Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userTerm Selection for Query Expansion in Medical Cross-lingual Information Retrieval
اختيار الكلمات لتوسيع الاستعلامات في نظم استرجاع المعلومات الطبية متعددة اللغات
2260
1 46
5.0
(
1
)
Added by
Springer ورقة بحثية
Publication date
2019
fields
Informatics Engineering
and research's language is
العربية
Created by
Shadi Saleh
Ask ChatGPT about the research
نعرض في هذا البحث مودل لتوسيع الاستعلامات التلقائية في نظم استرجاع المعلومات متعددة اللغات في المجال الطبي. يوظف المودل المستخدم ترجمة آلية للاستعلام في اللغة المصدر الى لغة المستندات وتابع انحدار خطي لتوقّع دقة الاسترجاع لكل استعلام مترجم عند توسيع هذا الاستعلام مع كلمة مرشحة. الكلمات المرشحة (في لغة المستندات) اختيرت من مصادر متعددة: الترجمات المقترحة للاستعلام التي تم الحصول عليها من نظام ترجمة آلي, مقالات ويكيبيديا, وملخصات PubMed. توسيع الاستعلام يُطبق فقط عندما يتوقّع المودل قيمة للكلمة المرشحة تتجاوز عتبة تم تدريبها مسبقا ليسمح ذلك لتوسيع الاستعلامات فقط بالكلمات المرتبطة بقوّة به. اختباراتنا تم تنفيذها على بيانات الاختبار الخاصة ب CLEF eHealth 2013-2015 وأظهرت تفوق ملحوظ في نظم استرجاع المعلومات متعددة اللغات واحادية اللغة.
We present a method for automatic query expansion for cross-lingual information retrieval in the medical domain. The method employs machine translation of source-language queries into a document language and linear regression to predict the retrieval performance for each translated query when expanded with a candidate term. Candidate terms (in the document language) come from multiple sources: query translation hypotheses obtained from the machine translation system, Wikipedia articles and PubMed abstracts. Query expansion is applied only when the model predicts a score for a candidate term that exceeds a tuned threshold which allows to expand queries with strongly related terms only. Our experiments are conducted using the CLEF eHealth 2013--2015 test collection and show %seven source languages and also in the monolingual case. The results show significant improvements in both cross-lingual and monolingual settings.
Artificial intelligence review:
Research summary
تقدم هذه الورقة البحثية طريقة لتوسيع الاستعلامات تلقائيًا لاسترجاع المعلومات عبر اللغات في المجال الطبي. تعتمد الطريقة على ترجمة الاستعلامات من لغة المصدر إلى لغة الوثائق باستخدام الترجمة الآلية وتوظيف نموذج الانحدار الخطي للتنبؤ بأداء الاسترجاع لكل استعلام مترجم عند توسيعه بمصطلح مرشح. يتم الحصول على المصطلحات المرشحة من مصادر متعددة مثل فرضيات ترجمة الاستعلامات من نظام الترجمة الآلية، مقالات ويكيبيديا، وملخصات PubMed. يتم تطبيق توسيع الاستعلام فقط عندما يتجاوز النموذج درجة معينة مضبوطة مسبقًا، مما يسمح بتوسيع الاستعلامات بالمصطلحات ذات الصلة القوية فقط. أظهرت التجارب التي أجريت باستخدام مجموعة بيانات CLEF eHealth 2013-2015 تحسينات كبيرة في كل من إعدادات الاسترجاع عبر اللغات وأحادية اللغة.
Critical review
دراسة نقدية: تقدم الورقة البحثية طريقة مبتكرة لتوسيع الاستعلامات في استرجاع المعلومات عبر اللغات في المجال الطبي، ولكن هناك بعض النقاط التي يمكن تحسينها. أولاً، تعتمد الطريقة بشكل كبير على جودة الترجمة الآلية، والتي قد تكون غير دقيقة في بعض الأحيان، مما يؤثر على دقة الاسترجاع. ثانيًا، يمكن أن تكون عملية ضبط العتبة لتحسين الأداء معقدة وتحتاج إلى مزيد من التجارب لضمان الدقة. أخيرًا، على الرغم من أن الطريقة أظهرت تحسينات كبيرة، إلا أنها قد تكون محدودة في تطبيقها على مجالات أخرى غير المجال الطبي.
Questions related to the research
-
ما هي المصادر التي تم استخدامها للحصول على المصطلحات المرشحة لتوسيع الاستعلامات؟
تم الحصول على المصطلحات المرشحة من فرضيات ترجمة الاستعلامات من نظام الترجمة الآلية، مقالات ويكيبيديا، وملخصات PubMed.
-
ما هو الهدف من استخدام نموذج الانحدار الخطي في هذه الدراسة؟
يهدف نموذج الانحدار الخطي إلى التنبؤ بأداء الاسترجاع لكل استعلام مترجم عند توسيعه بمصطلح مرشح، مما يسمح بتوسيع الاستعلامات بالمصطلحات ذات الصلة القوية فقط.
-
ما هي مجموعة البيانات التي استخدمت في التجارب؟
استخدمت مجموعة بيانات CLEF eHealth 2013-2015 في التجارب.
-
ما هي الفائدة الرئيسية من توسيع الاستعلامات في استرجاع المعلومات عبر اللغات؟
الفائدة الرئيسية هي تحسين دقة استرجاع المعلومات عن طريق إضافة مصطلحات ذات صلة قوية للاستعلامات المترجمة، مما يساعد في التغلب على مشكلة عدم تطابق المفردات بين الاستعلامات والوثائق.
References used
Amati, G., Carpineto, C., Romano, G.: Query diculty, robustness, and selective application of query expansion. In: European conference on information retrieval. pp. 127{137. Springer, Berlin, Germany (2004)
Aronson, A.R.: E ective mapping of biomedical text to the umls metathesaurus: the metamap program. Proc AMIA Symp pp. 17{21 (2001)
Cao, G., Nie, J.Y., Gao, J., Robertson, S.: Selecting good expansion terms for pseudo-relevance feedback. In: Proceedings of the 31st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. pp. 243{250. SIGIR '08, ACM, New York, NY, USA (2008)
Chandra, G., Dwivedi, S.K.: Query expansion based on term selection for Hindi- English cross lingual IR. Journal of King Saud University - Computer and Information Sciences (2017)
Chiang, W.T.M., Hagenbuchner, M., Tsoi, A.C.: The wt10g dataset and the evolution of the web. In: Special Interest Tracks and Posters of the 14th International Conference on World Wide Web. pp. 938{939. WWW '05, ACM, New York, NY, USA (2005)
Choi, S., Choi, J.: Exploring e ective information retrieval technique for the medical web documents: Snumedinfo at clefehealth2014 task 3. In: Working Notes of CLEF 2015 - Conference and Labs of the Evaluation forum. vol. 1180, pp. 167{175. CEUR-WS.org, Sheeld, UK (2014)
Dusek, O., Hajic, J., Hlavacova, J., Novak, M., Pecina, P., Rosa, R., et al.: Machine translation of medical texts in the Khresmoi project. In: Proceedings of the Ninth Workshop on Statistical Machine Translation. pp. 221{228. Baltimore, USA (2014)
Ermakova, L., Mothe, J.: Query expansion by local context analysis. In: Conference francophone en Recherche d'Information et Applications (CORIA 2016). pp. 235{ 250. CORIA-CIFED, Toulouse, France (2016)
Gabrilovich, E., Broder, A., Fontoura, M., Joshi, A., Josifovski, V., Riedel, L., Zhang, T.: Classifying search queries using the web as a source of knowledge. ACM Transactions on the Web 3(2), 5 (2009)
Goeuriot, L., Kelly, L., Li, W., Palotti, J., Pecina, P., Zuccon, G., Hanbury, A., Jones, G., Mueller, H.: ShARe/CLEF eHealth evaluation lab 2014, Task 3: Usercentred health information retrieval. In: Proceedings of CLEF 2014. pp. 43{61. CEUR-WS.org, Sheeld,UK (2014)
Goeuriot, L., Kelly, L., Suominen, H., Hanlen, L., Nevaol, A., Grouin, C., Palotti, J., Zuccon, G.: Overview of the CLEF eHealth evaluation lab 2015. In: The 6th Conference and Labs of the Evaluation Forum. pp. 429{443. Springer, Berlin, Germany (2015)
Harman, D.: Towards interactive query expansion. In: Proceedings of the 11th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. pp. 321{331. SIGIR '88, ACM, New York, NY, USA (1988)
Harman, D.: Information retrieval. chap. Relevance Feedback and Other Query Modi cation Techniques, pp. 241{263. Prentice-Hall, Inc., Upper Saddle River, NJ, USA (1992)
Hull, D.: Using statistical testing in the evaluation of retrieval experiments. In: Proceedings of the 16th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval. pp. 329{338. ACM, Pittsburgh, USA (1993)
Humphreys, B.L., Lindberg, D.A.B., Schoolman, H.M., Barnett, G.O.: The uni ed medical language system. Journal of the American Medical Informatics Association 5(1), 1{11 (1998)
Kalpathy-Cramer, J., Muller, H., Bedrick, S., Eggel, I., De Herrera, A., Tsikrika, T.: Overview of the clef 2011 medical image classi cation and retrieval tasks. In: CLEF 2011 - Working Notes for CLEF 2011 Conference. vol. 1177. CEUR-WS (2011)
Koehn, P., Hoang, H., Birch, A., Callison-Burch, C., Federico, M., Bertoldi, N., et al.: Moses: Open source toolkit for statistical machine translation. In: Proceedings of the 45th Annual Meeting of the Association for Computational Linguistics, Demo and Poster Sessions. pp. 177{180. Stroudsburg, PA, USA (2007)
Liu, X., Nie, J.: Bridging layperson's queries with medical concepts { GRIUM @CLEF2015 eHealth Task 2. In: Working Notes of CLEF 2015 Conference and Labs of the Evaluation forum. vol. 1391. CEUR-WS.org, Toulouse, France (2015)
McCarley, J.S.: 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. pp. 208{214. College Park, Maryland (1999)
Mikolov, T., Sutskever, I., Chen, K., Corrado, G., Dean, J.: Distributed representations of words and phrases and their compositionality. In: Proceedings of the 26th International Conference on Neural Information Processing Systems - Volume 2. pp. 3111{3119. NIPS'13, Curran Associates Inc., USA (2013)
Nikoulina, V., Kovachev, B., Lagos, N., Monz, C.: Adaptation of statistical machine translation model for cross-lingual information retrieval in a service context. In: Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics. pp. 109{119. Stroudsburg, PA, USA (2012)
Nogueira, R., Cho, K.: Task-oriented query reformulation with reinforcement learning. In: Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. pp. 574{583 (2017)
Nunzio, G.M.D., Moldovan, A.: A study on query expansion with mesh terms and elasticsearch. IMS unipd at CLEF ehealth task 3. In: Working Notes of CLEF 2018 - Conference and Labs of the Evaluation Forum, Avignon, France, September 10-14, 2018. CEUR-WS, Avignon, France (2018)
Oard, D.: A comparative study of query and document translation for crosslanguage information retrieval. In: Machine Translation and the Information Soup, vol. 1529, pp. 472{483. Springer, Berlin, Germany (1998)
Ounis, I., Amati, G., Plachouras, V., He, B., Macdonald, C., Johnson, D.: Terrier information retrieval platform. In: Proceedings of the 27th European Conference on Advances in Information Retrieval Research. pp. 517{519. ECIR'05, Springer- Verlag, Berlin, Heidelberg (2005)
Pakhomov, S.V., Finley, G., McEwan, R., Wang, Y., Melton, G.B.: Corpus domain e ects on distributional semantic modeling of medical terms. Bioinformatics 32(23), 3635{3644 (2016)
Pal, D., Mitra, M., Datta, K.: Query expansion using term distribution and term association. CoRR abs/1303.0667 (2013)
Pal, D., Mitra, M., Datta, K.: Improving query expansion using wordnet. J. Assoc. Inf. Sci. Technol. 65(12), 2469{2478 (2014)
Palotti, J.R., Zuccon, G., Goeuriot, L., Kelly, L., Hanbury, A., Jones, G.J., Lu pu, M., Pecina, P.: CLEF eHealth Evaluation Lab 2015, Task 2: Retrieving information about medical symptoms. In: CLEF (Working Notes). pp. 1{22. Spriner, Berlin, Germany (2015)
rate research
Read More
Transformers that are pre-trained on multilingual corpora, such as, mBERT and XLM-RoBERTa, have achieved impressive cross-lingual transfer capabilities. In the zero-shot transfer setting, only English training data is used, and the fine-tuned model i
s evaluated on another target language. While this works surprisingly well, substantial variance has been observed in target language performance between different fine-tuning runs, and in the zero-shot setup, no target-language development data is available to select among multiple fine-tuned models. Prior work has relied on English dev data to select among models that are fine-tuned with different learning rates, number of steps and other hyperparameters, often resulting in suboptimal choices. In this paper, we show that it is possible to select consistently better models when small amounts of annotated data are available in auxiliary pivot languages. We propose a machine learning approach to model selection that uses the fine-tuned model's own internal representations to predict its cross-lingual capabilities. In extensive experiments we find that this method consistently selects better models than English validation data across twenty five languages (including eight low-resource languages), and often achieves results that are comparable to model selection using target language development data.
Introducing biomedical informatics (BMI) students to natural language processing (NLP) requires balancing technical depth with practical know-how to address application-focused needs. We developed a set of three activities introducing introductory BM
I students to information retrieval with NLP, covering document representation strategies and language models from TF-IDF to BERT. These activities provide students with hands-on experience targeted towards common use cases, and introduce fundamental components of NLP workflows for a wide variety of applications.
Multilingual pre-trained contextual embedding models (Devlin et al., 2019) have achieved impressive performance on zero-shot cross-lingual transfer tasks. Finding the most effective fine-tuning strategy to fine-tune these models on high-resource lang
uages so that it transfers well to the zero-shot languages is a non-trivial task. In this paper, we propose a novel meta-optimizer to soft-select which layers of the pre-trained model to freeze during fine-tuning. We train the meta-optimizer by simulating the zero-shot transfer scenario. Results on cross-lingual natural language inference show that our approach improves over the simple fine-tuning baseline and X-MAML (Nooralahzadeh et al., 2020).
With the advent of contextualized embeddings, attention towards neural ranking approaches for Information Retrieval increased considerably. However, two aspects have remained largely neglected: i) queries usually consist of few keywords only, which i
ncreases ambiguity and makes their contextualization harder, and ii) performing neural ranking on non-English documents is still cumbersome due to shortage of labeled datasets. In this paper we present SIR (Sense-enhanced Information Retrieval) to mitigate both problems by leveraging word sense information. At the core of our approach lies a novel multilingual query expansion mechanism based on Word Sense Disambiguation that provides sense definitions as additional semantic information for the query. Importantly, we use senses as a bridge across languages, thus allowing our model to perform considerably better than its supervised and unsupervised alternatives across French, German, Italian and Spanish languages on several CLEF benchmarks, while being trained on English Robust04 data only. We release SIR at https://github.com/SapienzaNLP/sir.
Zero-shot cross-lingual information extraction (IE) describes the construction of an IE model for some target language, given existing annotations exclusively in some other language, typically English. While the advance of pretrained multilingual enc
oders suggests an easy optimism of train on English, run on any language'', we find through a thorough exploration and extension of techniques that a combination of approaches, both new and old, leads to better performance than any one cross-lingual strategy in particular. We explore techniques including data projection and self-training, and how different pretrained encoders impact them. We use English-to-Arabic IE as our initial example, demonstrating strong performance in this setting for event extraction, named entity recognition, part-of-speech tagging, and dependency parsing. We then apply data projection and self-training to three tasks across eight target languages. Because no single set of techniques performs the best across all tasks, we encourage practitioners to explore various configurations of the techniques described in this work when seeking to improve on zero-shot training.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
