
other than the original language. Parallel corpora are also used to decide the relationships, such as
co-occurrences, between terms of different languages. A parallel corpus is an important kind of

130
source of linguistic meta-knowledge, which forms the basis of techniques such as tokenization,
morphological and syntactic analysis [15], [10].
A comparable corpus is one of the important concepts in corpus-based translation study,
introduced by Baker [38]. Comparable corpora contain text in more than one language. The texts
in each language are not translations of each other, but cover the same topic area, and hence
contain an equivalent vocabulary. A good example of corpora is the multilingual news feeds
produced by news agencies such as Reuters, CNN, BBC, Xinhua News and BERNAMA. Such
texts are widely available on the Web for many language pairs and domains. They often contain
many sentence pair that are fairly good translations of each other [11], [12] [13].
2.3. Machine Translation Based Approach
Cross-lingual IR with query translation using machine translation [6] seems to be an obvious
choice compared to the other two above, as shown in figure4. The advantages of using the
machine translation is that it saves time while translating large texts. Manning and Schutze [16]
distinguished four different approaches to deal with machine translation: (a) Word-for-word
approach, (b) Syntactic transfer approach, (c) Semantic transfer approach, and (d) Interlingual
approach. The ultimate goal of CLIR machine translation (MT) systems is to translate queries
from one language to another by using a context. Many factors contribute the difficulties of
machine translation, including words with multiple meanings, sentences with multiple
grammatical structures, uncertainty about what a pronoun refers to, and other problems of
grammar.
Many researchers criticize MT-based CLIR approach. The reasons behind their criticisms mostly
stem from the fact that the current translation quality of MT is poor. Another reason is that MT
systems are expensive to develop and their application degrades the retrieval efficiency (run time
performance) due to the lengthy processing times associated with linguistic analysis.
Figure4. Machine Translation
MT based approaches seems to be the ideal solution for CLIR. It is mainly because MT systems
translate the sentence as a whole, and the translation ambiguity problem is solved during the
analysis of the source sentence. Table 1, describes the differences between various techniques of
query translation.
Table1: Comparison between Techniques of Query Translation
Parameters Dictionary Based
Translation Approach
Corpora Based
Translation Approach
Machine
Translation Based
Approach
Ambiguity High Low Low
Offline
Translation
Possible Possible Not possible
Working
Architecture
Visible as like white box
testing.
Visible as like white box
testing.
Works similar to
black box testing
Development Less expansive More expensive than More Expensive

131
expenses DBT
Translation
Availability
Highly available in many
languages
Available only in few
languages
Available only in
few languages
3. DOCUMENT TRANSLATION APPROACH
Document Translation [44], [45] can be the most desirable scenario in CLIR, if the purpose is to
allow the users to search the documents different from their own language and receive results
back in user‟s language, as given in figure5. In this sense, it is truly a better option which does not
require a passive knowledge of the foreign language from the user. In document translation
approach, all target languages are translated to the source language [63]. The function of this
translation is twofold. First, post translation or „as-and-when-needed‟ or „on-the–fly translation‟,
where documents of any other language being searched by user are translated into user language
at query time. IR process mostly uses indexing technique to speed up the searching process of
documents. But indexing is not possible in post translation, so this approach is infeasible because
it requires more time for translation.
Second, pre translation or „all together before any query is processed‟ used to browse through a
translated version of an original translation in user language or in a language which user can
understand (figure 5). This translation can be called as offline translation. In this approach,
documents that are written in different languages are translated to all desired source languages
and these documents are indexed before query time. This translation is impossible as a solution
for large collection of distributed documents, which are managed by different groups of people,
for example internet. Document translation has its own advantages and disadvantages compared
to query translation. Some researchers have used it to translate large sets of documents (e.g.,
Braschler & Schauble, 2001 [17]; Franz, Scott McCarley, & Todd Ward, 2000 [18]; Oard &
Hackett, 1998 [19]) since more varied context within each document is available for translation,
which can improve translation quality. The document translation approach has certain benefit
over query translation. These include the following:
(i) A long document provides more contexts to perform translation, so that terms in the target
language can be chosen more accurately.
(ii) Translations errors should not harm retrieval too much, as they are weighted against a whole
document.
(iii) The translation effort is done at indexing time, thus getting faster retrieval at run time.
However, there are certain issues with document translation as well, such as:
(i) Much more computational effort is needed to index collections.
(ii) Bad scaling performed in case of more than two languages.

132
Figure5. Document translation
4. DUAL TRANSLATION (BOTH QUERY AND DOCUMENT TRANSLATION
APPROACH)
In this approach – both queries and documents are translated into a common representation
(figure 6). This approach requires additional storage space for translated documents but provides
scalability when same collection of documents is required in multiple languages. One of the
examples of such approach is controlled vocabulary systems [20]. These systems represent all
documents using a pre-defined list of language-independent concepts, and enforce queries in the
same concept space. This concept space defines the granularity or precision of possible searching.
The major issue of controlled vocabulary systems is that, non-expert users usually require some
training and also require interfaces to the vocabularies in orderable to generate effective queries.
Dual translation approach also called as hybrid translation approach can be performed by pivot
language. Direct translation between two languages may not always be possible due to the
limitation of translation resources. To perform such type of translation, a resources or a third
language is required between these languages, called pivot language. In this process, two types of
approaches are possible: either the query or the document is translated first to pivot language,
then to the target language; translate both document and query into pivot language as shown in
figure 6.
Figure 6. Dual Translation (Pivot Language)

133
5. COMPARATIVE STUDY OF THE THREE APPROACHES
The need for translation has itself been questioned because non-translation based methods of
CLIR, such as cognate-matching [21] and cross-language Latent Semantic Indexing [22] have
been developed. Document translation into query language or query translation into documents
language are the two approaches that couples machines translation and information retrieval.
Query translation and document translation approaches are neither equivalent nor mutually
exclusive. They are not equivalent because machine translation is not an invertible operation.
Query translation and document translation become equivalent only if each word in one language
is translated into a unique word in other languages.
Various researches suggest that document translation should be competitive or superior to query
translation. Typical queries are short and may contain key words or phrases only. When these are
translated inappropriately, the IR engine has no chance to recover. Translating a long document,
MT engine offers the many more opportunities to translate key words and phrases. If some of
these are translated inappropriately, the IR engine has at least a chance of matching these to query
terms. Query translation approach is flexible and allows for more interactions with the user.
However, query translation often suffers from the problem of translation ambiguity, and this
problem is amplified due to the limited amount of context in short queries. From this perspective,
document translation seems to be more capable of producing more precise translation due to
richer contexts.
One of the critical aspects of document translation approach is that one has to determine in
advance to which language each document should be translated and that all the translated versions
of the document should be stored. In a multilingual IR environment, one would desire to translate
each document to all other languages. This is impracticable because of the multiplication of
document versions and the increase in storage requirement. Once a document is pre-translated
into the same language as the query, user can directly read and understand the translated version.
Otherwise, a post-retrieval translation is often needed to make the retrieved documents readable
by the user (if he/she does not understand the document language).
Query translation and document translation become equivalent only if each word in one language
is translated into a unique word in the other languages. Document translation can be performed
off-line and on-line but query translation is performed only on-line. Hybrid system that uses both
query and document translation are possible because of a trade off between computer resources
and quality of translation. Hybrid or dual translation approach provides the relationship between
multilingual and the key advantages of these systems are that queries can be expressed and
matched unambiguously. In this approach the additional storage space requirement is independent
to the number of languages supported. The major problems occurs in this approach are to define
the concept space, intermediate representation and conversion of documents into intermediate
representation. Differences between two approaches (query translation and documents
translation) of CLIR are described in table2. Table3 describes the comparative study of three
approaches of CLIR.
Table2: Difference between Query and Document Translation
Parameter Query Translation Document Translation
Size Small Large
Language Prior knowledge of translation language is
not required
Prior knowledge of translation
language is required
Overhead Low High
Recovery When these are translated inappropriately,
the IR engine has no chance to recover
Chance to recover
Ambiguity Maximum chances of occurring ambiguity Minimum chances of occurring

134
ambiguity
Cost Low cost High cost
Table 3: Comparison of three Translation Approaches
Parameter Query
Translation
Document Translation Both Query & Document
Translation
Ambiguity Maximum Minimum More than both
Additional
Storage Space
Not required Required Not required
Translation time Less More than query More than both
Information
retrieval
Bilingual Bilingual Bilingual and
Multilingual
Flexibility Highly Less Less
Working nature Can provide
interface between
two language at a
time
Can provide interface
between two language at
a time
Can provide interface
between more than two
language at a time
6. CHALLENGES IN CLIR
Queries from users are often too short, which produce more ambiguity in query translation, and
reduce the accuracy of the cross language retrieval results. Since the problem of language
mismatch in CLIR are more serious than in monolingual IR, it is necessary to exploit techniques
for improving the multilingual retrieval performance. In CLIR systems, users often present their
query in their native language, and then the system automatically searches documents written in
other languages. Therefore, it is a challenge for CLIR to conquer the barrier between the source
language (SL) in query sentences and the target language (TL) in documents to be searched. As
discussed in the previous section, most CLIR systems utilize MT technology to resolve this
problem. As MT research itself has a number of issues (such as accuracy), the research in CLIR
also faces critical issues and challenges that must be addressed.
6.1. Ambiguity
Ambiguity [15] occurs when words have multiple meaning which also referred to as homonymy
or polysemy. Ambiguity in IR are semantic and syntactic in nature, where as ambiguity in CLIR
are semantic and lexical. So the probability of occurrence of ambiguity in CLIR is higher than
normal IR, due to the availability of different languages [39].
6.2. Effective User Feedback
Effective user functionality can be incorporated by the user feedback, about their requirements
and information needs. It should also provide readable translations of the retrieved documents to
support document selection. System should also provide better support for query formulation and
reformulation based on some set of intermediate results.
6.3. Complexity in New Applications
Question/Answering is relatively a new stream of Information Retrieval. In Question/Answering
end-users throw a question in a form of query and retrieve answers for that. However, challenge
is to retrieve answers of English questions in different language.

135
6.4. Specialized Terminology and Proper Nouns
Specialized terminology, such as scientific names, is often difficult to translate and is often found
in specialized dictionaries or term banks. Specialized terminology tends to be less ambiguous
than regular vocabulary although regular vocabulary can have a specialized meaning when used
in a certain subject area.
7. CLIR RESEARCHES IN FOREIGN & INDIAN LANGUAGES
India is a multilingual country, with 22 constitutionally recognized languages and 12 scripts. A
huge amount of information on Web is available in English. In view of the fact that a small
number of people know English in India, others are familiar with Hindi and other local languages.
In order to solve the language gap between people of India, many government [14] and private
organisations are working on CLIR research field. Some of the prominent organisations working
in CLIR [47] field are:
(i) Indian Institute of Technology, Bombay, Maharashtra, India.
(ii) Indian Institute of Technology, Kharagpur, West Bengal, India.
(iii) Indian Institute of Technology, Hyderabad, Telangana, India.
(iv) Anna University- K B Chandrasekhar (AUC-KBC), research centre, Chennai, India.
(v) Indian Statistical Institute, Kolkata, West Bengal, India.
(vi) Jadavpur University, Kolkata, West Bengal, India.
(vii) Centre for Development of Advanced Computing (C-DAC), Pune, India.
(viii) Centre for Development of Advanced Computing (C-DAC), Noida, UP, India.
(ix) Utkal University Bhubaneswar and STDC, Orissa, India.
(x) Department of Information Technology (DIT), New Delhi.
(xi) Microsoft research centre, Bangalore, Karnataka, India.
(xii) Technology Development for Indian Langauge (TDIL), government of India.
The objective of these organisations is to help any user to give a query in one Indian language,
can access documents in any other language. Some Foreign organisation that working on CLIR
are as follows:
(i) Centre for Intelligent Information Retrieval (CIIR), School of Computer Science, University
of Massachusetts Amherst.
(ii) National Institute of Standard and Technology (NIST), U.S. department of research.
(iii) Johns Hopkins University, Baltimore, Maryland.
Some of the prominent researches in Foreign & Indian languages are described in table 4 and
table5 respectively as follows:
Table 4: Prominent CLIR research in foreign languages
Languages Authors Year
English and French David A. Hull, Gregory Grefenstette [48] 1996
Japanese and English Fujii, A., Ishikawa, T.,[49] 2001
English and Chinese Su Liu [50] 2001
Japanese, English and
Chinese
Wen- Cheng Lin Hsin- Hsi Chen [51] 2003
Greek, Latin and Old Norse Jeffrey A, Rydberg-Cox, Lara Vetter, Stefan M,
Ruger Daniel Heesch [52]
2004
Chinese, Japanese and Korea Chen-Yu Su, Tien-Chien Lin , Shih- Hung Wu [53] 2007
English and Chinese Hsin-Chang Yang, Chung-Hong Lee [54] 2008

136
Table 5: A brief description of some CLIR research in Indian languages
Languages Name of Researcher’s Year
Hindi, Telugu, English Prasad Pingali,Vasudeva Verma [56] 2006
Hindi,Tamil, Telugu, Bengali,
Malyalam and English
Jagdeesh, J. , Kumaran, K. [57] 2007
Telugu, English Ranbeer Makin, Nikita Pandey, Prasad Pingali,
Vasudeva Verma [58]
2007
English, Hindi Anurag Seetha, Sujoy Das, M. Kumar [59] 2007
Hindi, English Sethuramalingam S, Vasudeva Verma [60] 2008
English, Hindi, Telugu, Tamil Manoj Kumar Chinnakotla, Om P. Damani [61] 2009
English, Telugu, Tamil P. Sujatha, P. Dhavachelvan, V. Narasimhulu [62] 2010
Tamil to English S.Saraswathi & A. Siddhiqaa [55] 2010
8. CLIR TOOLS
Over the past few years, research in CLIR has progressed and a many systems have been
developed. Some of the prominent systems of CLIR are as follows:
8.1 SAPHIRE
The architecture of SAPHIRE system is based on multilingual aspects of UMLS (Unified Medical
Language System). In this system a dictionary based approach of CLIR is used [27]. It provides
an intelligent healthcare monitoring architecture for high quality health care services with
reasonable cost.
8.2. KEIZAI
KEIZAI [24] was developed at New Mexico State University and its aim is to provide the Web-
based cross language text retrieval system, which searches the documents of Korean or Japanese
language on the web for English query. Keizai examines the effectiveness of representing the
retrieved documents together with small images, which are called Document Thumbnail
Visualizations. The advantage of visualization is to improve the recall and efficiency.
8.3. MULINEX
MULINEX system [26] was developed at German Research Center for Artificial Intelligence
(DFKI), whose aim is to allow the user to search the collection of multilingual document,
supported by an effective combination of linguistic and IR technologies. There are three
document categorisation algorithms are used in Mulinex for different tasks: n-gram categoriser
for noisy input, k-nearest-neighbour (KNN) algorithm for normal documents, and pattern-based
categoriser for every short documents.
8.4. MIRACLE
MIRACLE (Maryland Interactive Retrieval Advanced Cross-Language Engine), deals with a
combination of statistical and linguistic resources, for monolingual, cross-lingual and multilingual
search. In MIRACLE two types of query translation are performed: fully automatic query
translation and user assistant query translation [25].

137
8.5. UCLIR
The core process of UCLIR (Unicode Cross-language Information Retrieval system) includes
machine translation and standard monolingual information retrieval, which accepts the query in
one language and retrieves relevant documents in other language. The UCLIR retrieval system is
based on URSA (Unicode Retrieval System Architecture), which is a high-performance text
retrieval system that can index and retrieve Unicode texts [28].
8.6. KANSHIN
KANSHIN collects and analyzes the multilingual articles of Japanese, Chinese, Korean and
English languages [23]. The system provides a various view points for user such as temporal,
focal, geographical, and network. It also provides a cross-lingual keyword navigation tool
between splog survey tools (called SplogExplorer) and inters language links of Wikipedia.
9. APPLICATION AREAS OF CLIR
The core field of information retrieval where research on CLIR is needed for effective results are:
9.1. Medical
A number of resources available on Web provide the public and healthcare professionals with the
most up-to-date findings in medical research, such as PubMed [68] and MedlinePlus [69].
Table6: List of some prominent CLIR tools from 1990‟s
Name Language Developed At Year
SAPHIRE
[27]
English, German,
French, Russian,
Spanish, and
Portuguese.
School of Medicine Oregon Health Sciences
University Portland, OR, USA by William R.
Hersh, M.D., Laurence C. Donohoe, M.L.I.S
1998
KEIZAI [24] English , Japanese
& Korean
Computer Research Lab New Mexico State
University, Las Cruces USA by William
Ogden & James Cowie et.al
1999
MULINEX[26] French, German
and English
German research Centre for artificial Intelligence
by Capstick et.al 2000
MIRACLE
[25]
English , French,
German, Spanish,
Cebuano and Hindi
Spanish University (Universal Politecnica de
Madrid, Universidad Autonoma de Madrid &
Universidad CarlosIII de Madrid) by Julio, Sara
et.al
2003
UCLIR
[28]
English , Japanese
Korean, Arabic
Computing Research Laboratory at New Mexico
State University by Ahmed Abdelali et.al
2004
KANSHIN
[23]
Japanese, Chinese ,
Korean & English
University of Tokyo Japan by Tomohiro
Fukuhara et.al.
2005
Medline Plus is a Web-based consumer health information resource, made available by the
National Library of Medicine. PubMed first released in 1996, is a free search engine for accessing
the Medline database of life sciences and biomedical topics.
Most of the high level quality resources that are freely available and unlimited for users all
around the world are available only in English language. Therefore Non-English users encounter

138
a great language barrier when trying to access medical information from these websites such are
also not familiar with medical terminology even in their first language (native language). So there
is a big platform for researcher to work on medical information retrieval system, in order solve
the problem of language barrier [29].
9.2. Multimedia
Multimedia Information Retrieval (MMIR or MIR) is a hot research discipline whose objective is
to extract the semantic information from multimedia data sources such as audio, video, and image
[1]. MMIR implies that multiple channels are employed for the understanding of media content,
each of these channels are described by media-specific features transformations [30].
The first version of the Multilingual Multimedia Information Retrieval (MMIR) prototype
involves short videos in the domain of news, that are selected from online web TV channels, from
UGC portals, or from online news agencies [31]. There has so far been very little work in the area
of Cross-Language Multimedia Information Retrieval (CLMIR). This is an important future
research topic as the growth of multilingual and multimedia document collections is likely to lead
inevitably to the growth of multilingual multimedia collections [32], [33].
9.3. Mobile Network
This research proposes a Cross-Lingual Information Retrieval approach that is used to search
Internet resources for appropriate content and summarize it into another form using the content
specification meta-language. This content is then mapped to the target language [34].
9.4. Video Question/Answering system
Question/Answering on multi-media is a new research issue in recent years. The cross-language
QA system have some fundamental problems like video processing, i.e. video Optical Character
Recognition (OCR) [32], [35], [36] and video segmentation.
9.5. Enterprise Competition
Along with the economic globalization, the information resource in a modern society becomes an
important element for modern enterprises competition. CLIR is introduced to the enterprise
competitive intelligence collections can effectively resolve the low recall and veracity rate of
intelligence collections to some extent and promote the development of CLIR in the enterprise
competition intelligence [37].
10. CONCLUSION AND FUTURE OF CLIR
Cross-lingual IR provides new mirror in searching documents through multitude varieties of
languages across the world and it can be the baseline for searching not only between two
languages but also in multiple languages. Today, most of the cross-lingual researches involved
only few famous languages like English, Hindi, Spanish, China and French. Research on
languages has increases the development of country. As the world becomes more connected by
technology, cross language IR in every language is needed. CLIR is a multidisciplinary area that
has been increasingly gaining more attention from the research community. Despite recent
advances and new developments, there are still many aspects to be explored.

139
In Indian context, which is one of the hotspots of linguistic diversity (350 languages) in the globe,
and the fact that a dominant language of one region may be a language of a linguistic minority in
other region, cross languages information retrieval systems would play a very important role in
allowing the people to go through the documents and literatures of other languages thus breaking
the language barrier. We work out here to give a broad overview of the speedy demanding work
in the field of CLIR by exploring its aspiration, difficulties, basic tools, major works and future
research goals. In reviewing this information, it becomes possible to gain a larger picture of the
CLIR field.
REFERENCES
[1] Gaillard, B., Bouraoui, J. L., De Neef, E. G., & Boualem, M., (2010, May), “Query expansion for
cross language information retrieval improvement”, In Research Challenges in Information Science
(RCIS), 2010 Fourth International Conference on (pp. 337-342). IEEE.
[2] Pigur, V. A., (1979, “Multilanguage information-retrieval systems: Integration levels and language
support”, Automatic Documentation and Mathematical Linguistics, 13(1), 36-46.
[3] Peng, Qu, Lu, Li & Zhang lili, (2008) “A Review of Advanced Topics in Information Retrieval”,
Library and Information Service, Vol.52.No 3. pp. 19-23.
[4] Ahmed, F., & Nurnberger, A., (2012), “Literature review of interactive cross language information
retrieval tools”, Int. Arab J. Inf. Technol., 9(5), 479-486.
[5] Narasimha Raju, B. N. V., Bhadri Raju, M. S. V. S., & Satyanarayana, K. V. V. (2014, December).
Translation approaches in Cross Language Information Retrieval. In Computer and Communications
Technologies (ICCCT), 2014 International Conference on IEEE, pp. 1-4.
[6] Carley Mc & S.J. (1999, June), “Should we translate the documents or the queries in cross-language
information retrieval?” In Proceedings of the 37th annual meeting of the Association for
Computational Linguistics on Computational Linguistics. Association for Computational Linguistics,
pp. 208-214.
[7] Aljlayl, M., & Frieder, O., (2001, October), “Effective arabic-english cross-language information
retrieval via machine-readable dictionaries and machine translation”, In Proceedings of the tenth
international conference on Information and knowledge management ACM, pp. 295-302.
[8] Pirkola, A., Hedlund, T., Keskustalo, H., & Jarvelin, K., (2001), “Dictionary-Based Cross-Language
Information Retrieval: Problems, Methods, and Research Findings,” Information Retrieval, 4(3/4), pp.
209-230.
[9] Picchi, E. & Peters, C, (2000) “Cross-language information retrieval: a system for comparable corpus
querying, in Cross-Language Information Retrieval,” G. Grefenstette, Editor. 2000, Kluwer Academic
Publishing: Massachusetts. pp. 81-90.
[10] Manning, C. D., Raghavan, P., & Schütze, H. (2008). Introduction to information retrieval (Vol. 1, p.
496). Cambridge: Cambridge University Press.
[11] Munteanu, D.S. and D. Marcu. (2005) “Extracting parallel subsentential fragments from non-parallel
corpora”, In Proceedings of the 21st international Conference on Computational Linguistics. Sydney,
Australia: Association of Computational Linguistics.
[12] Landauer, T.K. & Littman, M.L.,(1990) “Full Automatic Cross Language Document Retrieval using
Latent Semantic Indexing” , In Proc. of the 6th Conference of UW Center for New OED and Text
Research, pp. 31–38.
[13] Sheridan, P., & Ballerini, J.P., (1996) “Experiments in Multilingual Information Retrieval using the
SPIDER System”, In Proc. of the 19th Annual International ACM SIGIR, pp. 58–65.
[14]Sourabh, K., (2013) “An Extensive Literature Review on CLIR and MT activities in India”,
International Journal of Scientific & Engineering Research.
[15] Chandra, G.; Dwivedi, S.K., “A Literature Survey on Various Approaches of Word Sense
Disambiguation,” in Computational and Business Intelligence (ISCBI), 2014 2nd International
Symposium on , vol., no., pp.106-109, 7-8 Dec. 2014.
[16] Manning, C.D., & Schutze, H., (1999) Foundations of statistical natural language processing. MIT
Press.
[17] Braschler, M., & Schauble, P., (2000), “Using corpus-based approaches in a system for multilingual
information retrieval.” Information Retrieval, 3, pp. 273–284.

140
[18] Franz, M., Scott McCarley, J., & Todd Ward, R. (2000), Ad hoc, cross-language and spoken
document information retrieval at IBM. In Proceedings of TREC-8. Gaithersburg, MD: National
Institute of Standards and Technology. Available: http://trec.nist.gov/pubs/.
[19] Oard, D. W., & Diekema, A. R., (1998) “Cross-language information retrieval”, Annual Review of
Information Science and Technology, 33, pp. 223–256.
[20] Oard D and Dorr B. (1996), “A Survey of Multilingual Text Retrieval”, Technical Report UMIACS-
TR-96-19, University of Maryland, Institute for Advanced Computer Studies.
[21] Buckley, M. Mitra, Wals, J. & Cardie, C., (1998), “Using clustering and superconcepts within
SMART: TREC-6”, In E.M. Voorhees and D.K. Harman, editors, The 6th Text REtrieval Conference
(TREC-6).
[22] Dumais, S. T., Letsche, T. A., Littman, M. L., & Landauer, T. K., (1997, March) “Automatic cross-
language retrieval using latent semantic indexing”, In AAAI spring symposium on cross-language
text and speech retrieval, Vol. 15,pp. 21.
[23] Fukuhara, T., Kimura, A., Arai, Y., Yoshinaka, T., Masuda, H., Utsuro, T., & Nakagawa, H., (2008,
April) “KANSHIN: A cross-lingual concern analysis system using multilingual blog articles”, In
Information-Explosion and Next Generation Search, 2008. INGS'08. International Workshop on
IEEE, pp. 83-90.
[24] Ogden, W., Cowie, J., Davis, M., Ludovik, E., Nirenburg, S., Molina-Salgado, H., & Sharples, N.
(1999, September). Keizai: An interactive cross-language text retrieval system. In Proceeding of the
MT SUMMIT VII workshop on machine translation for cross language information retrieval, Vol.
416.
[25] He, D., Oard, D. W., Wang, J., Luo, J., Demner-Fushman, D., Darwish, K., & Leuski, A., (2003),
“Making miracles: Interactive translingual search for cebuano and hindi”, ACM Transactions on
Asian Language Information Processing (TALIP), 2(3), 219-244.
[26] De Luca, Ernesto William, Ernesto William, Stefan Hauke De Luca, Andreas Nürnberger, and Stefan
Schlechtweg., (2006) “MultiLexExplorer-Combining Multilingual web Search with Multilingual
Lexical Resources”, In Proceedings of the combined workshop on Language-Enabled educational
technology and Development and evaluation of Robust Spoken Dialogue Systems, Germany, pp. 17-
21.
[27] Hersh, William R., and Larry C. Donohoe, (1998) “SAPHIRE International: a tool for cross-language
information retrieval.”, Proceedings of the AMIA Symposium. American Medical Informatics
Association, pp673.
[28] Abdelali, A., Cowie, J. R., Farwell, D., & Ogden, W. C., (2004) “UCLIR: a Multilingual Information
Retrieval Tool”, Inteligencia Artificial, Revista Iberoamericana de Inteligencia Artificial, 8(22),
pp103-110.
[29] Lu, Wen-Hsiang, et al. (2008), “Using Web resources to construct multilingual medical thesaurus for
cross-language medical information retrieval.” Decision Support Systems 45.3: 585-595.
[30] Arrieta, K., Leturia, I., Iturraspe, U., de Ilarraza, A. D., Sarasola, K., Hernáez, I., & Navas, E. (2008,
December). AnHitz, development and integration of language, speech and visual technologies for
Basque. In Universal Communication, 2008. ISUC'08. Second International Symposium on IEEE, pp.
338-343.
[31] Wu, Y. C., Chang, C. H., & Lee, Y. S., (2004, December) “CLVQ: Cross-language video
question/answering system”, In Multimedia Software Engineering, 2004. Proceedings. IEEE Sixth
International Symposium on IEEE, pp. 294-301.
[32] Lin, C. J., Liu, C. C., & Chen, H. H., (2001) “A simple method for Chinese video OCR and its
application to question answering”, Computational linguistics and Chinese language processing, 6(2),
pp11-30.
[33]Wu, V., Manmatha, R., & Riseman, E. M., (1999) “Textfinder: An automatic system to detect and
recognize text in images”, IEEE Transactions on Pattern Analysis & Machine Intelligence, (11),
pp1224-1229.
[34] Kumar, B. A. (2012, January). Profound survey on cross language information retrieval methods
(CLIR). In Advanced Computing & Communication Technologies (ACCT), 2012 Second
International Conference on IEEE, pp 64-68.
[35] Shim, J. C., Dorai, C., & Bolle, R., (1998, August) “Automatic text extraction from video for content-
based annotation and retrieval”, In Pattern Recognition, 1998. Proceedings. Fourteenth International
Conference on IEEE, Vol. 1, pp. 618-620.

141
[36] Yang, H., Chaisorn, L., Zhao, Y., Neo, S. Y., & Chua, T. S. (2003, November). VideoQA: question
answering on news video. In Proceedings of the eleventh ACM international conference on
Multimedia ACM, pp. 632-641.
[37] Xiang-jun, L., Hong-Na, S., Li-Wen, Y., & Xiao-Kun, Y. (2009, May), “Research of Enterprise
Competitive Intelligence Collection System Based on Cross-Language Information Retrieval”, In
Electronic Commerce and Security, 2009. ISECS'09. Second International Symposium on IEEE, Vol.
1, pp. 601-604.
[38] Fernandes, Lincoln Paulo. (2006), “Corpora in Translation Studies: revisiting Baker's tipology.”
Fragmentos: Revista de Língua e Literatura Estrangeiras, 30.
[39] Diekema, Anne R., (2003) “Translation events in cross-language information retrieval: lexical
ambiguity, lexical holes, vocabulary mismatch, and correct translations”.
[40] Wu, D., & He, D., (2010, December) “A study of query translation using google machine translation
system”, In Computational Intelligence and Software Engineering (CISE), 2010 International
Conference on IEEE, pp. 1-4.
[41] Seetha, A., Das, S., & Kumar, M., (2007, December) “Evaluation of the English-Hindi Cross
Language Information Retrieval System Based on Dictionary Based Query Translation Method”, In
Information Technology,(ICIT 2007). 10th International Conference on IEEE, pp. 56-61.
[42] Pfeifer, U., Poersch, T., & Fuhr, N. (1996) “Retrieval effectiveness of proper name search methods”,
Information Processing & Management, 32(6), pp 667-679.
[43] Fluhr, C., Schmit, D., Ortet, P., Elkateb, F., Gurtner, K., & Radwan, K. (1998). Distributed cross-
lingual information retrieval. In Cross-Language Information Retrieval, Springer US, pp. 41-50).
[44] Croft, W. B., Turtle, H. R., & Lewis, D. D. (1991, September), “The use of phrases and structured
queries in information retrieval”, In Proceedings of the 14th annual international ACM SIGIR
conference on Research and development in information retrieval, ACM, pp. 32-45.
[45] Buckley, C., Singhal, A., Mitra, M., & Salton, G., (1995, November), “New retrieval approaches
using SMART: TREC 4”, In Proceedings of the Fourth Text Retrieval Conference (TREC-4), pp. 25-
48.
[46] Lyons, J, (1981), “Language and Linguistics: An introduction”, Cambridge University Press.
[47] Nagarathinam, A., & Saraswathi, S. (2011), “State of Art: Cross Lingual Information Retrieval
System for Indian Languages”, International Journal of Computer Applications, 35 (13).
[48] Hull, D. A., & Grefenstette, G., (1996, August), “Querying Across Languages: A Dictionary-Based
Approach to Multilingual Information Retrieval”, In Proceedings of the 19th annual international
ACM SIGIR conference on Research and development in information retrieval ACM, pp. 49-57.
[49] Fujii, A., & Ishikawa, T., (2001, March) “Evaluating multi-lingual information retrieval and
clustering at ULIS”, In Proceedings of the Second NTCIR Workshop Meeting on Evaluation of
Chinese & Japanese Text Retrieval and Text Summarization. Tokyo. Japan, pp. 5-144.
[50] Su Liu.: ECIRS, (2001), “An English-Chinese Cross-language Information-retrieval System”, In:
IEEE International Conference on Systems, Man, and Cybernetics, Volume 2.
[51] Lin, W. C., & Chen, H. H., (2003) “Description of NTU approach to NTCIR3 multilingual
information retrieval”, In Proceedings of the third NTCIR workshop.
[52] Jeffrey A. Rydberg-Cox, Lara Vetter, Stefan M. Ruger, Daniel Heesch, (2004) “Cross-lingual
searching and visualization for greek and latin and old norse texts”, In: JCDL Vol. 383.
[53] Su, C. Y., Lin, T. C., & Wu, S. H. (2007, May). Using Wikipedia to translate OOV terms on MLIR.
In Proceedings of NTCIR-6 Workshop Meeting (pp. 109-115).
[54] Yang, H. C., & Lee, C. H., (2008, November) “Multilingual Information Retrieval Using GHSOM”,
In Intelligent Systems Design and Applications, 2008. ISDA'08. Eighth International Conference on
IEEE, Vol. 1, pp. 225-228.
[55] Dr. Saraswathi, S., Siddhiqaa, Asma, Kalaimagal, M, K., and Kalaiyarasi M (2010, April) BiLingual
Information Retrieval System for English and Tamil, Journal Of Computing, 2, 4, 85-89.
[56] Pingali, P., & Varma, V. (2006, September) “Hindi and Telugu to English Cross Language
Information Retrieval at CLEF 2006”, In Working Notes of Cross Language Evaluation Forum.
[57] Jagadeesh, J. & Kumaran, K, (2007), “Cross-Lingual Information Retrieval System for Indian
Languages.In: Advances in Multilingual and Multimodal Information Retrieval: 8th Workshop of the
Cross-Language Evaluation Forum, CLEF, pages: 80-87.
[58] Makin, R., Pandey, N., Pingali, P., & Varma, V., (2007) “Approximate string matching techniques for
effective CLIR among Indian languages”, In Applications of Fuzzy Sets Theory, Springer Berlin
Heidelberg, pp. 430-437.

142
[59] Seetha, A., Das, S., & Kumar, M. (2007, December). Evaluation of the English-Hindi Cross Language
Information Retrieval System Based on Dictionary Based Query Translation Method. In Information
Technology,(ICIT 2007). 10th International Conference on IEEE, pp. 56-61.
[60] Sethuramalingam, S., & Varma, V. (2008). IIIT Hyderabad‟s CLIR experiments for FIRE-2008. In
The working notes of First Workshop of Forum for Information Retrieval Evaluation (FIRE),
Kolkata.
[61] Chinnakotla, M. K., & Damani, O. P., (2009, August) “Experiences with english-hindi, english-tamil
and english-kannada transliteration tasks at news 2009”, In Proceedings of the 2009 Named Entities
Workshop: Shared Task on Transliteration ,Association for Computational Linguistics, pp. 44-47.
[62] Sujatha, P., Dhavachelvan, P., & Narasimhulu, V., (2010) “Evaluation of English-Telugu and
English-Tamil Cross Language Information Retrieval System using Dictionary Based Query
Translation Method”, International Journal of Computer Science and Information Security, 8(2), 314-
319.
[63] Ramanathan, A. (2003). State of the Art in Cross-Lingual Information Retrieval.VIVEK-BOMBAY-,
15(2), 16-22.
[64] Voorhees, Ellen M., & Donna K. Harman, eds, (2005), “TREC: Experiment and evaluation in
information retrieval. Vol. 63. Cambridge: MIT press, Vol. 63.
[65] Peters, Carol, (2001), “Cross-language information retrieval and evaluation”, Berlin Heidelberg:
Springer.
[66] Gey, F., Larson, R., Sanderson, M., Joho, H., Clough, P., & Petras, V. (2006) “Geo CLEF: the CLEF
2005 cross-language geographic information retrieval track overview ,Springer Berlin Heidelberg, pp.
908-919.
[67] Salton, G., (1973), “Experiments in multi-lingual information retrieval”, Information Processing
Letters, 2(1), 6–11.
[68] Chang, D. T. H., Weng, Y. Z., Lin, J. H., Hwang, M. J., & Oyang, Y. J., (2006) “Protemot: prediction
of protein binding sites with automatically extracted geometrical templates”, Nucleic acids research,
34(suppl 2), W303-W309.
[69] Shatkay, Hagit, (2005) “Hairpins in bookstacks: information retrieval from biomedical text”,
Briefings in Bioinformatics 6.3 (2005): 222-238.
[70] Oard, D. W., He, D., & Wang, J. (2008). User-assisted query translation for interactive cross-language
information retrieval. Information Processing & Management, 44(1), 181-211.
AUTHORS
Prof. Sanjay K. Dwivedi is working as Professor & Head, Department of Computer
Science at Babasaheb Bhimrao Ambedkar Central University, Lucknow, India. His
research interest is in Artificial Intelligence, Information Retrieval, Web Mining, NLP
and WSD. He has published number of research papers in reputed journals and
conferences. He is approachable at skd200@yahoo.com.
Ganesh Chandra received his M. Tech. degree from KNIT Sultanpur, U.P. India and
is pursuing his Ph.D. degree in Department of Computer Science at Babasaheb
Bhimrao Ambedkar Central University, Lucknow, India. His major research interests
include Information Retrieval, WSD and Machine Translation. He has published
papers in various international journals and conferences. He is approachable at
ganesh.iiscgate@gmail.com.

A SURVEY ON CROSS LANGUAGE INFORMATION RETRIEVAL

More Related Content

What's hot (20)

Similar to A SURVEY ON CROSS LANGUAGE INFORMATION RETRIEVAL (20)

Recently uploaded (20)

A SURVEY ON CROSS LANGUAGE INFORMATION RETRIEVAL