--- title: "Using Coreference for Question Answering — Morton 1999" description: "The paper tackles a common problem with search engines: a ranked list of documents still leaves the user reading long texts to find a short answer." authors: ["Thomas S. Morton"] year: 1999 venue: "Coreference and Its Applications" source: "https://aclanthology.org/W99-0212/" licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)" licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/papers/morton-1999-coreference-question-answering/" --- [Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/morton-1999-coreference-question-answering/) [home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [1999](https://md.research.elamur.ai/free/papers/1999/) / Using Coreference for Question Answering # Using Coreference for Question Answering **Authors:** [Thomas S. Morton](https://md.research.elamur.ai/free/authors/thomas-s-morton/) · **Year:** 1999 · **Venue:** Coreference and Its Applications · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/) **Links:** [Publisher's page](https://aclanthology.org/W99-0212/) · [PDF](https://aclanthology.org/W99-0212.pdf) · [OpenAlex](https://openalex.org/W1560812011) **Citations:** 24 citations (openalex-snapshot · 2026-10-14) · [11](#cited-by) in our library, of the 140673 papers whose bibliographies we read ## Summary The paper tackles a common problem with search engines: a ranked list of documents still leaves the user reading long texts to find a short answer. Morton proposes a second processing stage that runs on a small set of already-retrieved documents and ranks individual sentences that likely contain the answer to a query. The method treats the query as a relation among entities and events and looks for coreferent mentions of them in the documents. It captures identity links through within-document coreference resolution, name and word-order variants, stemming and an acronym database. It captures part-whole and synonymy links through mutual-information co-occurrence statistics computed on a large Wall Street Journal corpus. Sentence scores combine idf weights with expansion weights. On ten real search-engine queries, the system beat a tf.idf baseline, although the baseline was stronger than expected. The main result is a roughly tenfold drop in how much text a user must read before reaching an answer. This shows that discourse-aware coreference modeling can make passage-level question answering practical. ## Key points - Identity coreference inside documents relies on the UPENN CAMP techniques (Baldwin et al., 1998). Query noun phrases are expanded into weighted head-word variants, and acronym matches receive a weight of 1. - Related verbs, nominalizations and proper nouns are found with mutual information over more than 62,000 automatically tagged and parsed Wall Street Journal articles. This is how links such as Sudan–Khartoum and bomb–attack are captured. - The baseline is a standard tf.idf sentence ranker following Salton (1989), with Porter stemming as its only query expansion. - The evaluation used 10 queries taken from a real online search engine. Answers were drawn from the top ten documents found in Britannica Online, CNN and the wider Web. On average the system surfaced an answer within about 290 bytes, compared with over 3000 bytes when reading documents in search-engine order. - Planned future work includes: - a probabilistic pronoun model; - handling of wh-words; - a POS tagger trained on more questions than the Penn Treebank contains; - a named entity recognizer; - language models that adapt to each user. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/) Topics: [Coreference resolution](https://md.research.elamur.ai/free/topics/coreference-resolution/) ## Cited by Cited by 11 papers of our library cite it (of the 140673 whose bibliographies we read); 5 of them have a page here, 6 not yet. - [Neural End-to-end Coreference Resolution for German in Different Domains](https://md.research.elamur.ai/free/papers/coref--schroder-neural-end-end-coreference-resolution/) - [Joint Determination of Anaphoricity and Coreference Resolution using Integer Programming](https://md.research.elamur.ai/free/papers/denis-2007-joint-determination-anaphoricity-coreference-resolution/) - [ParCorFull: a Parallel Corpus Annotated with Full Coreference](https://md.research.elamur.ai/free/papers/koltunski-2018-parcorfull-parallel-corpus-annotated-full/) - [A Probabilistic Annotation Model for Crowdsourcing Coreference](https://md.research.elamur.ai/free/papers/paun-2018-probabilistic-annotation-model-crowdsourcing-coreference/) - [SemEval-2010 Task 1: Coreference Resolution in Multiple Languages](https://md.research.elamur.ai/free/papers/recasens-2009-semeval-2010-task-1-coreference/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{morton1999using, title = {{Using Coreference for Question Answering}}, author = {Morton, Thomas S.}, booktitle = {Coreference and Its Applications}, year = {1999}, doi = {10.3115/1608810.1608826}, url = {https://doi.org/10.3115/1608810.1608826} } ``` - [← previous Using Coreference Chains for Text Summarization](https://md.research.elamur.ai/free/papers/azzam-1999-coreference-chains-text-summarization/) - [next → What is coreference, and what should coreference annotation be?](https://md.research.elamur.ai/free/papers/deemter-1999-what-is-coreference-what-should/) - [↑ back to papers of 1999](https://md.research.elamur.ai/free/papers/1999/) ## References 5 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Breck Baldwin and Thomas Morton. 1998. Dynamic coreference-based summarization. In Proceedings of the Third Conference on Empirical Methods in Natural Language Processing, Granada, Spain, June. [find](https://scholar.google.com/scholar?q=Breck+Baldwin+and+Thomas+Morton%2E+1998%2E+Dynamic+coreference%2Dbased+summarization%2E+In+Proceedings+of+the+Third+Conference+on+Empirical+Methods+in+Natural+Language+Processing%2C+Granada%2C+Spain%2C+June%2E) [our card](https://md.research.elamur.ai/free/papers/baldwin-1998-dynamic-coreference-based-summarization/) 2. B. Baldwin, T. Morton, Amit Bagga, J. Baldridge, R. Chandraseker, A. Dimitriadis, K. Snyder, and M. Wolska. 1998. [find](https://scholar.google.com/scholar?q=B%2E+Baldwin%2C+T%2E+Morton%2C+Amit+Bagga%2C+J%2E+Baldridge%2C+R%2E+Chandraseker%2C+A%2E+Dimitriadis%2C+K%2E+Snyder%2C+and+M%2E+Wolska%2E+1998%2E) 3. Description of the UPENN CAMP system as used for coreference. In Proceedings of the Seventh Message Understanding Conference (MUC-7), Baltimore, Maryland. [find](https://scholar.google.com/scholar?q=Description+of+the+UPENN+CAMP+system+as+used+for+coreference%2E+In+Proceedings+of+the+Seventh+Message+Understanding+Conference+%28MUC%2D7%29%2C+Baltimore%2C+Maryland%2E) [our card](https://md.research.elamur.ai/free/papers/baldwin-1998-description-upenn-camp-system-as/) 4. Inderjeet Mani and Eric Bloedorn. 1998. Machine learning of generic and user-focused summarization. In Proceeding of the Fifteenth National Conference on Artificial intelligence (AAAI-98). [find](https://scholar.google.com/scholar?q=Inderjeet+Mani+and+Eric+Bloedorn%2E+1998%2E+Machine+learning+of+generic+and+user%2Dfocused+summarization%2E+In+Proceeding+of+the+Fifteenth+National+Conference+on+Artificial+intelligence+%28AAAI%2D98%29%2E) 5. Gerald Salton. 1989. Automatic text processing: the transformation, analysis, and retrieval of information by computer. Addison-Wesley Publishing Company, Inc. [find](https://scholar.google.com/scholar?q=Gerald+Salton%2E+1989%2E+Automatic+text+processing%3A+the+transformation%2C+analysis%2C+and+retrieval+of+information+by+computer%2E+Addison%2DWesley+Publishing+Company%2C+Inc%2E)